HARLINGTON WOOD, JR., Circuit Judge.

Passengers collected their belongings and prepared to disembark as Amtrak Train Number 4 came to a halt in Chicago’s Union Station. When the doors finally opened, several hundred people, some of whom had boarded the train at its Los Angeles origination point, spilled out onto the platform and made their way toward the terminal. In the middle of this frenzied activity, one person was moving slower than the rest. Derek Foster was struggling to pull two new-looking, hard-sided suitcases and was therefore unable to keep up with his fellow passengers.

The encumbered Foster appeared nervous and repeatedly looked over his shoulder as if someone were following him. In fact, someone was; his unwieldy luggage had caught the attention of several members of a Drug Enforcement Administration (“DEA”) task force. Two of the agents approached Foster as he entered the terminal, identified themselves as law enforcement officers, and requested a few moments of his time.

Foster agreed and began to set down his suitcases. One of them fell over, emitting a puff of white powder from its seam. Foster thereafter attempted to return the suitcase to an upright position but was not immediately successful. The suitcase fell over once again and more white powder billowed from its seam. An agent commented on the apparent weight of the bag, and Foster replied that he had a “box” inside. A third agent approached the group, touched and smelled the particles that had escaped from the suitcase, and determined that the substance was talcum powder.

*449In response to the agents’ requests, Foster identified himself and produced a driver’s license but claimed that he had lost his train ticket. Foster also disclaimed any ownership of the suitcases, telling the agents that he had agreed to carry them from the train to the baggage claim area as a favor to a young black male. Foster did not know this person’s name, but he described the man as tall, wearing a blue jacket, and carrying two suitcases that were identical to the ones at his feet. None of the law enforcement officers had seen this man and an immediate search failed to locate anyone fitting the description.

Foster did admit, however, to ownership of the blue duffel bag that was strapped across his shoulder. He even volunteered it for search by the agents. Upon opening it, the agents discovered clothing, a digital-display beeper, a notebook, and train tickets. The tickets were issued to Mark Wynn and permitted one-way travel from Los Angeles to Chicago and from Chicago to Rockville, Maryland. They had been purchased with cash less than an hour and a half before the train had departed Los Angeles.

A quick perusal of the notebook disclosed notations reflecting weights, money, and names. It also revealed the following handwritten verse:

Key for Key, Pound for pound I’m the biggest Dope Dealer and I serve all over town. Rock 4 Rock Self 4 Self. Give me a key let me go to work more Dollars than your average bussiness [sic] man.1

A receipt lodged in between the notebook’s pages indicated that Foster had signed for the beeper.2

The agents then asked if they could search the two suitcases. Foster replied that he did not care, again explaining that this luggage did not belong to him. The group then made its way to the DEA station office and the agents opened the suit-eases.3

One suitcase contained a kilogram of cocaine and a five-gallon can containing a liquid form of phencyclidine (“PCP”). The other suitcase contained another five-gallon can filled with liquid PCP. Talcum powder, diapers, and foam rubber had been used to pack the narcotics.

Foster, who was watching the agents’ activity, commented that he had not known “that there was that much in there.” When the agents completed their search, Foster was placed under arrest. He signed a waiver of his rights and, while relaying information about his personal history, asked to speak in private with his interviewer, Detective Kinsella.

After the other agents left the room, Foster gave Kinsella a very different account of how he had obtained the bags. Under this revised version, a friend promised to pay him for transporting the suitcases from Los Angeles to Rockville, Maryland. Foster knew that the suitcases contained “something,” but he did not know how much.

After Foster failed in his bid to make out a violation of his fourth amendment rights, his case proceeded to trial on charges of possession with intent to distribute cocaine and PCP. See 21 U.S.C. § 841(a)(1). And as Foster concedes, the only disputed issue at that trial was whether he knew that he was transporting controlled substances. During the course of that trial, he believes that the district court erred when it allowed Kinsella to testify as an expert regarding the methods used by narcotics traffickers. More specifically, he charges error in the admission of testimony that Los Angeles is a major source city for cocaine and liquid PCP and that, as a general rule, drug couriers try to:

(1) Use cash instead of checks or credit cards to avoid creating a record of their activities;

*450(2) Avoid using their real names in travelling;

(3) Purchase travel tickets shortly before departure to avoid having their travel plans known ahead of time;

(4) Purchase one-way tickets for drug-smuggling runs;

(5) Use masking agents such as talcum powder to disguise the odor of narcotics;

(6) Use hard-sided suitcases to keep the odor of drugs from escaping;

(7) Travel by train in part because they can watch their luggage more easily;

(8) Use trains because train station security is minimal compared to security systems used in airports;

(9) Use beepers in order to permit anonymity and mobility; and

(10) Conduct countersurveillance activities to avoid detection by competitors or the police.

Foster also finds error in the admission of the verse found in his notebook and in the admission of testimony that PCP in its liquid form is dangerous to touch and highly explosive. He concludes that the admission of this evidence impermissibly contaminated his trial, but we cannot agree.4

I.

This court applies a deferential standard in analyzing a district court’s evidentiary rulings; we will presume their correctness unless the defendant “can show a clear abuse of discretion.” United States v. Carter, 910 F.2d 1524, 1530 (7th Cir.1990), cert. denied, — U.S. —, 111 S.Ct. 1628, 113 L.Ed.2d 724 (1991). Even if the district court has erred, moreover, reversal is required only if the error has affected “substantial rights.” Fed.R.Crim.P. 52(a). In considering whether a nonconstitutional error affects substantial rights, “ ‘[o]ur task is to gauge “what effect the error had or reasonably may be taken to have had upon the jury’s decision.” ’ ” United States v. Zapata, 871 F.2d 616, 622 (7th Cir.1989) (quoting United States v. Shackleford, 738 F.2d 776, 783 (7th Cir.1984)). Those errors that did not influence the jury or had only a very slight effect are disregarded as harmless.

A. Kinsella’s Expert Testimony

Foster’s first challenge is in actuality a sextet of arguments that is more conveniently divided into two sub-groups. One subgroup charges that the district court should have excluded Kinsella’s expert testimony in the first place because: (1) the testimony is irrelevant outside of a probable cause hearing; (2) an expertise as to the methods used by narcotics traffickers is in reality no expertise at all; (3) the significance of the underlying evidence (for example, the beeper) was well within the common understanding of the jury and did not require expert testimony; and (4) the testimony would constitute an opinion as to Foster’s mental state and would therefore violate Rule 704(b) of the Federal Rules of Evidence. A second sub-group charges that, after erring in its decision to admit the evidence, the district court erred further by failing to: (1) take precautionary efforts to avoid any confusion that might have been engendered by Kinsella’s role as both eyewitness and expert; and (2) allow Foster a wider latitude in cross-examining Kinsella.

Contrary to Foster’s apparent assumption in briefing these arguments, his wide-ranging challenge on appeal is considerably broader than the scope of the objections that he articulated at trial. Even charitably read, the transcript reveals that he preserved only two issues for appeal. First, he objected to Kinsella’s expert testimony about the beeper on the ground that it spoke to matters that the jury could evaluate for itself. Second, he attempted but was prevented from pursuing more extensive cross-examination of Kinsella. The unpreserved issues, therefore, must be analyzed under the plain error standard; we will reverse only if: (1) the district court made an error; and (2) that error repre*451sented a “miscarriage of justice”5 such that Foster “ ‘probably would have been acquitted but for the erroneously admitted evidence.’ ” United States v. Carroll, 871 F.2d 689, 692 (7th Cir.1989) (quoting United States v. Wynn, 845 F.2d 1439, 1443 (7th Cir.1988)); see also United States v. Myers, 892 F.2d 642, 645 (7th Cir.1990).

According to Foster, the government’s use at trial of this type of expert testimony is a relatively new phenomenon. This circuit, however, is quite familiar with the use during trial of expert testimony as to the methods used by drug dealers. Indeed, we were confronted with and upheld the use of such testimony as early as United States v. Ramirez, 796 F.2d 212, 214, 216-17 (7th Cir.1986). Thereafter, in United States v. Rollins, 862 F.2d 1282, 1292 (7th Cir.1988), cert. denied, 490 U.S. 1074, 109 S.Ct. 2084, 104 L.Ed.2d 648 (1989), we indicated that “narcotics code words and the operations of drug dealers are generally an appropriate subject for expert testimony.” Most recently, in United States v. Gonzalez, 933 F.2d 417 (7th Cir.1991), we upheld the admissibility of expert testimony that “narcotic-related phone calls” are usually short in nature so as to avoid detection. Id. at 428-29.6

This solid line of cases dispenses with many of the challenges raised by Foster. United States v. Teslim, 869 F.2d 316 (7th Cir.1989), for example, clearly rejects any per se bar to the admission of drug courier profile evidence once the probable cause hearing has concluded. If the testimony is relevant to the defendant’s guilt or innocence, then it is potentially admissible at trial. Id. at 324; see United States v. Solis, 923 F.2d 548, 551 (7th Cir.1991) (expert testimony relevant as circumstantial evidence of defendant’s intent to distribute cocaine).

The issue here was whether Foster knew that he was carrying narcotics. As placed in context by expert testimony, Foster’s behavior was material to the issue of knowledge and made it “more probable” that he knew what he was carrying; Kinsella’s testimony was thus relevant to the issue of Foster’s knowledge. See Fed.R.Evid. 401. And if relevant, Rule 702 allows for the admission of expert testimony if the witness is qualified and if the testimony would be helpful to the jury. Fed.R.Evid. 702.7

Kinsella was clearly qualified by reason of “knowledge, skill, experience, training, or education.” Id. His credentials included extensive training in the area of narcotics investigation, over twenty years of police work focused primarily on narcotics investigation, and approximately 2000 narcotics-related arrests. Kinsella had served as an instructor in some twenty DEA seminars and had helped to establish drug interdiction teams at airports and train stations across the country. He had also been qualified as an expert witness on approximately fifty prior occasions.

It is our continued belief, moreover, that this type of expert testimony is helpful to a jury. Despite our country’s “war on drugs” and its accompanying media cover*452age, it is still a reasonable assumption that jurors are not well versed in the behavior of drug dealers. “The investigator and the expert witness both serve as a link to the drug culture in providing the jury with [an] understanding of the intricate patterns and modus operandi ” of those involved in narcotics trafficking. Gonzalez, 933 F.2d at 428-29; see also United States v. De Soto, 885 F.2d 354, 360 (7th Cir.1989) (activity that “may appear, to the outside observer, to be perfectly normal and innocent” may take on an entirely new significance when placed in context by expert testimony); Daniels, 723 F.2d at 33 (“The methods by which drug dealers attempt to conceal their activities ‘[are] not something with which most jurors are familiar.’ ”) (quoting United States v. Scavo, 593 F.2d 837, 844 (8th Cir.1979)).8

Our general acceptance of this type of testimony notwithstanding, we are not unmindful of its potential for undue prejudice. Indeed, we have held that the district court and the prosecutor must be vigilant in ensuring that the expert’s testimony does not become so broad that it speaks to matters that the jury can evaluate for itself. De Soto, 885 F.2d at 361. When the expert witness also serves as an eyewitness, the district court and the prosecutor should exercise special caution to ensure that the jury understands its function in evaluating the evidence and is not confused by the witness’s dual role. Id. at 360, 361.

Our review of the record assures us that the district court and the prosecutor were aware of their special responsibilities and proceeded with the requisite degree of caution. Kinsella’s testimony was within that range that would be helpful to the jury in determining whether Foster knew what he was carrying. It may be innocent behavior to purchase a one-way train ticket, for cash, on the same day as departure from a source city for illegal drugs, under a false name, and carrying a beeper, but it is a fair use of expert testimony to offer another explanation for such behavior. The same rationale holds true for the remainder of Kinsella’s testimony. See Gonzalez, 933 F.2d at 428-29 (upholding admission of expert testimony that narcotics-related conversations are generally brief); Solis, 923 F.2d 548 (upholding admission of expert testimony that beepers are tools of narcotics trade); United States v. Briscoe, 896 F.2d 1476, 1497 (7th Cir.) (upholding admission of expert testimony concerning meaning of code terms and phrases used by narcotics traffickers), cert. denied, — U.S. —, 111 S.Ct. 173, 112 L.Ed.2d 137 (1990); De Soto, 885 F.2d at 360-61 (upholding admission of expert testimony concerning countersurveillance techniques employed by narcotics traffickers); Rollins, 862 F.2d at 1292 (upholding admission of expert testimony as to use of code words by narcotics traffickers).9

The district court was sensitive, moreover, to the need to confine Kinsella’s testimony to comments that would be helpful to the jury. As one example, the district court prevented Kinsella from expressing an opinion as to whether it would be unusual for a narcotics trafficker to deny possession of luggage that contained narcotics. The district court then gave a *453contemporaneous instruction: “What people do when accused of wrongdoing is an issue that the jury can determine based on their common experiences in life and their common sense.”

The district court was also sensitive to the need to prevent any jury confusion that might be engendered by Kinsella’s dual role. A witness may testify as both an eyewitness and an expert witness during the course of the same trial, see De Soto, 885 F.2d at 860; see also Solis, 923 F.2d at 551, but this dual role may increase the possibility that the jury will not understand its function in evaluating the evidence. De Soto, 885 F.2d at 360. As in De Soto, however, our concerns have been addressed by the use of cautionary instructions and thorough cross-examination.

As to cautionary instructions, the district court qualified Kinsella with the statement, “You may proceed by way of Rule 702”; scrupulously neutral terms that did not serve to overemphasize his role as an expert. Later, in qualifying another expert witness, the district court read the text of Rule 702 to the jury. The jury instruction on expert testimony, taken directly from this court’s Pattern Instruction 3.27, followed up these preliminary remarks with another explanation of Rule 702 and also told the jurors that they were free to credit as much or as little of Kinsella’s testimony as they saw fit.10 While Foster would in hindsight demand more in the way of instructions from the district court, we believe that the district court did enough under the circumstances. At least part of the blame for this alleged “error,” moreover, must be attributed to Foster; he failed to request more extensive instructions on this issue and failed even to raise the issue until this appeal.

Foster’s prosecutor was also cognizant of the need to prevent jury confusion. While a portion of Kinsella’s testimony alternated between eyewitness observations and expert opinions, the manner in which the examination was conducted made sufficient attempts to distinguish the two. During closing arguments, moreover, the prosecutor was careful not to place undue emphasis on Kinsella’s expert testimony:

You heard testimony the pager is a tool of the trade in drug [trafficking]. And we’re not saying to you it’s not used for other reasons; of course it is. We know that in our daily lives. Doctors use them, lots of people use them.

But sometimes they’re not used for appropriate purposes. And sometimes the people that use them are narcotics couriers; and they use them, as you heard from the testimony, to keep in touch with the people they’re transacting their narcotic[s] business with.

As to the scope of cross-examination, Foster points only to a few areas in which he was restricted in his cross-examination of Kinsella. This portrayal, however, unfairly belittles the fact that the cross-examination was otherwise quite extensive; the district court gave Foster wide latitude. For example, Foster was allowed to ask Kinsella, in his expert capacity, whether narcotics traffickers generally received assistance from their relatives in purchasing airline or train tickets. The district court also permitted Foster to relitigate, almost in its entirety, the propriety of the original stop. Compare Teslim, 869 F.2d at 324 (probable cause is issue for judge, not jury). There were some restrictions on- Foster’s efforts, but these restrictions were not as serious as Foster now suggests. De Soto, moreover, does not require the district court to permit highly irrelevant or otherwise impermissible cross-examination.11 See United States v. De-*454gaglia, 913 F.2d 372, 377 (7th Cir.1990) (“Although criminal defendants enjoy the constitutional guarantee that they will have an effective opportunity to test the accuracy of adverse evidence, ‘this guarantee does not include the opportunity to test the evidence “in whatever way, and to whatever extent the defense might wish.” ’ ”) (quoting Briscoe, 896 F.2d at 1492).

The only remaining issue in this original sextet is whether Kinsella’s testimony included or comprised an impermissible opinion on Foster’s mental state. Fed.R.Evid. 704(b).12 Foster has not identified, however, and this court could not locate, any specific statement in which Kinsella opines that Foster had the requisite mental state. Instead, Foster asserts that Kinsella's testimony was impermissible merely because the jury could use it to infer that Foster had the requisite mental state.

As the D.C. Circuit has persuasively noted, however, acceptance of Foster’s logic “would swallow the permissive aspects of Rule 704.” United States v. Dunn, 846 F.2d 761, 762 (D.C.Cir.1988) (expert’s testimony that quantities of drugs, drug packaging material, drug paraphernalia and weapons located in town house indicated retail drug operation did not violate Rule 704(b)).

All expert evidence assists jurors in analyzing and drawing inferences from other evidence; in so doing it may support inferences as to ultimate intent.... Suppose, for example, that an expert testifies at a homicide trial that the victim died of a poison administered daily in small doses over a long period. The evidence goes not only to what happened, but suggests extreme premeditation on the part of whomever doled out the poison. It is only as to the last step in the inferential process — a conclusion as to the defendant’s actual mental state — that Rule 704(b) commands the expert to be silent.

Id. Kinsella’s testimony thus fell within permissible bounds. It merely assisted the jury in coming to a conclusion as to Foster’s mental state; it did not make that conclusion for them. Compare United States v. Windfelder, 790 F.2d 576, 582 (7th Cir.1986) (expert witness testified that defendant “ ‘intentionally understated his income,’ ” which was element of crime).

In sum, both the type of testimony given by Kinsella and the circumstances surrounding its admission indicate that the district court was well aware of the problems inherent in admitting this relevant but potentially volatile testimony. These *455factors also indicate that the district court was careful to address those problems in a manner that avoided unfair prejudice to Foster. We will not, under these circumstances, label the district court’s decision an abuse of discretion.

B. The Verse

Foster’s second evidentiary challenge to his conviction is that the district court erred in admitting the verse found in his notebook. Although the government argued that the verse was admissible without limitation because it was “ ‘intricately related to the facts of the case,’ ” United States v. Monzon, 869 F.2d 338, 343 (7th Cir.), cert. denied, 490 U.S. 1075, 109 S.Ct. 2087, 104 L.Ed.2d 650 (1989), the district court concluded otherwise and admitted the verse under the government’s fall-back argument — Rule 404(b).13 It then gave a contemporaneous instruction:

The document is received for a limited purpose. It is not received to establish that the defendant is, in fact, the biggest dope dealer. It is not received that the defendant makes more dollars than the average businessman. It is not received for that purpose. It is received for a limited purpose.

The admissibility of evidence of other acts or crimes is governed by Rule 404(b) of the Federal Rules of Evidence, which provides that such evidence may not be used to prove a person’s bad character or his propensity to commit crimes in conformity with that character, but may be used for other purposes, such as proof of motive, opportunity, intent, preparation, plan, knowledge or absence of mistake or accident.

The limited purpose for which the document is received is only as to evidence of knowledge and intent. The defendant is accused in the indictment of having knowledge and intention; that he knowingly and intentionally did something,

This instruction illustrates the care with which the district court treated this issue, and in large part supports our conclusion that the decision to admit the verse did not constitute an abuse of discretion.

Foster’s argument is straightforward; he claims that the prejudice from admitting the verse clearly outweighed its minimal relevance to the issue of knowledge. As to relevance, he argues that the verse “certainly was nothing that could show knowledge of what was in the suit-eases” because the verse “made no reference to the suitcases he carried, or to the trip he was making.” This view of relevance, however, is unduly restrictive. The verse, standing alone, need not have been enough to prove knowledge; it is sufficient that the verse made it more probable that Foster had knowledge (and, therefore, more probable that he was guilty of the crime charged). United States v. York, 933 F.2d 1343, 1351 (7th Cir.1991) (citing Fed.R.Evid. 401). In our view, the verse clearly meets that test; it indicated, at a minimum, that Foster was familiar with drug code words and, to a certain extent, narcotics trafficking, a familiarity that made it more probable that he knew that he was carrying illegal drugs.

Foster’s knowledge, moreover, was relevant to the charges that he faced and was, in Foster’s words, “the only issue in the case.” Compare United States v. Wright, 901 F.2d 68 (7th Cir.1990) (Rule 404(b) evidence inadmissible to prove intent and “identity” if they are not at issue). Foster even defended on that basis, offering an innocent explanation for his activities and maintaining that he had no knowledge of the suitcases’ contents. And under these circumstances, the verse achieved heightened relevance by virtue of the fact that it also rebutted Foster’s protestations *456of naiveté. York, 933 F.2d at 1350 (“When the defendant affirmatively denies having the requisite intent by proffering an innocent explanation for his actions, the government is entitled to rebut that argument” by means of Rule 404(b) evidence).

Before moving on to prejudice, however, it is necessary to discuss one further point. Foster attempts to destroy, or at least diminish, the relevance of the verse by claiming that it was written for eventual incorporation into a rap song. These are rap lyrics, he claims. They have artistic value. They are fiction, just like Dashiel Hammett’s description of violent acts in The Maltese Falcon. And as such, they cannot be relevant to his guilt.

This court has not faced such an argument in the past nor, does it appear, have many others. Indeed, the parties have cited no cases that are really on point, and this court has discovered only one case discussing the admission of a defendant’s own literary or artistic work under a Rule 404(b) theory. In State v. Hanson, 46 Wash.App. 656, 731 P.2d 1140, review denied, 108 Wash.2d 1003 (1987), a state appellate court reversed a conviction under the rationale that the defendant’s fictional writings were not probative of the defendant’s character and too prejudicial to be admitted under the state’s version of Fed.R.Evid. 404(a). The court suggested, however, that there may be instances in which a defendant’s fictional writings would be admissible under the state equivalent of Fed.R.Evid. 404(b).

If nothing else, Hanson underscores the need to recall that the rap verse was not admitted to show that Foster was, in fact, “the biggest dope dealer”; it was not admitted to establish that Foster was the character portrayed in the lyrics. But in writing about this “fictional” character, Foster exhibited knowledge of an activity that is far from fictional. He exhibited some knowledge of narcotics trafficking, and in particular drug code words. It was for this limited purpose that the verse was admitted, and it is for this limited purpose that its relevance is clear. Compare Monzon, 869 F.2d at 344-45 (certain evidence indicating “use of controlled substances”— i.e., that defendant had marijuana butts in his car and that he sported a long pinky fingernail — held inadmissible to establish intent to distribute cocaine).

Much of Foster’s argument on this point is therefore of limited usefulness because, to answer his concerns by the same type of analogy, admitting the rap verse was not the equivalent of admitting The Godfather as evidence that Mario Puzo was a mafia don or admitting “The Pit and the Pendulum” as evidence that Edgar Allen Poe had tortured someone. It was, instead, the equivalent of admitting The Godfather to illustrate Puzo’s knowledge of the inner workings of an organized crime family and admitting “The Pit and the Pendulum” to illustrate Poe's knowledge of medieval torture devices. Rap music, under Foster’s definition, “constitutes a popular musical style that describes urban life”; it describes the reality around its author. And it is Foster’s knowledge of this reality, as evidenced by the verse that he has admittedly authored, that was relevant to the crimes for which he was charged.

This brings us to the matter of prejudice. Foster asserts that the admission of the verse, given its allegedly nonexistent probative value, was clearly prejudicial. We agree that the verse was prejudicial to Foster, but that concession benefits him little. All evidence offered by the prosecutor is prejudicial to the defendant; there would be no point in offering it if it were not. See United States v. Sophie, 900 F.2d 1064, 1076 (7th Cir.) (“[A]ll probative evidence is prejudicial to the party against whom it is introduced.”), cert. denied, — U.S. —, 111 S.Ct. 124, 112 L.Ed.2d 92 (1990). The real focus should be on whether the evidence is “unduly prejudicial.” Huddleston v. United States, 485 U.S. 681, 691, 108 S.Ct. 1496, 1502, 99 L.Ed.2d 771 (1988).

The Supreme Court in Huddleston identified four ways in which a defendant is protected from undue prejudice arising from Rule 404(b) evidence. Id. at 691-92, 108 S.Ct. at 1502. These protections are: (1) the evidence must be admissible for a *457proper purpose under Rule 404(b); (2) the evidence must be relevant under Rule 402; (3) the evidence must survive Rule 403’s balancing test; and (4) the evidence must be the subject of a jury instruction, if requested under Rule 105, that the evidence is to be considered only for the proper purpose for which it is admitted. Id,.; see also United States v. Murphy, 935 F.2d 899, 901 (7th Cir.1991). Of the four we have already found that three are present in this case. This leaves only Rule 403, in which the district court is directed to use a balancing test to determine whether the probative value of the evidence “is substantially outweighed by the danger of unfair prejudice.” Fed.R.Evid. 403.

As with any evidence admitted under Rule 404(b), there is always a possibility of unfair prejudice, i.e., that a jury could use the verse to draw forbidden inferences. York, 933 F.2d at 1349. There is also, as we have just explained, a legitimate inference that may be drawn from the verse. The task of assessing the relative impact of these inferences, and any accompanying potential for unfair prejudice, is one that, “to a large extent, requires a contemporaneous assessment of the presentation, credibility, and impact of the challenged evidence.” Id. at 1352. The district court is thus uniquely suited to that task, and we have rarely hesitated to uphold the results of such a balancing act where, as here, the district court has exercised such great care.

C. The Attributes of Liquid PCP

Foster’s last challenge takes issue with the district court’s decision to allow Officer Boyle to testify that liquid PCP is dangerous to touch (contact with the skin is the equivalent of ingestion) and highly explosive, claiming that the sole purpose of offering this evidence was “to inflame the jury.” See Ferrier v. Duckworth, 902 F.2d 545, 548 (7th Cir.), cert. denied, — U.S. —, 111 S.Ct. 526, 112 L.Ed.2d 536 (1990). It is far from clear that this testimony was “inflammatory,” however. We are inclined to agree with the district court’s observation on the matter: “I don’t think it’s any shocking ... shocking thing for the jury to know that narcotics substances are dangerous in one form or another. That this particular one happens to be dangerous also in terms of being an explosive is simply one of the facts of the case.” Foster’s present argument is also somewhat difficult to understand in light of his behavior at trial; his closing argument openly ridiculed the idea that liquid PCP could be explosive.

The record also belies Foster’s argument about the lack of a legitimate purpose for introducing the evidence. Foster’s cross-examination of several of the DEA agents had focused on the activity that had occurred when Foster, the suitcases, and the agents had arrived at the DEA’s station office. Detective Boyle had been present during the events in question, but his focus for most of that time had been elsewhere; he was concentrating on the contents of the suitcase, in particular the liquid PCP. The prosecutor anticipated that Foster would follow the same line of cross-examination as to Boyle and wanted to preempt the effect of those questions by offering a reason why Boyle was not paying attention to the activity around him. The questions, moreover, were narrowly tailored to this purpose and the prosecutor made use of a side bar to give advance notice of the government’s intentions to the district court and the defense attorneys.

The evidence thus had purpose beyond mere “inflammation” of the jury. Compare id. (“The only conceivable reason” for placing photographs of a bloody murder scene in evidence was to inflame jury against defendant). This purpose arose by virtue of Foster’s own trial tactics. And these facts, in combination with the fact that it is open to question whether the evidence was really inflammatory at all (or, at best, that the potential for unfair prejudice was minimal), support our decision to uphold the admission of this testimony as properly within the district court’s discretion.

II.

Foster portrayed himself as a young man taking his first train ride who, in a text*458book example of the golden rule, offered to assist a fellow passenger in transporting some heavy suitcases to the baggage claim area. The jury did not believe him, but the blame for this disbelief cannot be pinned upon the district court. The challenged evidence was relevant for the purposes for which it was admitted and the district court exercised the requisite degree of care when allowing the evidence to go before the jury. There was no abuse of discretion here, and Foster’s convictions and sentence are therefore

Affirmed.

MORGAN, Senior Circuit Judge.

Appellants Tom and Penny Crutchfield were convicted by a jury of charges involving the illegal importation and the intent to sell certain Figi banded iguanas in the United States, in violation of 18 U.S.C. §§ 545 and 371, and .16 U.S.C. §§ 3372(a)(1) and 3373(d)(1)(A) and (B). Both appellants subsequently filed timely motions for a new trial, asserting pervasive prosecutorial misconduct throughout the course of their first trial. The district court denied these motions. Appellants now appeal their respective convictions, contending that the prosecutor’s numerous instances of misconduct during their two-week trial unfairly prejudiced the outcome of the proceeding. We have jurisdiction pursuant to 28 U.S.C. § 1291, and we REVERSE and REMAND for a new trial.

I. Factual Background

Appellants Tom and Penny Crutchfield are commercial importers and distributors of reptiles. Prior to their indictment in this action, the Crutchfields were the proprietors and managers of Herpetofauna, Inc., one of the largest dealers and importers of exotic reptiles in the United States. This case revolves around four rare lizards known as “Figi banded iguanas” which appellants, through Herpetofauna, Inc., owned from May of 1989 until August of 1990.

Figi Iguanas are internationally recognized as an endangered species. Convention on International Trade in Endangered Species of Wild Fauna and Flora, Mar. 3, 1973, 27 U.S.T. 1089, 1140 (CITES). Because of the Figis’ endangered status, CITES regulates the importation of this special iguana into the United States. CITES requires that .an individual seeking to import a Figi first obtain a special authorizing permit from both the United States and the exporting country. Id. at 1095-96. Before the enactment of CITES, however, no permitting requirements for the importation of Figis were in place. Therefore, those Figi iguanas that found their way into the United States before the enactment of CITES (as well as their captive-bred progeny) may be possessed lawfully in the United States without a permit.

The crucial issue before the jury in the Crutchfield case was whether the Figis possessed by Herpetofauna between 1989 and 1990 were illegally imported into the United States by the Crutchfields without the requisite CITES permits, or whether these iguanas were the captive-bred progeny of legally imported “pre-act” Figis. Both sides offered contradictory testimony as to this question at trial. After hearing the evidence, the jury returned a verdict against the Crutchfields.

II. Discussion

Appellants urge this Court to reverse their respective convictions based on the prosecutor’s pervasive misconduct during their trial. “Reversal on the basis of prosecutorial misconduct requires that the conduct be ‘so pronounced and persistent that it permeates the entire atmosphere of the trial.’ ” United States v. McLain, 823 F.2d 1457, 1462 (11th Cir.1987) (quoting United States v. Weinstein, 762 F.2d 1522, 1542 (11th Cir.1985), cert. denied, 475 U.S. 1110, 106 S.Ct. 1519, 89 L.Ed.2d 917 (1986)). Moreover, this Court recognizes that in evaluating a prosecutorial misconduct claim, reversal is warranted only if the misconduct prejudicially affects the substantial rights of the accused. *1100See United States v. Eyster, 948 F.2d 1196, 1206 (11th Cir.1991). The focus of our inquiry, therefore, is whether the appellants received a fair trial. After reviewing the record in this case, we find it is replete with examples of unquestionable prosecutorial misconduct that prejudiced both appellants to such a degree as to warrant a reversal of their convictions.

A. Prosecutor’s Personal Interest in this Case

The record in this case clearly indicates that the prosecutor, from the outset of the Crutchfields’ trial, was more than professionally interested in its outcome. The prosecutor himself was a Herpetologist (an expert in the field of reptiles) and apparently had been a previous customer of the Crutchfields’. Because of the prosecutor’s expertise, he was well versed in “reptile jargon” and familiar with the business aspect of both collecting and selling exotic reptiles. While this case concerned the fairly simple issue of whether the Crutchfields had illegally imported an endangered iguana species into the United States, many of the prosecutor’s inquiries on direct and cross examination had little, if anything, to do with the resolution of this issue. Indeed, his examination of the witnesses included countless irrelevant inquiries seemingly designed only to display to the jury his own expertise in the reptile field.1

In response to the prosecutor’s numerous lines of irrelevant questioning, appellants’ counsel raised several relevancy objections. The court repeatedly sustained these objections and instructed the prosecutor to “move along”; however, the prosecutor ignored the court’s instructions. Finally, as the quantity of irrelevant inquiries increased, the court was prompted to offer the following instruction sua sponte:

. Now, seriously, you have impressed all of us with your knowledge of this subject matter and it is quite interesting. However, some of it is a little far afield from the issues which are framed by the indictment ... but it will help us move along if you try to be a little more careful about the areas you go into.”

Even this specific instruction by the court, however, proved ineffective. The prosecutor continued throughout the course of the trial to waste the valuable resources of the court by refusing to focus his inquiries on the issue before the jury.2

B. Improper Questioning of Witnesses

Several lines of questioning pursued by the prosecutor in this case were not only completely irrelevant, but also constituted improper character evidence under Rules 404, 608 and 609 of the Federal Rules of Evidence. Although the record provides several examples of these improper and highly prejudicial prosecutorial inquiries, we will specifically address in detail only the two most egregious illustrations.3

1. Direct Examination of Nora Dietlein.

The first example arose during the prosecutor’s direct examination of Nora Dietlein, a previous close friend and business partner of both Penny and Tom Crutchfield. After asking several preliminary questions regarding Dietlein’s past relationship with the Crutch-fields and their children, the prosecutor began his inquiry into the cause of the eventual disintegration of Dietlein’s friendship and business association -with the Crutchfields. The following colloquy occurred:

*1101“Q. And did there come a time in your life when you had moved to British Columbia when you did something to hurt Tom and Penny that, perhaps, you wish you hadn’t done now?

“A. Yes.

“Q. What did you do?

“A After repeated phone calls from Tom harassing us in Canada, and my 85 year old mother had arrived to live with us and she got the last one and heard it, I phoned down at 8:80 p.m. our time, which meant there would be no one in their business premises, and I left a message on their answering machine that was hurtful to Penny.

“Q. What did you say in the message?

“A. I, I said, does Penny, do Penny’s children know, as far as I remember now, that she was pregnant with another man’s child when she married Tommy?”

Immediately after hearing this response from Dietlein, appellants’ counsel objected. The court called both counsel to the bench and inquired as to the question’s relevancy. The prosecutor then attempted to justify his inquiry by stating that he was “anticipating” that appellants’ counsel would use the recorded message on cross examination in an effort to impeach Dietlein. He claimed to have introduced the substance of the message on direct to “draw the sting” from this anticipated attack. When the court asked appellants’ counsel if he intended to use the recording on cross, however, counsel responded that he had never even heard of the incident, much less the existence of such a recording. The court properly struck Diet-lein’s testimony and instructed the jury to disregard the last answer given by Dietlein. Additionally, the court warned the prosecutor to be more careful in the future of what he brought up “in anticipation” of the defense.

2. Cross Examination of Robert Harding

A second example of the prosecutor’s misconduct in his questioning of the witnesses came during his cross examination of defense witness Robert Harding. Harding was introduced by the defense as both a friend and former employee of Tom Crutchfield. During the course of his direct testimony, Harding stated that he had on occasion smoked marijuana with another of Crutchfield’s former employees. The prosecutor began his cross-examination of Harding by asking:

“Q. You live down in that Ft. Myers Beach area for a long time?

“A. No sir.

“Q. Where did you live during most of that 20 years in Lee County?

“A. Most of the time I lived in either Ft. Myers or for the past approximately eight and a half years in Lehigh Acres.

“Q. Are you aware of the mullet boats coming in at night without lights around that area carrying the square grouper off of the boats there?”

At this point, appellants’ counsel lodged an immediate objection. Before the court could rule on the objection, however, the prosecutor quickly asked:

“Q. Do you know what a square grouper is?”

Appellants’ counsel again objected. The district court sustained this objection and began to instruct the prosecutor to wait until it had ruled on objections before continuing with his inquiry. The prosecutor, however, interrupted the court mid-sentence and posed the following question to Harding:

“Q. Weren’t you involved in obtaining bales of pot?”

After hearing this final question, the court was forced to clear the courtroom. Appellants’ counsel then moved for a mistrial based on the prosecutor’s repeated improper inquiries designed solely to suggest the poor character of Harding in violation of Federal Rule of Evidence 608(b). The prosecutor responded by suggesting that his questions to Harding were somehow appropriate because of his personal expertise in the prosecution of drug cases and because of his knowledge of many illegal activities in the area where the witness resided. He also contended that defense counsel, during his direct examination of Harding, had opened the door to this line of questioning. The court disagreed and sustained appellants’ objection. Upon their return, the court instructed the jury to disregard the last several questions asked by the prosecutor as they had been ruled improper.

We do not believe that a detailed explanation regarding the impropriety of ei*1102ther of these lines of questioning is necessary. As to the testimony elicited from Nora Dietlein, we would hope that even the most inexperienced trial attorney would recognize that this testimony was clearly irrelevant, improper and prejudicial to the accused.4 The prosecutor’s decision to-place the “sexual character” of Penny Crutchfield before the jury in a case involving the importation of iguanas is simply inexcusable. Such testimony could have been elicited only in an effort to damage Mrs. Crutchfield’s character in the eyes of the jury.

Similarly, the prosecutor’s questioning of Robert Harding on cross examination was unquestionably out of line and an obvious violation of Federal Rule of Evidence 608(b).5 We disagree with the prosecutor’s contention at trial that Harding’s single statement regarding his past occasional marijuana use entitled the prosecutor to follow up ■with questions about Harding’s knowledge of “mullet boats” and “square grouper.”6 While questions regarding the extent of Harding’s personal use of marijuana may have been proper on cross examination, questions designed to implicate Harding as a large scale buyer or distributor of marijuana clearly were not. These questions were outside the scope of direct examination. Moreover, the prosecutor apparently lacked even a good faith basis for these allegations and was simply attempting to discredit Harding’s testimony with the jury.7

C. Disobedience of the District Court’s Instructions and Rulings

Throughout the two-week trial, the prosecutor displayed his apparent disrespect for the court by continuously disobeying the trial judge’s instructions and rulings. To demonstrate, the record reflects numerous instances in which the prosecutor simply ignored the court’s rulings on relevancy and improper character evidence objections. Instead of moving to another line of questioning after the court sustained these objections by appellants’ counsel, the prosecutor continued on several occasions to make the same types of inquiries.8

Moreover, the prosecutor completely ignored the court’s in limine evidentiary ruling excluding any testimony by the government regarding Tom Crutchfield’s offer to sell the Figis to a particular group of Italians. During the prosecutor’s opening argument, he mentioned that the jury would hear testimony from government witness Tom Stanley that Tom Crutchfield had offered to sell the Figis to Italian customers. After opening statements were completed, howev*1103er, the court excluded any testimony by Stanley regarding this alleged sale to the Italians. Ignoring this ruling, the prosecutor asked during his direct examination of Stanley:

“Q. Do you know whether Mr. Crutch-field ever attempted to sell these animals to anybody?

“A. Yes.

“Q. Did he?

“A. As far as I know, what I would assume would be a sales, yeah.

“Q. Was there any particular customers that come to mind?

“Appellants’ Counsel: Objection, Your Honor.

“The Court: You’re not going to ask him what we have already discussed at sidebar, are you?

“The Prosecutor: Well, I’d like to revisit that issue at some time.

“The Court: Well, I’ve ruled on that issue so, if you have other information, then go ahead and ask the question.”

The prosecutor apparently had no other information to offer the court other than that which the court had already ruled was improper. By posing these questions to Stanley, however, the prosecutor was able to leave with the jury at least the suggestion that Crutchfield had attempted to sell the Figis when no admissible evidence as to this sale existed.

III. CONCLUSION

We conclude that the record supports appellants’ contention that the prosecutor was guilty of multiple and continuing instances of intentional misconduct that prejudicially affected the substantial rights of both appellants. While we recognize that a United States Attorney should “prosecute with earnestness and vigor,” we also acknowledge “his duty to refrain from improper methods calculated to produce a wrongful conviction.” United States v. Eason, 920 F.2d 731, 736 n. 8 (11th Cir.1990) (quoting Berger v. United States, 295 U.S. 78, 88, 55 S.Ct. 629, 633, 79 L.Ed. 1314, 1321 (1935)).

On numerous occasions during this trial, the prosecutor flagrantly violated Rules 404, 608 and 609 of the Federal Rules of Evidence by repeatedly and improperly attacking the character of the defendants and their witnesses. These attacks often continued even after the court instructed the prosecutor as to their impropriety. The prejudicial effect of this misconduct cannot be disputed, as this ease turned largely on the jury’s credibility determinations of the several witnesses who testified. Furthermore, we are unpersuaded by the government’s argument that any misconduct which may have occurred during the trial was neutralized by the court’s several curative instructions and, therefore, did not affect the outcome of the proceeding. As we noted in United States v. McLain, 823 F.2d 1457, 1462 n. 8 (11th Cir.1987), “a jury cannot always be trusted to follow instructions to disregard improper statements.” When improper inquiries and innuendos permeate a trial to such a degree as occurred in this case, we do not believe that instructions from the bench are sufficient to offset the certain prejudicial effect suffered by the accused. Accordingly, we REVERSE and REMAND this case for a new trial.

REVERSED and REMANDED.

The original panel decision in this case affirming a jury’s verdict against the defendants was vacated, Sherrod v. Berry, 835 F.2d 1222 (7th Cir.1988), and the case set for rehearing en bane. We reverse and remand for a new trial.

COFFEY, Circuit Judge.

I.

Because we are remanding the case for a new trial, we recite only those facts necessary to our decision. (A more complete recitation of the facts is detailed in Sherrod v. Berry, 827 F.2d 195, 198-200 (7th Cir.1987) and 827 F.2d at 211-13 (dissenting opinion)). On December 8, 1979, the operator of Ziggy’s Plant and Gift Shop in Joliet, Illinois, reported to the police that a robbery had just taken place. Willie Berry, a Joliet police officer on patrol in the area, heard a police radio dispatch recounting the robbery and describing the suspect. Officer Berry told his partner, Officer Richard Klepfer, that the description of the robbery suspect fit Gary Duckworth. Berry testified at trial that he previously knew Duck-worth to have been a suspect in other as-saultive crimes, including “robberies and purse snatchings.” After receiving another radio communique that the suspect had returned to the area of the robbery, Officer Berry and his partner decided to investigate and proceeded to the crime scene. En route, the officers observed two men sitting in a 1969 Cadillac in the bank parking lot adjacent to Ziggy’s.

The 1969 Cadillac exited the parking lot and pulled onto the street where Officers Berry and Klepfer were patrolling. Officer Berry activated the squad’s “mars lights” and directed the driver of the Cadillac to pull over. As the officer’s car approached the Cadillac, Berry recognized Duckworth as the passenger in the Cadillac. At that point in time, Officer Berry believed he was apprehending the perpetrator of the Ziggy’s robbery, who was probably armed and thus considered dangerous. As the suspect’s car slowed to a stop, Officer Berry and his partner exited the patrol car and removed their guns from their holsters believing the automobile stop to be of high risk. Officer Berry and his partner assumed positions outside the police vehicle with their guns pointed at the occupants of the Cadillac, covering the suspects from separate angles. Following accepted police procedures, Berry ordered the suspects to raise their hands. The two suspects failed to comply with the command, and Berry had to order them to raise their hands three times before the suspects complied; this recalcitrance on their part to follow the raised arm order further aroused Berry’s suspicion as to the imminent danger confronting him. Berry testified that “it seemed to me as though the passenger [Duckworth] was looking at the driver [Ronald Sherrod] as more or less ‘what are we going to do next?’ ”

Officer Berry asked his partner, who had also drawn his gun, if he had the suspects under cover. Officer Klepfer responded in the affirmative. At this time, Officer Berry raised his gun and cautiously approached the Cadillac. Patrolman Berry testified that while looking into the vehicle and approaching the suspect, he observed the driver make a “quick movement with his hand into his coat ... [as if] he was going to reach for a weapon.” At that point, Officer Berry fired his revolver at Sherrod, killing him instantly.

Lucien Sherrod, the father of the deceased, filed a 42 U.S.C. § 1983 action individually and as administrator of his son’s estate against Officer Berry, the City of Joliet, and the Joliet Chief of Police. The first count of the complaint alleged that: (a) Officer Berry violated 42 U.S.C. § 1983 when he shot and killed Ronald Sherrod; (b) Chief of Police Breen and the City of Joliet violated 42 U.S.C. § 1983 through their improper policy regarding the use of deadly force; and (c) the City failed to *804adequately train the Joliet police officers concerning correct procedures for making felony stops of vehicles as well as the use of deadly force. The second count alleged that the defendants, in depriving the plaintiff-appellee (Sherrod) of his right to raise a family, deprived him of his constitutional right of due process of law. During the trial, evidence was received over the objection of the defendants-appellants that a search of the deceased (Sherrod) failed to disclose that he was armed with a weapon. The trial judge admitted the evidence, reasoning that “had plaintiff been prevented from introducing this evidence, the record would have been such that the jury would have been left to speculate on whether Berry was justified in thinking that the claimed movement by Sherrod posed a danger to the police officer. This would not have been fair.” Sherrod v. Berry, No. 80 C 4117, mem. op. at 38 (N.D.Ill. November 15, 1985) [available on WESTLAW, 1985 WL 3926]. The jury ultimately found for the plaintiff on both counts and awarded $1,601,700 in damages.

II.

Defendants urge this court to reverse the jury’s verdict, arguing that the trial court’s receipt of evidence demonstrating that Ronald Sherrod was unarmed when Officer Berry discharged his weapon is not relevant to the question of whether Officer Berry reasonably believed that the use of deadly force was justifiable at the time of the shooting. The district court not only found that evidence as to whether Sherrod was unarmed was both relevant and material to determining whether Berry acted reasonably under the circumstances, but even implicitly stated on the record in its written findings dealing with a motion for a new trial that it would have been prejudicial to the plaintiff had the evidence not been received. Generally a “district court has broad discretion to determine the admissibility of evidence, and thus [this court] will reverse the court’s evidentiary rulings only upon a clear showing of abuse of discretion.” United States v. Garver, 809 F.2d 1291, 1297 (7th Cir.1987); see also Davis v. Lane, 814 F.2d 397, 399 (7th Cir.1987). But “[discretion does not mean immunity from accountability.” 1 Weinstein’s Evidence, ¶ 401[01], Where a district court applies an improper legal standard as the basis for allowing to be received in evidence facts not only irrelevant but also prejudicial to the determination of a party’s liability (in this case Officer Berry’s liability), the court’s decision obviously constitutes an abuse of discretion.

Under Fed.R.Evid. 401,1 “relevancy is a relationship between a proferred item of evidence and a ‘fact that is of consequence to the determination of the action.’ ” 1 Weinstein’s Evidence, ¶ 401[03], pp. 401-17. “Whether or not a fact is of consequence is determined not by the Rules of Evidence but by substantive law.” Id. at 401-19. Thus, before the district court could properly have received evidence that Sherrod was unarmed at the time of the shooting, the district court had to find that this fact was relevant to the determination of Officer Berry’s liability in the first instance.

In Lester v. City of Chicago, 830 F.2d 706 (7th Cir.1987) this court applied an “objective reasonableness under the circumstances” standard to Fourth Amendment excessive force and arrest claims. Lester stated that “the Fourth Amendment test measures [the] ... objective reasonableness [of an officer’s actions] under the circumstances.” 830 F.2d at 711. In phrasing the test set forth in Lester as one of “objective reasonableness under the circumstances,” it is obvious that “under the circumstances” refers only to those circumstances known and information available to the officer at the time of his action (firing the fatal shot). When a jury measures the objective reasonableness of an officer’s action, it must stand in his shoes and judge the reasonableness of his actions based *805upon the information he possessed and the judgment he exercised in responding to that situation.

Knowledge of facts and circumstances gained after the fact (that the suspect was unarmed) has no place in the trial court’s or jury’s proper post-hoc analysis of the reasonableness of the actor’s judgment. Were the rule otherwise, as the trial court ruled in this instance, the jury would possess more information than the officer possessed when he made the crucial decision. Thus, we are convinced that the objective reasonableness standard articulated in Lester requires that Officer Berry’s liability be determined exclusively upon an examination and weighing of the information Officer Berry possessed immediately prior to and at the very moment he fired the fatal shot. The reception of evidence or any information beyond that which Officer Berry had and reasonably believed at the time he fired his revolver is improper, irrelevant and prejudicial to the determination of whether Officer Berry acted reasonably “under the circumstances.”

The record of the first trial substantiates Officer Berry’s testimony that he fired at Sherrod because he reasonably believed in the split second he had to react to Sher-rod’s furtive, rapid movement, that he and his partner were in imminent danger of death or great bodily harm. As Officer Klepfer testified, Sherrod made a “quick movement with his hand into his coat” and “there was no doubt in my mind when he started to move, he was going to reach for a weapon of some type.” Officer Berry never claimed that he actually saw a weapon, but stated that he simply reacted to what a reasonable person would consider to be a life-threatening and imminently dangerous situation. Thus, absent a constitutional violation, appellate judges would be well advised not to second-guess an officer’s split-second reasonable judgment to protect himself and those around him through the use of deadly force; rather, courts and juries must determine the propriety of the officer’s actions based upon a thorough review of the knowledge, facts and circumstances known to the officer at the time he exercised his split-second judgment as to whether the use of deadly force was warranted.

Our holding is consistent with both today’s opinion in Ford v. Childers, 855 F.2d 1271 (7th Cir.1988), and the Supreme Court’s decision in Tennessee v. Garner, 471 U.S. 1, 105 S.Ct. 1694, 85 L.Ed.2d 1 (1985). In both cases, the respective courts reviewed the totality of the circumstances known to the police officer at the time of the incident in determining whether the officer acted reasonably. (Ford also involved an officer’s use of deadly force in the apprehension of a suspect.) In Garner, the Court stressed the rule that it is reasonable to use deadly force if the officer, when exercising his or her reason and judgment, has probable cause to believe that the suspect poses a threat of death or serious physical harm to the officer or others and, whenever possible, warns the suspect before firing. Id. at 11-12, 105 S.Ct. at 1701. The Court specifically stated:

“Thus, if the suspect threatens the officer with a weapon or there is probable cause to believe that he has committed a crime involving the infliction or threatened infliction of serious physical harm, deadly force may be used if necessary to prevent escape and if, where feasible, some warning has been given.”

Id. Similarly, as here, when an officer believes that a suspect’s actions places him, his partner, or those in the immediate vicinity in imminent danger of death or serious bodily injury, the officer can reasonably exercise the use of deadly force. As aptly noted in Young v. City of Killeen, Tx., 775 F.2d 1349, 1353 (5th Cir.1985), “no right is guaranteed by federal law that one will be free from circumstances where he will be endangered by the misinterpretation of his acts.” Id. at 1353 (emphasis added).

Our holding today is likewise consistent with Davis v. Freels, 583 F.2d 337 (7th Cir.1978). There Wallace Davis sued Joseph Freels, a Chicago police officer, under 42 U.S.C. § 1983 after Freels shot him in the back. The facts in Davis establish that Officer Freels and his partner believed Davis and another individual were wanted *806in connection with a shooting incident. Police officers confronted Davis and his companion and ordered the two men to walk over and place their hands on the car. Officer Freels testified that as Davis turned to put his hands on the car, he (Officer Freels) “saw a sudden movement with his [Davis’s] right elbow in a backward direction.” Responding to this sudden and suspicious movement, Officer Freels fired his revolver, previously drawn as a precautionary measure, at the suspect. This court upheld the jury’s verdict that Davis was not deprived of any rights under 42 U.S.C. § 1983 and quoted with approval from 6th Am.Jur.2d Assault & Battery, § 161 at 135 (1963):

“In a civil action for assault, the defendant’s belief that the plaintiff intended to do him bodily harm cannot support a plea of self defense unless it was such a belief as a reasonable person of average prudence would have entertained under similar circumstances. It is not necessary that the danger which gave rise to the belief actually existed; it is sufficient that the person resorting to self defense at the time involved reasonably believed in the existence of such a danger, and such reasonable belief is sufficient even where it is mistaken. In forming such reasonable belief a person may act upon appearances. In other words, it is sufficient that the danger was reasonably apparent."

583 F.2d at 341 (emphasis added). In this case the investigation conducted after the shooting revealed that Sherrod was unarmed. But at the time Officer Berry made the crucial decision to discharge his revolver, he was unaware of this fact. Officer Berry testified that he believed that he was apprehending an armed and dangerous suspect who had just committed a robbery and further that this suspect made a “quick movement with his hand into his coat ... [as if] he was going to reach for a weapon.” The veracity of Officer Berry’s testimony and the reasonableness of his actions based upon the totality of the information he possessed at the time of the shooting are questions we leave for a properly informed and instructed jury on remand. In short, we categorically reject the district court’s assertion that fairness requires that the jury be presented with facts unknown and unavailable to Officer Berry at the time of the shooting (that Sherrod was unarmed). To the contrary, we hold that justice requires that the jury analyze and weigh the reasonableness of Officer Berry’s conduct just as he was compelled to do in a split second on that fateful day, without the knowledge that Sherrod was unarmed. The Sioux Indians have a prayer that asks for this wisdom: “Grant that I may not judge another until I have walked a mile in his moccasins.”

Our holding today should not be interpreted as establishing a black-letter rule precluding the admission of evidence which would establish whether the individual alleging a § 1983 violation was unarmed at the time of the incident. Clearly, the credibility of the witness “can always be attacked by showing that his capacity to observe, remember or narrate is impaired.” 3 Weinstein’s Evidence ¶ 607[04] p. 607-55. Further, “impeachment by contradiction is a technique well recognized in the federal courts by which specific errors in the witness’s testimony are brought to the attention of the trier of fact.” Id. at ¶ 607[05] p. 607-76. For example, if an officer testifies that “I saw a shiny, metallic object similar to a gun or a dangerous weapon in the suspect’s hand,” then proof that the suspect had neither gun nor knife would be material and admissible to the officer’s credibility on the question of whether the officer saw any such thing (and therefore had a reasonable belief of imminent harm). See, e.g., Davis v. Lane, 814 F.2d 397, 399 (7th Cir.1987) (wherein we held that the existence of a homemade knife or shank was clearly relevant to the question of the reasonableness of Officer Hull’s reaction where Officer Hull thought that he saw the inmate reach for something shiny, perhaps a homemade knife or shank). But, as in this case, if the officer says “I saw the suspect reach quickly for his pocket,” then proof of the contents of the pocket does not contradict the officer’s testimony. Here, Officer Berry did not testify that he saw an *807object in Sherrod's coat or in his hand; rather, Officer Berry testified only that he saw Sherrod make a “quick movement with his hand into his coat.” Thus, evidence that Sherrod was unarmed is irrelevant for impeachment purposes, and the jury must determine the reasonableness of Berry’s actions limited to the facts known to Berry when he acted, no more and no less. Having ruled that evidence establishing that Sherrod was unarmed is immaterial to the question of the reasonableness of the officers’ actions under the totality of the circumstances, we are impelled to conclude that the trial court’s error in admitting the evidence was also unfairly prejudicial to the defendants.

Evidence that Sherrod was unarmed (information not available to Officer Childers) is not only irrelevant, as explained above, but also tends to induce the trier of fact to premise its ultimate determination of liability on an improper basis: namely, to infer from the fact that Sherrod was unarmed that Officer Childers’ use of his weapon was unreasonable. Such an inference is manifestly inappropriate, and a new trial is the only way to remedy the evidentiary error.

Judges must never forget that “[pjeace officers stand at the front of law and the ordering processes of society,” Grandstaff v. City of Borger, 767 F.2d 161, 166 (5th Cir.1985), cert. denied, 480 U.S. 916, 107 S.Ct. 1369, 94 L.Ed.2d 686 (1987).2 Nevertheless, we hasten to point out that we must and will continue to be vigilant in ensuring that officers, in exercising the critical judgment of when and where to use deadly force, will always be subject to review based upon the very information known to the officer at the time of his or her actions under the Davis v. Freels standard:

“It is not necessary that the danger which gave rise to the belief actually existed; it is sufficient that the person resorting to self defense at the time involved reasonably believed in the existence of such a danger, and such reasonable belief is sufficient even where it is mistaken. In forming such reasonable belief a person may act upon appearances. In other words, it is sufficient that the danger was reasonably apparent.”

583 F.2d at 341 (emphasis added). On remand it will be for the jury to determine whether Officer Berry ultimately acted reasonably based upon the knowledge and information he possessed at the time he fired the fatal shot. It is precisely this unknown danger that our law enforcement officers face daily, and we refuse in this instance to second-guess objectively reasonable police decisions.

III.

Because we reverse and remand for a new trial, we need not discuss the district court’s other evidentiary rulings or jury instructions. We leave these questions for the district court on remand to decide in light of this court’s prior discussions of *808those matters, specifically those found in our earlier vacated opinion.

For the afore-mentioned reasons, we reverse and remand this case for further proceedings in the district court consistent with this decision.

RIPPLE, Circuit Judge,

concurring.

I concur in the judgment. In my view, the district court’s failure to give defendant’s instruction No. 10 and the giving of plaintiffs instruction No. 28.11 constitute reversible error and requires a new trial. With respect to the admissibility of evidence as to whether Mr. Sherrod had a weapon, I agree essentially with the view stated by Judge Cummings in these en banc proceedings.

CUMMINGS, Circuit Judge,

with whom CUDAHY, Circuit Judge, joins, dissenting.

The majority’s opinion misrepresents the parties and events involved in the killing of Ronald Sherrod, misinterprets the applicable legal standards and rules and allows an argument waived by the defendants to prevail. The result is a miscarriage of justice. Ford v. Childers, 855 F.2d 1271 (7th Cir.1988) (en banc) (Cudahy, J., concurring).

The majority’s version of the facts presents a good police officer in a bad situation confronting a recalcitrant suspect. The record presents a bad cop with a history of unnecessary violence, acting in violation of police procedure, confronting an innocent victim. This characterization is important because the issue on rehearing is whether the district court clearly abused its discretion when it admitted evidence it determined was relevant to the credibility of Berry’s version of the events leading to his shooting and killing Ronald Sherrod. Therefore, before addressing the legal issues we find it necessary to correct several crucial misstatements of the facts.

I

The majority opinion states that “[f]ol-lowing accepted police procedures, Berry ordered the suspects to raise their hands,” then it characterizes their reaction as “recalcitrance” which “further aroused Berry’s suspicion.” Maj. op., supra at 803. Berry testified to, and the majority makes much of, the fact that he had to order Sherrod and Duckworth to raise their hands three times before they complied. Although Sherrod and Duckworth may not have responded until Berry completed all three of his vulgar and abusive commands to “get their motherfucking hands up” (Tr. at 406),1 this “failure” cannot accurately or fairly be labeled recalcitrance. It is not clear from the record how much time elapsed between each of these commands. But if any time at all elapsed, it must have been very little because Officer Klepfer testified that from the time Sherrod responded and stopped his car until the time Berry shot and killed him only “10, 12 seconds maybe” elapsed. (Tr. at 1241).

The majority also mischaracterizes the nature of the stop and the danger that Berry’s partner Officer Klepfer perceived. The majority presents Duckworth as probably armed and dangerous. Nothing in the record supports this view. The police dispatcher described the incident at Ziggy’s over the radio as follows, “[apparently was a sneak thief. He took the money, was chased by an employee. The employee got the money back.” (Tr. at 1117). Nothing in this or any of the other radio dispatches concerning the alleged sneak thievery at Ziggy’s indicated that the suspect was armed or dangerous. Although Berry stated that he never heard the broadcast *809calling the suspect a “sneak thief,” he admitted that he had no reason to believe that Duckworth had ever committed any violent acts. (Tr. at 417-418). As to reasonable suspicions concerning Sherrod, the radio dispatches reported the suspect as one man, on foot, so that Berry had no information indicating that Sherrod was in any way involved in the sneak thievery or was otherwise armed or dangerous.

In addition, neither the manner in which the stop was made nor Sherrod and Duck-worth’s response indicated any danger. Sherrod responded immediately to Berry’s hand signal to pull over (Tr. at 445), and sat “peacefully” in the car (Tr. at 502 and 1156), doing nothing that indicated to Berry that he was going to create any problems. (Tr. at 1203). The majority states that “Officer Berry and his partner ... be-liev[ed] the automobile stop to be of high risk.” However, the record indicates that Officer Klepfer did not feel he was in any danger at the time the stop was made (Tr. at 1280), or when Sherrod moved his hand toward his pocket. (Tr. at 1280-1281). Rather, Officer Klepfer’s sense of security was so great, due to Sherrod’s “compliance,” that when Berry asked Klepfer “Do you have him?” Klepfer responded “Yes,” but at the same time relaxed his finger from the trigger. (Tr. at 1285).

Finally, the majority’s version creates the impression that the stop was a proper cautious one in which the officers assumed protective positions. This was not the case. In a proper felony stop, Berry would have stopped his car behind Sherrod’s and, from a position of relative safety behind the open doors of the squad car, ordered Sherrod and Duckworth out of their vehicle. Berry admitted that the circumstances would have allowed him to make a proper felony stop (Tr. at 525), but that instead he incautiously chose to confront Sherrod’s car head-on. By violating proper police procedure, Berry unnecessarily placed himself in the position he later perceived as dangerous to himself and his partner (Tr. 502-503), and which led to the death of Ronald Sherrod.

II

The majority is able to hold that the district court clearly abused its discretion because the majority finds that the district court applied an improper legal standard. Maj. op., supra at 804. Unfortunately, it is the majority, not the district court, that applies an improper legal standard. The majority correctly identifies the applicable legal standard in a Fourth Amendment excessive force claim as a test of “ ‘objective reasonableness under the circumstances.’ ” Id. at 804 (quoting Lester v. City of Chicago, 830 F.2d 706, 711 (7th Cir.1987)), but the majority then fails to apply this objective test to the circumstances that resulted in Sherrod’s death. Instead, by narrowly defining what constitutes the circumstances and what constitutes impeachment, the majority uses a subjective test that gives to the law enforcement officer the wide discretion it denies the district court.

In the majority’s narrow view “it is obvious that ‘under the circumstances’ refers only to those circumstances known and information available to the officer at the time of his action (firing the fatal shot).” Maj. op., supra at 804. Therefore, the “[kjnowledge of facts and circumstances gained after the fact (that the suspect was unarmed) has no place in the trial court’s or jury’s proper post-hoc analysis of the reasonableness of the actor’s judgment.” Id.2

The limitations the majority imposes on what constitutes the circumstances are neither intuitively obvious nor, more impor*810tantly, obvious from this Court’s opinion in Lester. In Lester we cited Tennessee v. Garner, 471 U.S. 1, 105 S.Ct. 1694, 85 L.Ed.2d 1, for the proposition that the balancing of interests between the individual and the government depends on “whether the totality of circumstances justifies a particular sort of ... seizure.” Lester, 830 F.2d at 711 (emphasis added). Then, as an example of what factors courts should consider in determining the totality of the circumstances, we cited Gilmere v. City of Atlanta, 774 F.2d 1495 (11th Cir.1985) (en banc), certiorari denied, 476 U.S. 1115, 106 S.Ct. 1970, 90 L.Ed.2d 654 (1986). In Gil-mere, the police received a report that Thomas Patillo had pulled a gun from the trunk of his car and threatened the driver of another vehicle. When the police arrived at Patillo’s home and ordered him to the police car for questioning, he put up some resistance and attempted to flee. A scuffle ensued and an officer shot Patillo in the stomach and killed him. Id. at 1496-97. In evaluating Patillo’s estate’s Fourth Amendment claim against the police and evaluating the reasonableness of the officer’s action under the totality of the circumstances, the Eleventh Circuit en banc considered that “the police had little cause to believe Patillo himself to be dangerous. Given his small size, intoxicated state, and lack of a weapon, he clearly posed little threat to their safety when they initially began escorting him to the patrol car.” Id. at 1502 (emphasis added). Obviously, the after-the-fact information that the suspect/victim was unarmed had a place in the Eleventh Circuit’s post-hoc analysis of the totality of the circumstances. Consistent with Lester and Gilmere, the after-the-fact information that Sherrod was unarmed had a place in the jury’s post-hoc analysis of the totality of the circumstances leading to Berry’s shooting Sherrod at point-blank range.

In addition to its relevance as part of the totality of the circumstances leading to the killing of Ronald Sherrod, the evidence that Sherrod was unarmed is relevant as part of the testimony helping the jury understand the type of movement Sherrod made that precipitated his death. Evidence is relevant if it has “any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence.” Fed.R.Evid. 401. The manner in which Sherrod reached into his pocket is of great consequence to the determination of the action — it was this movement that caused Berry to shoot and kill him.

The jury was presented with conflicting versions of how Sherrod reached into his jacket. The descriptions of the action, both verbal and demonstrative, were repetitive and confusing. Adding to the confusion was the element that different versions came not only from different witnesses but also from the same witness on direct examination, cross-examination and through impeaching prior testimony read to the jury as admissions. See Tr. at 1133, 1181-1186, 1240, 1266, 1280-1281, 1282-1283. Whether Berry’s response was reasonable depends in large measure on the nature of Sherrod’s movement, and the fact that Sherrod did not have a gun has the tendency to make some of the conflicting versions more probable. The fact that there was no gun was relevant to help the jury evaluate the conflicting testimony and determine the type of movement Sherrod made.

In addition to holding that the fact that Sherrod was unarmed is irrelevant to the determination of the reasonableness of Berry’s action, the majority holds that “evidence that Sherrod was unarmed is irrelevant for impeachment purposes.” Maj. op., supra at 806. Once again the majority has taken too narrow a view of the applicable law. To be relevant for impeachment purposes evidence need not directly counter “black” with “white” or “yes” with “no;” it need only detract from the credibility of a witness or the veracity of his testimony.

The majority admits that “if an officer testifies that T saw a shiny, metallic object similar to a gun or a dangerous weapon in the suspect’s hand,’ then proof that the suspect had neither a gun nor a knife would be material and admissible to the officer’s credibility on the question of *811whether the officer saw any such thing.” Maj. op., supra at 806. The majority sees this as an example of “no” countering “yes;” however, that is not why this evidence would be impeaching. The officer in the majority’s example testified that he saw a gun. The after-the-fact determination that no gun existed does not mean that the officer did not sincerely believe that he saw one, but it does make it less likely that the officer believed he saw one or, if he did believe he saw one, tends to make that belief unreasonable. It is also relevant to the observational abilities of the officer and reflects upon the veracity of his testimony. Simply stated, in the majority’s example the impeachment goes not to the existence of a weapon, but to the probability that the officer’s version of the events is credible.

For this reason, the majority’s attempt to differentiate the above scenario from Berry’s situation fails. The majority states that “Officer Berry did not testify that he saw an object in Sherrod’s coat or in his hand; rather, Officer Berry testified only that he saw Sherrod make a ‘quick movement with his hand into his coat;’ ” from this the majority concludes “[t]hus, evidence that Sherrod was unarmed is irrelevant for impeachment purposes.” Id. at 807. But the fact that Sherrod was unarmed does not necessarily mean that Berry acted unreasonably.3 Instead, as in the majority’s example, the fact that no gun existed does make it less likely that Berry thought he was in danger and makes the evidence impeaching in the same manner as in the majority’s example. The object of the relevancy determination is the reasonableness of Berry’s conclusion that he was in danger and the credibility of his testimony at trial. Thus the fact that Sherrod was unarmed is relevant for the same reasons that the evidence is relevant in the majority’s example.

In conclusion, I find the evidence that Sherrod was unarmed relevant to the totality of the circumstances and to determining the credibility of Berry’s testimony. The majority finds the evidence irrelevant. However, the outcome of this appeal should not depend on which version garners a majority of votes. At this stage of the proceedings we are not making a de novo determination of whether to admit the evidence; we are reviewing the determination of the district court and can reverse only if the district court clearly abused its discretion. United States v. Laughlin, 772 F.2d 1382, 1392 (7th Cir.1985) (“a trial court possesses broad discretion in determining the relevance of proffered evidence, and its decision will not be disturbed on appeal absent a clear showing that the court abused its discretion”). Regardless of which view would prevail in an evidentiary ruling made today, the district court’s decision to admit this evidence cannot be reversed because there has been no clear showing that the court abused its discretion.

Ill

Finally, although the majority never argues in the alternative that if the evidence that Sherrod was unarmed is relevant it should still be excluded because it is more prejudicial than probative, the majority nevertheless attempts to buttress its holding that the evidence should be excluded by stressing its prejudicial impact. See maj. op., supra at 805 and 807. But the issue of whether its prejudicial impact outweighs its probative value has been waived by the defendants.

To preserve an issue for appellate review, a party must make a proper objection that alerts the trial court and the opposing party to the specific grounds for the objection. United States v. Wynn, 845 F.2d 1439, 1442 (7th Cir.1988); see also Laughlin, 772 F.2d at 1391-92. “An objection is proper only if ‘a timely objection or motion to strike appears of record, stating the *812specific ground of objection, if the specific ground was not apparent from the context....’” Wynn, 845 F.2d at 1442 (quoting Fed.R.Evid. 103(a)(1)). The specific ground for reversal of an evidentiary ruling on appeal must be the same specific ground as that stated in the proper objection at trial. Id.; see also United States v. Taylor, 800 F.2d 1012, 1017 (10th Cir.1986), certiorari denied — U.S. —, 108 S.Ct. 123, 98 L.Ed.2d 81; United States v. Medina, 755 F.2d 1269, 1275-76 (7th Cir.1985); United States v. Hickerson, 732 F.2d 611, 613 (7th Cir.), certiorari denied, 469 U.S. 846, 105 S.Ct. 159, 83 L.Ed.2d 95 (1984).

On appeal, defendants argue for the first time that even if the evidence that Sherrod was unarmed is relevant under Rule 401, it should have been excluded because under Rule 403 “its probative value is substantially outweighed by the danger of unfair prejudice and confusion of the issue.” (Br. at 37). Defendants failed to preserve this argument.

When evidence that Sherrod was unarmed was offered at trial, the defendants did not object on any ground whatsoever. (Tr. 144-147, 277, 438 and 1285). Despite their failure to object at trial, defendants argue that the issue is preserved for appeal by their motion in limine. Then, because their motion in limine cited no authority but simply requested, without reason, that the court prohibit the introduction of evidence concerning the nonexistence of a weapon (R. at 176), defendants rely on the arguments made to the trial court on the first day of trial.4 (Reply Br. at 1-3). During the argument, the trial court emphasized that it believed the evidence was relevant, even critical. The defendants correctly interpreted this ruling as definitive, leaving no room for reconsideration, and thus relieving them of the responsibility of objecting at trial on relevancy in order to preserve the issue for appeal. See American Home Assurance Co. v. Sunshine Supermarket, Inc., 753 F.2d 321, 324-25 (3d Cir.1985).

Relevancy, however, is a separate issue from prejudice and each is governed by a separate rule of evidence. See Fed.R.Evid. 401 and 403. Consequently, the issue of whether the prejudicial effect of the admission of evidence outweighs its probative value is not preserved by an objection to the evidence on the separate ground of relevancy. See, e.g., Laughlin, 772 F.2d at 1392-93. Although the district court’s ruling on relevancy appeared firm, making further discussion at the time or future objections at trial on that issue futile and thus unnecessary to preserve the issue for appeal, the court did not rule on the issue of prejudice. The court’s firmness on one issue does not excuse counsel from interjecting an objection on the ground that the court’s ruling on relevancy would unfairly prejudice his client. See, United States v. Warner, 855 F.2d 372, 374 (7th Cir. Aug. 10, 1988) (counsel has a duty to object, even at the risk of incurring the wrath of the trial court that has stated it does not want to discuss the issue further). But the defendants did not mention prejudice in their written motion in limine, at argument on that motion, or in an objection at trial. This failure to present the argument to the district court waives it on appeal.

*813IV

The majority states that its holding is consistent with today’s opinion in Ford. Maj. op., supra at 805. It is not. In Ford we “emphasize[d] that this court must and will continue to be vigilant in ensuring that when an officer makes the critical judgment to use deadly force, his or her actions will be subject to scrutiny and review.... [We noted that] an officer oftentimes has only a split second to make the critical judgment of whether to use his weapon. But this fact alone will never immunize an otherwise unreasonable use of deadly force.” Ford, 855 F.2d at 1276. Yet in Sherrod, the majority allows this vigilance to lapse in favor of the wisdom of a Sioux Indian prayer. Maj. op., supra, at 806. We can only hope that in the Sherrod family’s third attempt to resolve this tragedy, the jury restores the vigilant review the majority has eviscerated.

As shown in the panel opinion, 827 F.2d 195, and herein, affirmance was warranted.

FLAUM, Circuit Judge,

dissenting.

Federal Rule of Evidence 402 provides that all relevant evidence is initially to be viewed as admissible.1 In my judgment, the evidence presented at trial on the issue of whether the deceased was armed was relevant, but its prejudicial effect should have precluded its admission. Had the evidence been properly objected to, its admission would have warranted reversal of this case. However, because no objection was ever interposed on the specific ground of prejudice, the jury’s verdict should stand. I write separately because my view regarding the admission of this evidence is not advanced by or reflected in the other opinions of the court.

I.

Evidence is relevant if it has “any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence.” Fed.R.Evid. 401. This definition of relevant evidence is an expansive one. To be relevant, evidence need not conclusively decide the ultimate issue in a case; it must simply tend to make a witness’ testimony more or less credible. 1 J. Weinstein & M. Berger, Weinstein’s Evidence ¶ 401[05]. See Davis v. Lane, 814 F.2d 397, 399 (7th Cir.1987) (in a § 1983 claim, evidence that Davis may have been armed with a shank was relevant to the reasonableness of officer’s decision to shoot him); United States v. Miroff, 606 F.2d 777, 781 (7th Cir.1979), cert. denied, 445 U.S. 928, 100 S.Ct. 1315, 63 L.Ed.2d 761 (1980) (evidence of guns found in defendant’s bedroom was admissible as relevant; even though the guns did not demonstrate that defendant was guilty of transporting stolen goods, it was at least a "circumstance” the jury could consider); Saladino v. Winkler, 609 F.2d 1211, 1214 (7th Cir.1979) (in § 1983 excessive use of force case, fact that plaintiff was intoxicated at time of arrest was relevant to the issue of the reasonableness of plaintiff’s conduct).

As complicated as it is to construct a factual picture at the trial court level, it is even more difficult to recreate it on appeal. It is for this reason that great deference must be given to a district judge’s eviden-tiary rulings. Accordingly, on appeal a district court’s decision on the relevance of proffered evidence should not be disturbed absent an abuse of discretion. In this case, I would initially accept the district judge’s finding that the evidence that Sherrod was unarmed (or if it had been otherwise, the evidence that he was armed) was relevant. Applying our highly deferential standard of review in this area, I cannot conclude that a clear showing has been made that the trial court abused its discretion in determining the relevance of this evidence.

*814The contents of Ronald Sherrod’s clothing, and the fact that he did not have a gun in his possession, were relevant because this information could have arguably aided the jury in evaluating the probative value of various witnesses’ conflicting testimony regarding the manner in which Sherrod reached toward his shirt pocket — the action that allegedly precipitated Officer Berry’s decision to shoot him. See 1 J. Weinstein & M. Berger, Weinstein’s Evidence ¶ 401[05] (evidence is relevant if it “will aid the court or jury in evaluating the probative value of other evidence offered to affect the probability of the existence of a consequential fact.”). Furthermore, the fact that Sherrod was unarmed may have tended to make less credible Officer Berry’s testimony about Sherrod’s actions, and might have informed the jury’s consideration of whether a reasonable officer would have believed that he or she was in imminent danger. See supra at 810-811 (Cummings, J., dissenting). While the appropriate test to be applied in this case is the “objective reasonableness” of Officer Berry’s actions, I believe this evidence satisfied the very broad standards of Federal Rule of Evidence 401.2

II.

Having concluded that the evidence in question can be appropriately viewed as relevant, the critical question in my judgment is whether it should have been excluded under Federal Rule of Evidence 403. Under the balancing test of Rule 403, relevant evidence “may be excluded if its probative value is substantially outweighed by the danger of unfair prejudice_” Fed. R.Evid. 403. Evidence is unfairly prejudicial if it creates “an undue tendency to suggest decision on an improper basis, commonly, though not necessarily, an emotional one.” Id. (advisory committee notes). As we have frequently recognized, all relevant evidence is inherently prejudicial. See, e.g., United States v. Thomas, 676 F.2d 239, 244 (7th Cir.1980), cert. denied, 450 U.S. 931, 101 S.Ct. 1392, 67 L.Ed.2d 364 (1981). Relevant evidence should be excluded only if it contains a risk of unfair prejudice that substantially outweighs its probative value. As with the determination that evidence is relevant under Rule 401, a district judge has broad discretion to determine whether the probative value of certain evidence is outweighed by its possible prejudice.

In the present case, as a thorough review of the record clearly establishes, the defendants did not specifically object to the admission of the evidence that Sherrod was unarmed on the basis of Rule 403.3 The district court therefore never explicitly determined whether this evidence was so inherently prejudicial that its probative value was outweighed by the potential prejudice. See supra at 812-13 (Cummings, J., dissenting). Admittedly, the district court’s seemingly unwavering position on the admissibility of this evidence made an objection on Rule 403 grounds uninviting. However, it was incumbent upon counsel to at least proffer a Rule 403 objection. See Sadowski v. Bombardier Ltd., 539 F.2d 615, 623 n. 7 (7th Cir.1976) (quoting Phillips v. Kitt, 290 F.2d 377, 378 (D.C.Cir.1961) (Counsel has a duty “ ‘even at the risk of incurring the displeasure of the trial court, to insist upon his [or her] objection. Having failed to do so, it is too late to urge this as error here.’ ”). Counsel’s lack of diligence resulted in a waiver of this crucial argument, a controlling fact that we should not ignore.

Informing the jury that Sherrod did not have a weapon most probably tainted its decision with an impermissible emotional consideration. Thus, had this objection been preserved for appeal, I would have been persuaded to find the admission of this evidence to be a clear abuse of discretion. I would therefore have reversed the *815verdict on the ground that this evidence was substantially more prejudicial than probative.4

III.

The lamentable events of this case starkly reflect the ofttimes tragic nature of citizen/police contact. In my view, however, this appeal is not the appropriate avenue by which to review the propriety of the police procedures of the city of Joliet, severely lacking as they might be in this particular case. Our purpose should be simply to ensure the essential fairness of the trial.

We can never be certain what the district judge’s ruling would have been if the defendants had objected to the admission of the evidence that Sherrod was unarmed under Rule 403. Thus, we cannot reverse on that basis. Although I strongly believe that this evidence was unduly prejudicial, the defendants’ failure to clearly preserve this specific issue should mandate that we leave the jury’s verdict untouched. I am therefore compelled to respectfully dissent.

McCarthy, J.

Appeal from a judgment of the County Court of Franklin County (Main Jr., J.), rendered June 23, 2013, upon a verdict convicting defendant of the crime of criminal contempt in the first degree.

Amber Montour observed defendant in the front passenger seat of a vehicle driven by the mother of defendant’s infant son (hereinafter the mother). At that time, a stay-aw ay order of protection was in place prohibiting defendant from having any contact with the mother, which had been entered by County Court upon his prior conviction of criminal contempt in the second degree. Montour, who knew the vehicle occupants and was aware of the order of protection, contacted the police. A police officer stopped the vehicle, then driven by the mother’s brother, who indicated that he had dropped off the mother at school, and arrested defendant, a passenger in the vehicle.

At trial, the parties stipulated to the terms of the order of protection and that defendant had a prior criminal contempt conviction. Montour testified that as she drove past the distinct vehicle driven by the mother, with which she was familiar, she saw the occupants from a “few feet” away and made eye contact with defendant. The mother testified, under subpoena, acknowledging that she was aware of the terms of the order of protection but wanted to see defendant that day and still loved him. She explained that she left home for school with her brother in *1423the distinct car and picked up defendant; defendant sat in the front passenger seat, and she had a discussion with him regarding their child. She recounted that defendant got into her vehicle willingly, remained for 10 to 15 minutes and never asked to get out. Upon his conviction of criminal contempt in the first degree, County Court sentenced defendant, as a second felony offender, to 2 to 4 years in prison. Defendant appeals.

Defendant challenges County Court’s trial ruling regarding his postarrest remarks to police. This argument centers on the court’s ruling permitting a police officer to recount the substance of defendant’s postarrest statements over his objection that the prejudicial effect of this evidence outweighed its probative value. Specifically, the officer testified that, after defendant was taken to the police station, he stated to an officer, “You’re a nightmare. You were after me. You’re doing this to me. You’re harassing me, you fucking b*tch.”

While relevant evidence is generally admissible unless it violates an exclusionary rule, “relevant evidence . . . may still be excluded by the trial court in the exercise of its discretion if its probative value is substantially outweighed by the danger that it will unfairly prejudice the other side or mislead the jury” (People v Scarola, 71 NY2d 769, 777 [1988]; see People v Caruso, 6 AD3d 980, 984-985 [2004], lv denied 3 NY3d 704 [2004]). However, evidence is relevant only “if it tends to prove the existence or non-existence of a material fact, i.e., a fact directly at issue in the case” (People v Primo, 96 NY2d 351, 355 [2001]; see People v Mateo, 2 NY3d 383, 424 [2004], cert denied 542 US 946 [2004]), and “[e]vidence of merely slight, remote or conjectural significance will ordinarily be insufficiently probative to outweigh [the] countervailing risks [of prejudice]” (People v Primo, 96 NY2d at 355 [internal quotation marks and citations omitted]).

Here, the People were required to prove that defendant intentionally disobeyed an order of protection that was “duly served” or of which he had “actual knowledge” because he was present in court when the order was issued, that the order required him to stay away from the protected party, and that he had been convicted of criminal contempt in the first or second degree in the prior five years (Penal Law § 215.51 [c]; see Penal Law § 215.50 [3]). The only disputed issues at trial were whether defendant had violated the order of protection and had done so intentionally. Defendant’s remarks to the police officer were directed at the officer’s motives and conduct and reflected defendant’s disdain for her or the police, which were not relevant to or probative of any element of the charged *1424crime or any material fact in issue (compare People v Austin, 115 AD3d 1063, 1065 [2014], lv denied 23 NY3d 960 [2014]). The remarks do not suggest a motive for defendant’s violation of the order of protection and did not constitute an admission by defendant that he had intentionally violated the order of protection. Thus, as defendant’s statements are not relevant to or probative of his motive or any disputed fact or element of the crime charged, they should not have been admitted into evidence. However, given the overwhelming evidence of defendant’s guilt — including his stipulation that the order of protection was in place, Montour’s testimony placing defendant in the vehicle with the mother and the mother’s testimony that they voluntarily remained together in the vehicle for 10 to 15 minutes — there is no “significant probability” that he would have been acquitted had it not been for the admission of his postarrest statements (People v Crimmins, 36 NY2d 230, 241-242 [1975]; see People v Byer, 21 NY3d 887, 889 [2013]).

We are unpersuaded by defendant’s further contention that the prosecutor’s references in summation to Montour’s lack of a motive to testify against defendant were improper and referred to facts not in evidence. While defense counsel’s cross-examination of Montour did not directly address her motives for testifying, it did question her credibility and ability to observe the incident. In his opening statement, defense counsel referred generally to witness “biases” and asserted, among other things, that “[p]eople sometimes make things up because they have reasons to make things up.” Where, as here, defense counsel attacks the credibility and alludes to the motives and biases of prosecution witnesses, the prosecutor is entitled in summation to fairly respond and argue that the witness is not biased and had no motive to lie (see People v Marcus, 112 AD3d 652, 653 [2013], lv denied 22 NY3d 1140 [2014]; People v Barber, 13 AD3d 898, 900 [2004], lv denied 4 NY3d 796 [2005]; People v Alexander, 255 AD2d 708, 710 [1998], lv denied 93 NY2d 897 [1999]; see generally People v Halm, 81 NY2d 819, 821 [1993]). Viewing the prosecutor’s remarks in the context of the trial and summations as a whole, defendant was not deprived of a fair trial. Even if the prosecutor’s remarks were not a fair comment on the evidence or a measured response to defense counsel’s arguments, they “did not rise to the flagrant and pervasive level of misconduct which would deprive defendant of due process” or a fair trial (People v Robinson, 16 AD3d 768, 770 [2005], lv denied 4 NY3d 856 [2005]).

Finally, defendant’s contention that County Court’s charge was inadequate was not preserved for appellate review, as he *1425did not make a request for a particular charge on intent and did not object to the charge as given (see CPL 470.05 [2]; People v Green, 119 AD3d 23, 30 [2014], lv denied 23 NY3d 1062 [2014]). In any event, as County Court closely followed the model jury charge, explaining clearly that intent means “conscious objective or purpose” (see CJI2d[NY] Penal Law § 215.51 [c]), no corrective action is warranted in the interest of justice (see CPL 470.15 [6] [a]).

Peters, P.J., Rose and Clark, JJ., concur.

Ordered that the judgment is affirmed.

Hines, J.

After a jury-waived trial in the Boston Municipal Court, the defendant, Jimmy Warren, was convicted of unlawful possession of a firearm, G. L. c. 269, § 10 (a).2 The complaint *531arose from the discovery of a firearm after an investigatory stop of the defendant in connection with a breaking and entering that had occurred in a nearby home approximately thirty minutes earlier. Prior to trial, the defendant filed a motion to suppress the firearm and statements made after his arrest, arguing that police lacked reasonable suspicion for the stop. The judge who heard the motion denied it, ruling that, at the time of the stop, the police had reasonable suspicion that the defendant was one of the perpetrators of the breaking and entering. The defendant appealed, claiming error in the denial of the motion to suppress.3 The Appeals Court affirmed, Commonwealth v. Warren, 87 Mass. App. Ct. 476, 477 (2015). We allowed the defendant’s application for further appellate review and conclude that because the police lacked reasonable suspicion for the investigatory stop, the denial of the motion to suppress was error. Therefore, we vacate the conviction.

Background. We summarize the facts as found by the judge at the hearing on the motion to suppress, supplemented by evidence in the record that is uncontroverted and that was implicitly credited by the judge. Commonwealth v. Melo, 472 Mass. 278, 286 (2015). On December 18, 2011, Boston police Officer Luis Anjos was patrolling the Roxbury section of Boston in a marked police cruiser when, at 9:20 p.m., he received a radio call alerting him to a breaking and entering in progress on Hutchings Street, where the suspects were fleeing the scene. The dispatcher gave several possible paths of flight from Hutchings Street, one toward Seaver Street and the other toward Jackson Square, locations that are in the opposite direction from one another.4

Anjos went to the scene and spoke to the victims, a teenage male and his foster mother. The male reported that as he was leaving the bathroom in the residence, his foster mother said that she heard people in his bedroom. The victim opened his bedroom door and saw a black male wearing a “red hoodie” (hooded sweatshirt) jump out of the window. When the victim looked out *532the window he saw two other black males, one wearing a “black hoodie,” and the other wearing “dark clothing.” When the victim checked his belongings, he noticed that his backpack, a computer, and five baseball hats were missing. The victim saw the three males run down Hutchings Street, but he could only guess which direction they took thereafter. Anjos peered out the window but could only see twelve to fifteen yards up the street to the intersection of Hutchings and Harold Streets. After speaking to the victims for approximately eight to twelve minutes, Anjos left the scene and broadcast the descriptions of the suspects.

For the next fifteen minutes or so, Anjos drove a four to five block radius around the house, searching for persons fitting the suspects’ descriptions. Because of the cold temperature that night, Anjos did not come across any pedestrians as he searched the area. At around 9:40 p.m., Anjos headed back toward the police station. While on Martin Luther King Boulevard, he saw two black males, both wearing dark clothing, walking by some basketball courts near a park. One male wore a dark-colored “hoodie.” Neither of the two carried a backpack. Anjos did not recognize either of the males, one of whom was the defendant, as a person he had encountered previously in the course of his duties as a police officer.

When Anjos spotted the defendant and his companion, he had a hunch that they might have been involved in the breaking and entering. He based his hunch on the time of night, the proximity to the breaking and entering, and the fit of the males to the “general description” provided by the victim. He decided “to figure out who they were and where they were coming from and possibly do [a field interrogation observation (FIO)].”5 He rolled down the passenger’s side window of the cruiser and “yelled out,” “Hey guys, wait a minute.” The two men made eye contact with Anjos, turned around, and jogged down a path into the park.

After the two men jogged away, Anjos remained in the police cruiser and radioed dispatch that three men6 fitting the descriptions provided by the victim were traveling through the park *533toward Dale Street. Boston police Officers Christopher R. Carr and David Santosuosso, who had heard the original broadcast of the breaking and entering, were very near Dale Street and headed in that direction. Arriving quickly, Carr and Santosuosso observed two males matching Anjos’s description walking out of the park toward Dale Street. Carr parked the cruiser on Dale Street and both officers approached the defendant and his companion as they left the park. The defendant and his companion walked with their hands out of their pockets. Carr saw no bulges in their clothing suggesting the presence of weapons or contraband.

Carr was closer to the two males, approximately fifteen yards away. When he uttered the words, “Hey fellas,” the defendant turned and ran up a hill back into the park. His companion stood still. Carr ordered the defendant to stop running. After the command to stop, Carr observed the defendant clutching the right side of his pants, a motion Carr described as consistent with carrying a gun without a holster.7

Ignoring the command to stop, the defendant continued to run and eventually turned onto Wakullah Street. Carr lost sight of the defendant for a few seconds before catching up with him in the rear yard of a house on Wakullah Street. Carr drew his firearm, pointed it at the defendant, and yelled several verbal commands for the defendant to show his hands and to “get down, get down, get down.” The defendant moved slowly, conduct that Carr interpreted as an intention not to comply with his commands. After a brief struggle, Carr arrested and searched the defendant but found no contraband on his person. Minutes after the arrest, police recovered a Walther .22 caliber firearm inside the front yard fence of the Wakullah Street house. When asked if he had a license to carry a firearm, the defendant replied that he did not.

Discussion. The defendant challenges the judge’s denial of the motion to suppress, claiming error in the judge’s ruling that at the time of the stop on Dale Street, the police had a sufficient factual basis for reasonable suspicion that the defendant had committed the breaking and entering.8 In sum, he argues that the police *534pursued him with the intent of questioning him, while lacking any basis for doing so. Accordingly, he claims that any behavior observed during the pursuit and any contraband found thereafter must be suppressed.

1. Standard of review. “In reviewing a ruling on a motion to suppress evidence, we accept the judge’s subsidiary findings of fact absent clear error and leave to the judge the responsibility of determining the weight and credibility to be given oral testimony presented at the motion hearing” (citation omitted). Commonwealth v. Wilson, 441 Mass. 390, 393 (2004). However, “[w]e review independently the application of constitutional principles to the facts found.” Id. We apply these principles in deciding whether the seizure was justified by reasonable suspicion that the defendant had committed the breaking and entering on Hutchings Street. Commonwealth v. Scott, 440 Mass. 642, 646 (2004).

2. Reasonable suspicion. The judge ruled, and the Commonwealth concedes, that the seizure occurred when Officer Can-ordered the defendant to stop running and pursued him onto Wakullah Street. If a seizure occurs, “we ask whether the stop was based on an officer’s reasonable suspicion that the person was committing, had committed, or was about to commit a crime.” Commonwealth v. Martin, 467 Mass. 291, 303 (2014). “That suspicion must be grounded in ‘specific, articulable facts and reasonable inferences [drawn] therefrom’ rather than on a hunch.” Commonwealth v. DePeiza, 449 Mass. 367, 371 (2007), quoting Scott, 440 Mass. at 646. The essence of the reasonable suspicion inquiry is whether the police have an individualized suspicion that the person seized is the perpetrator of the suspected crime. Commonwealth v. Depina, 456 Mass. 238, 243 (2010) (stop is lawful only if “information on which the dispatch was based had sufficient indicia of reliability, and ... the description of the suspect conveyed by the dispatch had sufficient particularity that it was reasonable for the police to suspect a person matching that description”).

According to the judge’s ruling, the following information established reasonable suspicion for the investigatory stop: the defendant and his companion “matched” the description of two of *535the three individuals being sought by the police; they were stopped in close proximity in location (one mile) and time (approximately twenty-five minutes) to the crime; they were the only persons observed on the street on a cold winter night as police canvassed the area; and they evaded contact with the police, first when both men jogged away into the park, and later when the defendant fled from Carr after being approached on the other side of the park.9

We review the judge’s findings as a whole, bearing in mind that “a combination of factors that are each innocent of themselves may, when taken together, amount to the requisite reasonable belief’ that a person has, is, or will commit a particular crime. Commonwealth v. Feyenord, 445 Mass. 72, 77 (2005), cert. denied, 546 U.S. 1187 (2006), quoting Commonwealth v. Fraser, 410 Mass. 541, 545 (1991). We are not persuaded that the information available to the police at the time of the seizure was sufficiently specific to establish reasonable suspicion that the defendant was connected to the breaking and entering under investigation.

a. The description of the suspects. First, and perhaps most important, because the victim had given a very general description of the perpetrator and his accomplices, the police did not know whom they were looking for that evening, except that the suspects were three black males: two black males wearing the ubiquitous and nondescriptive “dark clothing,” and one black male wearing a “red hoodie.” Lacking any information about facial features, hairstyles, skin tone, height, weight, or other physical characteristics, the victim’s description “contribute[d] nothing to the officers’ ability to distinguish the defendant from any other black male” wearing dark clothes and a “hoodie” in Roxbury. Commonwealth v. Cheek, 413 Mass. 492, 496 (1992) (insufficient detail in generalized description of suspect to justify stop where defendant was observed walking on street approximately one-half mile from scene of reported stabbing, without indication he was fleeing crime scene or had engaged in criminal activity).

With only this vague description, it was simply not possible for the police reasonably and rationally to target the defendant or any *536other black male wearing dark clothing as a suspect in the crime. If anything, the victim’s description tended to exclude the defendant as a suspect: he was one of two men, not three; he was not wearing a red “hoodie”; and neither he nor his companion was carrying a backpack.10 Based solely on this description, Anjos had nothing more than a hunch that the defendant might have been involved in the crime. He acknowledged as much when he explained that the purpose of the stop was “to figure out who they were and where they were coming from and possibly do an FIO.” As noted, an FIO is a consensual encounter between an individual and a police officer. Therefore, the defendant was not a “suspect” subject to the intrusion of a threshold inquiry. Unless the police were able to fortify the bare-bones description of the perpetrators with other facts probative of reasonable suspicion, the defendant was entitled to proceed uninhibited as he walked through the streets of Roxbury that evening.

b. Proximity. We agree with the motion judge that proximity of the stop to the time and location of the crime is a relevant factor in the reasonable suspicion analysis. Commonwealth v. Foster, 48 Mass. App. Ct. 671, 672-673, 676 (2000) (reasonable suspicion established where police observed persons matching physical description on same street and headed in same direction as indicated by informant). Proximity is accorded greater probative value in the reasonable suspicion calculus when the distance is short and the timing is close. See Commonwealth v. Doocey, 56 Mass. App. Ct. 550, 555 n.8 (2002), and cases cited. Here, the defendant was stopped one mile from the scene of the crime approximately twenty-five minutes after the victim’s telephone call to the police. Several considerations, however, weigh against proximity as a factor supporting an individualized suspicion of the defendant as a suspect in the breaking and entering.

The location and timing of the stop were no more than random occurrences and not probative of individualized suspicion where the direction of the perpetrator’s path of flight was mere conjecture. Although the police appropriately began their investigation with the information available to them, this lack of detail made it less likely that a sighting of potential suspects could be elevated beyond the level of a hunch or speculation. As noted by the dissenting Justices in the Appeals Court opinion, given the nearly *537thirty-minute time period between the breaking and entering and the stop on Dale Street, the suspects could have traveled on foot within a two mile radius of the crime scene, a substantial geographic area comprising 12.57 square miles.11 Warren, 87 Mass. App. Ct. at 499 n. 1 (Rubin, J., dissenting). See id. at 488-489 (Agnes, J., dissenting). Other than the victim’s report that the perpetrators fled toward Harold Street, the responding officers had nothing more than the information in the dispatch suggesting that the perpetrators could have fled toward Seaver Street or Walnut Avenue. Depending on the direction taken, these paths of flight would lead to different Boston neighborhoods, Dorchester or Jamaica Plain, in different areas of the city.

In addition, Anjos testified to two important geographical facts that undermine the proximity factor. He acknowledged that Dale Street is in the opposite direction from where either of the reported paths of flight might lead. And, most important, Anjos also stated that if the perpetrators had headed in the direction of Dale Street, they likely would have reached that location well before his first encounter with the defendant and his companion. Thus, where the timing and location of the stop lacked a rational relationship to each other, proximity lacks force as a factor in the reasonable suspicion calculus.

c. Lack of other pedestrians. The judge considered in her analysis that the defendant and his companion were the only people observed on the street as Anjos canvassed the four to five block radius of the Hutchings Street address, traveling “up and down Harold Street, Walnut Avenue and Holworthy Street” before turning onto Martin Luther King Boulevard to return to the station.12 This factor also is of questionable value in the analysis given the lapse of time and the narrow geographical scope of the *538search for suspicious persons. Anjos spoke to the victim for approximately fifteen minutes and thereafter canvassed only four to five blocks surrounding the location of the breaking and entering. The lapse of time between the victim’s report and the canvassing suggests that the perpetrators could have fled the immediate area before Anjos began his search. Thus, the defendant’s presence on the street, some distance away from the crime, within a time frame inconsistent with having recently fled the scene, is hardly revelatory of an individualized suspicion of the defendant as the perpetrator of the crime.

d. Flight. We recognize that the defendant’s evasive conduct during his successive encounters with police is a factor properly considered in the reasonable suspicion analysis. Commonwealth v. Stoute, 422 Mass. 782, 791 (1996) (failure to stop combined with accelerated pace contributed to officer’s reasonable suspicion). But evasive conduct in the absence of any other information tending toward an individualized suspicion that the defendant was involved in the crime is insufficient to support reasonable suspicion. Commonwealth v. Mercado, 422 Mass. 367, 371 (1996) (“Neither evasive behavior, proximity to a crime scene, nor matching a general description is alone sufficient to support . . . reasonable suspicion”). Commonwealth v. Thibeau, 384 Mass. 762, 764 (1981) (quick maneuver to avoid contact with police insufficient to establish reasonable suspicion). “Were the rule otherwise, the police could turn a hunch into a reasonable suspicion by inducing the [flight] justifying the suspicion.” Stoute, supra at 789, quoting Thibeau, supra. Although flight is relevant to the reasonable suspicion analysis in appropriate circumstances, we add two cautionary notes regarding the weight to be given this factor.

First, we perceive a factual irony in the consideration of flight as a factor in the reasonable suspicion calculus. Unless reasonable suspicion for a threshold inquiry already exists, our law guards a person’s freedom to speak or not to speak to a police officer. A person also may choose to walk away, avoiding altogether any contact with police. Commonwealth v. Barros, 435 Mass. 171, 178 (2001) (breaking eye contact and refusing to answer officer’s initial questions did not provide reasonable suspicion for detention or seizure as “[i]t was the defendant’s right to ignore the officer”). Yet, because flight is viewed as inculpatory, we have endorsed it as a factor in the reasonable suspicion analysis. See Commonwealth v. Sykes, 449 Mass. 308, 315 (2007) (defendant’s *539abandonment of bicycle in “effort to dodge further contact with the police was significant” in determining reasonable suspicion); Commonwealth v. Grandison, 433 Mass. 135, 139-140 (2001) (attempt to avoid contact with police may be considered with other factors in establishing reasonable suspicion). Where a suspect is under no obligation to respond to a police officer’s inquiry, we are of the view that flight to avoid that contact should be given little, if any, weight as a factor probahve of reasonable suspicion. Otherwise, our long-standing jurisprudence establishing the boundary between consensual and obligatory police encounters will be seriously undermined. Thus, in the circumstances of this case, the flight from Anjos during the inihal encounter added nothing to the reasonable suspicion calculus.

Second, as set out by one of the dissenting Justices in the Appeals court opinion, where the suspect is a black male stopped by the police on the streets of Boston, the analysis of flight as a factor in the reasonable suspicion calculus cannot be divorced from the findings in a recent Boston Police Department (department) report documenting a pattern of racial profiling of black males in the city of Boston. Warren, 87 Mass. App. Ct. at 495 n.18 (Agnes. J., dissenting), citing Boston Police Commissioner Announces Field Interrogation and Observation (FIO) Study Results, http://bpdnews.eom/news/2014/10/8/boston-police-commis sioner-announces-field-interrogation-and-observation-fio-study-re suits [https://perma.cc/H9RJ-RHNB].13 According to the study, based on FIO data collected by the department,14 black men in the city of Boston were more likely to be targeted for police-civilian encounters such as stops, frisks, searches, observations, and interrogations.15 Black men were also disproportionally targeted for *540repeat police encounters.16 We do not eliminate flight as a factor in the reasonable suspicion analysis whenever a black male is the subject of an investigatory stop. However, in such circumstances, flight is not necessarily probative of a suspect’s state of mind or consciousness of guilt. Rather, the finding that black males in Boston are disproportionately and repeatedly targeted for FIO encounters suggests a reason for flight totally unrelated to consciousness of guilt. Such an individual, when approached by the police, might just as easily be motivated by the desire to avoid the recurring indignity of being racially profiled as by the desire to hide criminal activity. Given this reality for black males in the city of Boston, a judge should, in appropriate cases, consider the report’s findings in weighing flight as a factor in the reasonable suspicion calculus.

Here, we conclude that the police had far too little information to support an individualized suspicion that the defendant had committed the breaking and entering. As noted, the police were handicapped from the start with only a vague description of the perpetrators. Until the point when Carr seized the defendant, the investigation failed to transform the defendant from a random black male in dark clothing traveling the streets of Roxbury on a cold December night into a suspect in the crime of breaking and entering. Viewing the relevant factors in totality, we cannot say that the whole is greater than the sum of its parts.

Conclusion. For the reasons stated above, the police lacked reasonable suspicion for the investigatory stop of the defendant. Therefore, we vacate the judgment of conviction and remand the matter to the Boston Municipal Court for further proceedings consistent with this opinion.

So ordered.

*697OPINION OF THE COURT

Rivera, J.

In this medical malpractice and wrongful death action, we conclude that the trial court erroneously admitted evidence concerning defendant’s negligent treatment of 12 other patients, and that this evidence tainted the jury’s deliberative process. On the facts of this case, the trial court abused its discretion by admitting evidence that was irrevelant to defendant’s liability and that unduly prejudiced the jury. Therefore, the Appellate Division should be reversed and a new trial ordered.

*698I.

Plaintiff Janice Mazella commenced the instant action against defendant William Beals, M.D. and codefendant Elisabeth Mashinic, M.D., claiming that their substandard medical treatment of her husband, Joseph Mazella, proximately caused his suicide. At trial, defendant Beals admitted he deviated from accepted medical practice by prescribing decedent the antidepressant drug Paxil for over a decade while failing to adequately monitor his condition. However, defendant Beals maintained that he was not liable for malpractice because superceding acts severed the causal connection between his conduct and the suicide, including medical care provided by Dr. Mashinic. For her part, plaintiff argued that defendant Beals’ treatment and conduct towards decedent was a contributing factor leading to decedent’s death. A jury found defendant Beals solely liable and he now appeals.

Defendant began treating decedent in October 1993, when he diagnosed him with major depression, obsessive-compulsive disorder (OCD) and generalized anxiety disorder. Defendant prescribed 20 mg of Paxil and eventually discontinued decedent’s antianxiety Klonopin medication, previously prescribed to decedent by his family physician. In April 1994, defendant tapered off decedent’s Paxil dosage and instructed decedent to discontinue it the following month, and to call him if there were any problems.

Decedent next contacted defendant on April 7, 1998, following an episode of depression. Defendant concurred with the recommendation of decedent’s family physician that decedent should be placed on the antianxiety drug Ativan and 40 mg of Paxil. Within a few weeks decedent showed improvement and defendant reduced the Ativan dosage, eventually discontinuing it within the month. Defendant also reduced decedent’s Paxil dosage to 20 mg.

For more than 10 years, defendant refilled the prescriptions for Paxil by telephone or facsimile, without seeing or examining decedent. Then on August 9, 2009, decedent called defendant complaining about anxiety, an increase in obsessive thoughts, and difficulty sleeping. Defendant, who was away on vacation at the time, was unable to see decedent but instructed him to double the Paxil dosage to 40 mg. He also prescribed the antipsychotic medication Zyprexa, for decedent’s anxiety and sleep problems. The following day, on August 10th, *699decedent and plaintiff called defendant. They told defendant that decedent was pale, nauseous, and light-headed, and did not feel well. Defendant instructed decedent to double the Zy-prexa and that he would call him the next day in the late afternoon.

On August 11th, plaintiff observed decedent’s condition worsen and she took him to the emergency room. After decedent was cleared medically he was transferred to the hospital’s Community Psychiatric Emergency Program (CPEP) for overnight observation. According to the hospital records admitted into evidence, decedent complained of suicidal ideations, difficulty sleeping and controlling his thoughts, and feeling as if his body was on fire inside. That night he was taken off Zyprexa and given Ativan. Upon his discharge the following day, decedent was told to discontinue Zyprexa, take Klonopin, and reduce his Paxil dosage to 30 mg.

For the next five days decedent appeared stable. On August 17th, plaintiff and decedent visited defendant, now returned from vacation. This would be the last time defendant had contact with decedent before the suicide.

Both parties presented differing accounts of decedent’s August 17th visit to defendant’s office. According to plaintiff, defendant’s conduct had a devastating adverse impact on decedent’s condition. Plaintiff testified that defendant yelled at them, and that he appeared angry that she had taken her husband to CPEP because defendant viewed this as decedent trying to get help from someone else. She also claimed that defendant degraded decedent, accusing him of not taking more Paxil in the past “because [decedent] couldn’t get an erection.” In response to defendant’s comments, decedent pulled his shirt over his head, even while plaintiff tried to comfort him. According to plaintiff, defendant abruptly ended the session by standing up, waving them off and telling decedent, “Just go to CPEP. That’s where you belong.” Plaintiff further claimed that defendant “threw [them] out of his office. He turned his back,” and never said good-bye. When decedent left he was a “crumbling mess,” and went to CPEP because he believed defendant was refusing to take care of him.

In contrast, defendant testified that during the August 17th visit, decedent was unresponsive and cried, and that when decedent spoke he was very upset because he felt that his wife thought he was acting like a baby. It was also the first time *700decedent could not assure defendant that he would not act upon his suicidal thoughts. Defendant advised decedent that the only option left was inpatient treatment at CPEP. Decedent rejected this advice because he did not want to be seen in a local psychiatric facility, and because decedent felt he could not go a period of time without working. Despite the differences in their respective accounts of the August 17th visit, defendant corroborated plaintiffs testimony that decedent pulled his shirt over his head, adding that decedent had been sobbing, and that he had never seen decedent act this way. He also admitted that he raised his voice, but claimed that he did so to emphasize that he could not be sure outpatient treatment would be adequate to address decedent’s suicidal thoughts. Defendant testified that decedent eventually agreed to go to CPEP, and, as far as defendant knew, decedent remained his patient.

There is no dispute as to what happened after decedent last saw defendant. Decedent went to CPEP later that day and, while he initially declined inpatient care, after he complained of being suicidal he was placed on 15-minute safety checks for the next 27 hours and his access to “lethal means of suicide” was restricted. The following day, August 18th, he complained of feeling hopeless and worthless, and repeated that he would kill himself. His medications were adjusted and he was discharged.

After a difficult and restless night, decedent returned to CPEP on August 19th. He was administered Ativan, and placed on 15-minute safety checks for about 12 hours. That evening he was involuntarily transferred to the psychiatric unit at Auburn Memorial Hospital.

On August 20th, decedent met with Dr. Mashinic. She adjusted his medication and placed him on a multidrug regimen of increased Paxil, Klonopin, Zyprexa, Ativan, and another antipsychotic drug. That night, after Dr. Mashinic discontinued the one-on-one suicide watch, decedent attempted suicide by tying the belt of his hospital gown around his neck. Dr. Mash-inic reinstated the suicide watch, and again changed decedent’s medications, replacing Paxil with another antidepressant, and added Risperdal. Over the course of a week, doctors at Auburn adjusted decedent’s medications as he continued to complain about anxiety and depression, and increased repulsive thoughts of a sexual nature. At times he reported a decrease in depression, but still complained of difficulty sleeping and relaxing.

*701Decedent was discharged on August 27th, and referred to the Brownell Center for outpatient psychiatric care. Brownell had a three-part screening and intake process, which decedent commenced on September 3, 2009, when he met with a social worker. At this time he complained of suicidal and obsessive sexual thoughts. During his second intake visit, on September 9th, he met with a psychotherapist and told her that everything was overwhelming, that he felt “as if someone ha[d] ‘taken his brain out,’ ” and that he had “suicidal thoughts come and go.” The Brownell psychotherapist scheduled an accelerated third intake appointment for September 11th. However, Brownell was unable to obtain decedent’s previous medical records in time for this appointment. As a consequence, on September 11th decedent met instead with an independent licensed social worker and psychotherapist recommended by a family member. Decedent told the psychotherapist that he had suicidal thoughts, but could not act on them because of his daughters. The psychotherapist concluded decedent was not at risk of committing suicide and made plans to check up on him the next day.

As it turned out, decedent did not have any further contact with any medical professionals. Early on September 12, 2009, decedent went to his garage and committed suicide by stabbing himself with a knife. Shortly after, plaintiff found him there, face down in a pool of blood.

II.

In June 2010, plaintiff, as administrator of decedent’s estate, commenced this medical malpractice and wrongful death action against defendant and Dr. Mashinic. She alleged that defendant’s treatment of decedent was negligent, as demonstrated, in part, by his failure to properly prescribe and monitor decedent’s medication, and adequately diagnose decedent’s worsening condition during the August 17, 2009 office visit. She further claimed defendant’s negligence was a direct and proximate cause of his suicide. With respect to Dr. Mashinic, plaintiff alleged that her treatment at Auburn was negligent, and that her conduct was also a direct and proximate cause of decedent’s suicide.

Prior to trial, defendant filed a motion in limine to preclude, among other things, the admittance of a consent agreement between defendant and the Office of Professional Medical *702Conduct (OPMC).1 The OPMC is part of the New York State Board for Professional Medical Conduct and is responsible for investigating complaints against physicians, coordinating disciplinary hearings and enacting sanctions as required. In January 2012, OPMC brought misconduct charges against defendant, alleging that he “deviated from accepted standards of medical care” by prescribing medications to 13 patients over several years without adequately monitoring and evaluating them, and often without any face-to-face visits. Decedent was one of the listed patients. By consent agreement and order dated and finalized in February 2012 (consent order), defendant agreed not to contest charges of negligence based on allegations involving his treatment for 12 of the 13 patients, specifically excluding decedent.2

Defendant argued, in part, that the consent order was not probative evidence of his negligence with respect to decedent, and was unduly prejudicial because none of the uncontested charges involved decedent or addressed the proper treatment for a patient with a long history of depression, anxiety and OCD. Defendant contended that introduction of the consent order would serve only to unfavorably “sway” the jury. The court denied the motion and determined that the consent order “would be admissible in full with regard to the issues surrounding not only the [decedent’s] case, but also [the other patients], based on testimony of habit and credibility.” Prior to trial, defendant conceded that prescribing Paxil to decedent over the course of more than 10 years without any face-to-face contact was a deviation from acceptable medical practice. On the day trial was scheduled to begin, defendant renewed his motion to preclude the consent order, arguing that, in light of his concession, it was no longer probative of any disputed issue. The court again denied the motion.

The consent order was later admitted into evidence during defendant’s testimony. When plaintiff called defendant as a witness, he testified that he failed to appropriately monitor decedent from 2000 through 2009 while decedent was on Paxil, *703but denied that this constituted malpractice. Over defense counsel’s objection, the court admitted the consent order and allowed plaintiff to question defendant about its contents. During that questioning defendant was repeatedly confronted with the fact that OPMC had charged him with “gross negligence” with regard to 13 patients, including decedent, and that defendant signed the consent order in satisfaction of the charges, receiving a reprimand and censure as punishment for his misconduct.

Defendant also sought to preclude admission of a photograph of decedent taken after his suicide, arguing that it lacked probative value because there was no dispute as to the manner of decedent’s death. The court allowed the photograph into evidence “on not only the issue of how [decedent] went about what he did, but also the pain and suffering issues and the other related issues.”

At trial, each party submitted expert testimony to persuade the jury of the party’s respective theory of negligence and causation. Plaintiff relied on Dr. Peter Breggin, a licensed physician in New York with a specialty in psychiatry. Dr. Breg-gin concluded that defendant deviated from accepted medical standards by failing to monitor decedent for years while prescribing Paxil, and by later abandoning him as a patient, and that defendant’s conduct was a significant contributing factor to decedent’s suicide. He explained that following more than 10 years of unmonitored Paxil dosing, defendant worsened decedent’s condition by doubling his Paxil prescription and adding Zyprexa after decedent telephoned him on August 9th. He described this as a “turning point” with catastrophic results for decedent. According to Dr. Breggin, doubling decedent’s Paxil was hazardous because it greatly increased the impact of a very potent drug, and notably is not recommended by the Federal Food and Drug Administration (FDA).

Dr. Breggin also testified to the impact on decedent’s already vulnerable condition when he finally had a face-to-face visit with defendant on August 17, 2009. He explained that when a patient visits a psychiatrist the patient is feeling hurt and self-conscious. According to Dr. Breggin a person who is very distressed, having a great deal of emotional difficulty, is particularly sensitive to humiliation — to being rejected, abandoned and invalidated. A doctor cannot turn a patient away, but instead has to ensure there is adequate follow-up. Dr. Breggin concluded that after August 17th, decedent never *704again established a secure relationship with a physician and had “really been cast at sea by” defendant. He also testified, as established by the photograph and the autopsy report, that decedent’s suicide was “very violent and bloody,” and that such suicides are often associated with the use of antidepressants.3 Therefore, in Dr. Breggin’s opinion, defendant’s actions on August 17th, after years of failing to monitor decedent’s prescription medication and doubling the Paxil dosage over the telephone without an in-person assessment of decedent, were a significant contributing factor to decedent’s suicide.

For his part, defendant presented testimony from Dr. Benson Zoghlin, a family physician, who explained that defendant’s 10 years of prescribing Paxil without seeing decedent did not contribute to the suicide because decedent was doing well during that period. According to Dr. Zoghlin, decedent only appeared to decompensate when he was hospitalized and his medication was substantially readjusted. In his opinion, decedent’s major depressive disorder caused his death, rather than any action taken by defendant.

Dr. Thomas Schwartz, a licensed doctor board certified in psychiatry, also testified on behalf of defendant. He explained that individuals, like decedent, who are suffering from a major depressive disorder and obsessive-compulsive disorder at the same time pose a high risk of suicide. He also opined that the benefits of the different medications that defendant prescribed to decedent outweighed any risks.

The jury returned a verdict for plaintiff, finding both defendant and Dr. Mashinic negligent, but that only defendant’s negligence proximately caused decedent’s suicide. The jury awarded $1,200,000 in damages and apportioned $800,000 to plaintiff and $400,000 to be divided among decedent’s three surviving daughters. The court denied defendant’s motion to set aside the verdict, and entered an amended judgment for plaintiff in accordance with the money damages awarded by the jury.4

*705The Appellate Division affirmed, with one Justice dissenting (122 AD3d 1358 [4th Dept 2014]). We granted defendant leave to appeal (25 NY3d 901 [2015]), and now reverse.

III.

Defendant asserts several grounds for reversal.5 First, he claims the verdict is legally insufficient because plaintiff failed to establish defendant was the proximate cause of the suicide. Second, defendant argues that he was denied a fair trial by the trial court’s admission into evidence of the consent order and the photograph of decedent’s body. Third, defendant argues that the trial court abused its discretion when it denied his request for a special verdict sheet on liability and damages. We conclude that although the evidence was sufficient to support the verdict, the trial court committed reversible error when it admitted the consent order and permitted defendant to be questioned regarding its contents.

A. Legal Sufficiency of the Evidence

To succeed on his legal insufficiency claim, defendant must establish “there is simply no valid line of reasoning and permissible inferences which could possibly lead rational [persons] to the conclusion reached by the jury on the basis of the evidence presented at trial” (Cohen v Hallmark Cards, 45 NY2d 493, 499 [1978]). This is a “basic assessment of the jury verdict” and prohibits a holding of insufficiency “in any case in which it can be said that the evidence is such that it would not be utterly irrational for a jury to reach the result it has determined upon” (id.).

In a medical malpractice action, the plaintiff must show that the defendant “deviated from acceptable medical practice, and that such deviation was a proximate cause of the plaintiff’s injury” (James v Wormuth, 21 NY3d 540, 545 [2013]). Defendant conceded that he deviated from accepted medical standards by failing to properly monitor decedent, and on appeal he *706argues only that the evidence does not support a jury determination that his negligence was a proximate cause of the suicide.6

A defendant’s negligence qualifies as a proximate cause where it is “a substantial cause of the events which produced the injury” (Derdiarian v Felix Contr. Corp., 51 NY2d 308, 315 [1980]). However, “[w]here the acts of a third person intervene between the defendant’s conduct and the plaintiff’s injury, the causal connection is not automatically severed” (id.). As this Court has explained, “liability turns upon whether the intervening act is a normal or foreseeable consequence of the situation created by the defendant’s negligence” (id., citing Parvi v City of Kingston, 41 NY2d 553, 560 [1977]). Only where “the intervening act is extraordinary under the circumstances, not foreseeable in the normal course of events, or independent of or far removed from the defendant’s conduct,” may it possibly “break[ ] the causal nexus” (id.). The mere fact that other persons share some responsibility for plaintiff’s harm does not absolve defendant from liability because “there may be more than one proximate cause of an injury” (Argentina v Emery World Wide Delivery Corp., 93 NY2d 554, 560 n 2 [1999]; see also NY PJI 2:71).

Defendant contends that decedent’s hospitalization at Auburn and treatment by other medical professionals after defendant last saw him on August 17, 2009, were intervening and superceding events that broke any casual connection between defendant’s conduct and decedent’s suicide. He further argues that the suicide is too far removed from defendant’s treatment of decedent to be considered proximate. We hold his claims to be without merit.

Although several events transpired after his last meeting with decedent on August 17th, there was sufficient trial evidence for the jury to conclude that, regardless of these events, defendant proximately caused decedent’s suicide. Defendant admitted to negligently treating decedent for over a decade, which was further corroborated by evidence of the specific manner in which he prescribed Paxil for over 10 years without properly monitoring or meeting with decedent. There *707was also trial evidence supporting plaintiff’s argument that the violent nature of the suicide indicated it was connected to decedent’s prescription drug use. Furthermore, the jury could have credited plaintiff’s version of the August 17th office visit and concluded that defendant’s conduct worsened decedent’s condition, leading to his suicide.

Significantly, plaintiff’s expert provided testimony to assist the jury in connecting defendant’s negligence with decedent’s suicide, lending further record support for the verdict. Dr. Breggin testified that, in his considered expert opinion, what led to decedent’s suicide was a multistage process.

“I think it begins with ten years unmonitored on Paxil, so that he’s inevitably going to have a horrific withdrawal reaction when he’s abruptly stopped. In other words, even though it’s covered over with other drugs, the brain just can’t bounce back after ten years. . . . The lack of monitoring, very likely, contributed. By his not having anyone that he had a relationship with to go to, by his not having someone to observe whether he was, in some way, getting worse on the drug because the family isn’t going to necessarily notice, and it happens over time. . .. Then the August 9th, 2009, prescription of the Zyprexa . . . and the doubling of the dose of the Paxil on the phone, sight unseen, with no records, was the real beginning of the catastrophe, because at that moment, he seemed as though he was having a problem like he had had twice before .... But, now, we have this new complication that he’s doubling in the dose, which is, the FDA recommend [s], no more than 10 milligrams at a time. He’s given 20 additional milligrams. And adding Zyprexa, I think that’s a real turning point for him, even though he’s only on the Zyprexa for a few days.”

He described how defendant’s actions on August 17th tragically impacted decedent at a moment when he was most vulnerable.

“It’s the start of the big decline, and I think that’s very important. I think his visit later in August with the patient . . . where the patient feels humiliated, invalidated, rejected, abandoned, remarkable words to find in a medical record, rejec*708tion, abandonment and invalidated, marked and unseen in a medical record and by a doctor. This was a very, very big impact, and it left him with no relationship. The main preventative of suicide that we know of is a good relationship with a therapeutically-oriented professional. He was bereft of that. He has nobody he’s going to trust, and it’s going to make it hard to trust after that.”

It was then for the jury to decide the persuasiveness of this testimony, and to consider it along with the opinions of defendant’s opposing experts (see People v Drake, 7 NY3d 28, 33 [2006] [“jurors remain always free to accept or reject expert evidence”]; People v Negron, 91 NY2d 788, 792 [1998] [“a jury is entitled to assess the credibility of witnesses and determine, for itself, what portion of their testimony to accept and the weight such testimony should be given”]).

To the extent defendant claims there is legally insufficient evidence of a causal nexus because the third-party acts were unforeseeable, we disagree. There is no superceding event if “the intervening act is a normal or foreseeable consequence of the situation created by the defendant’s negligence” (Derdiarian, 51 NY2d at 315). Defendant concedes that when he last saw decedent alive on August 17th, decedent’s conduct was unusual. Decedent was anxious and very upset and, for the first time, he was unable to assure defendant that he could control his suicidal thoughts. According to plaintiff, defendant threw decedent out of his office when he was desperately in need of help, leaving decedent a “crumbling mess.” The jury could have fully credited plaintiff’s version of these events and Dr. Breggin’s opinion about decedent’s condition when he left defendant’s office. Thus, the jury could have concluded that it was foreseeable that decedent would seek treatment by others and that the treatment could potentially be lacking. Under these circumstances, we cannot say that the intervening acts are “of such an extraordinary nature or so attenuate [ ] defendant’s negligence from the ultimate injury that responsibility for the injury may not be reasonably attributed to the defendant” (Kush v City of Buffalo, 59 NY2d 26, 33 [1983]).

This case required the jury to consider decedent’s mental health treatment and the delicate and complex functioning of the brain and human emotion under prescription drug use. The jury was presented with evidence of the long-term impact of defendant’s negligence on decedent’s condition, as well as *709evidence that subsequent medical treatment could be a foreseeable consequence of defendant’s actions. Since a valid line of reasoning and permissible inferences could lead rational persons to find defendant liable for medical malpractice based on this evidence, we conclude defendant’s legal insufficiency claim is without merit (see Cohen, 45 NY2d at 499).

B. Evidentiary Rulings

Defendant also claims that, even if there was evidence sufficient to support the verdict, certain evidentiary rulings by the trial court denied him a fair trial. He argues that the court erroneously admitted evidence of the photograph of decedent’s body and the consent order. Although the court did not abuse its discretion in admitting the photograph, we agree with defendant that admission of the consent order was an abuse of discretion warranting reversal and a new trial.

Since “[t]rial courts are accorded wide discretion in making evidentiary rulings . . . , absent an abuse of discretion, those rulings should not be disturbed on appeal” (People v Carroll, 95 NY2d 375, 385 [2000]). To be admissible, evidence must be relevant and its probative value outweigh the risk of any undue prejudice (People v Morris, 21 NY3d 588, 595 [2013]).

Defendant’s claim that the photograph should have been precluded because it lacked probative value and served only to arouse the jury’s emotions is without merit. The photograph depicted the manner in which decedent committed suicide and was relevant to plaintiff’s theory that the violent nature of the suicide — death by self-inflicted knife wounds — was a result of decedent’s extreme mental and emotional condition, induced by the long-term use of prescription drugs. Nor was its admission unduly prejudicial since there was already testimony from a paramedic describing the condition in which he found the body, and the official autopsy report from the Medical Examiner’s Office was admitted into evidence without objection. Therefore, the court did not abuse its discretion in admitting the photograph.

The trial court’s admission of the consent order into evidence is a wholly different matter. Generally, “it is improper to prove that a person did an act on a particular occasion by showing that he did a similar act on a different, unrelated occasion” (Matter of Brandon, 55 NY2d 206, 210-211 [1982], citing Prince, Richardson on Evidence §§ 170, 184 [10th ed]; see also Coopersmith v Gold, 89 NY2d 957, 959 [1997]). Contrary *710to plaintiff’s argument, none of the exceptions to this rule— motive, intent, the absence of mistake or accident, a common scheme or plan, or identity — apply in this case (see Matter of Brandon, 55 NY2d at 211). Moreover, even though the consent order was a public document, and under Public Health Law § 10 (2) possibly admissible as “presumptive evidence of the facts . . . stated therein” if otherwise properly rendered admissible evidence, under these facts it should not have been admitted.

The record establishes that the consent order was probative of neither defendant’s negligence nor the question of proximate cause. As part of the consent order defendant agreed not to contest negligent treatment of certain anonymous patients, none of whom was the decedent. As such, defendant preserved his objections to factual allegations related to decedent and any charges of misconduct based on those allegations. Since the consent order did not establish facts concerning defendant’s treatment of decedent, it was not probative as to that issue. In any event, given defendant’s pretrial concession that he deviated from accepted medical practice, the issue of negligent treatment did not require resolution by the jury.

Further, any possible relevance of the consent order’s contents was outweighed by the obvious undue prejudice of his repeated violations of accepted medical standards (see Maraziti v Weber, 185 Misc 2d 624, 626 [Sup Ct, Dutchess County 2000] [court denied admittance of an OPMC report detailing previous instances of defendant’s negligence since such evidence was “of marginal relevance at best, but would be likely to unduly prejudice the jury”]). The consent order was nothing more than evidence of unrelated bad acts, the type of propensity evidence that lacks probative value concerning any material factual issue, and has the potential to induce the jury to decide the case based on evidence of defendant’s character (see People v Arafet, 13 NY3d 460, 464-465 [2009] [“Evidence of uncharged crimes is inadmissible where its only purpose is to show bad character or propensity towards crime”]; Hosmer v Distler, 150 AD2d 974, 975 [3d Dept 1989] [trial court properly excluded defendant’s prior convictions for driving while intoxicated and that he had a habit of excessive drinking as unfairly prejudicial propensity evidence]).

Plaintiff’s additional argument that the consent order was admissible to impeach defendant as an admission and as a prior inconsistent statement is also without merit. Plaintiff *711claims she was entitled to present this evidence to the jury once defendant testified that although he was negligent in prescribing Paxil for 10 years without monitoring decedent, his conduct was not malpractice. As a preliminary matter, defendant’s testimony was not inconsistent because the consent order did not include any assertions or concessions regarding defendant’s treatment of decedent. In addition, since medical malpractice requires a finding of causation (James, 21 NY3d at 545), defendant could concede negligent treatment and still maintain his conduct did not constitute malpractice as a legal matter.

We are also unpersuaded by plaintiff’s claim that the evidence was admissible to impeach defendant’s credibility. Collateral matters relevant only to credibility are properly excluded because they distract the jury from the central issues in the case, and bear the risk of prejudicing the jury based on character and reputation (see Badr v Hogan, 75 NY2d 629, 635 [1990]; People v Schwartzman, 24 NY2d 241, 245 [1969]; see also Prince, Richardson on Evidence § 4-410 [11th ed]). It is an abuse of a trial court’s discretion to admit evidence of bad acts when such evidence lacks any probative value, or bears only marginal relevance, outweighed by its prejudicial effect (see Badr, 75 NY2d at 635; Prince, Richardson on Evidence §§ 4-410, 4-501 [11th ed]). Here, given defendant’s concession at trial that he deviated from accepted medical practices, the consent order was unquestionably collateral, without probative value, and, regardless, improperly prejudicial.

On the facts of this case, there were no permissible grounds to allow the consent order into evidence. Morever, notwithstanding that under CPLR 2002 “[a]n error in a ruling of the court shall be disregarded if a substantial right of a party is not prejudiced,” here, admission of the consent order tainted the deliberative process, and sufficiently prejudiced defendant, such that we cannot disregard this error. Given the multiple allegations of defendant’s negligent monitoring of prescription drug treatment, and the numerous patients referenced in the consent order, we cannot say that the verdict was not influenced by this powerful evidence of defendant’s professional misconduct. Indeed, it is difficult to imagine how a jury could simply ignore that defendant negligently treated 12 other patients for years in a similar manner as decedent, namely failing to monitor them, and that this conduct resulted in OPMC charges leading to its oversight of his medical practice.

*712This point was not lost on plaintiff, who repeatedly referred to defendant’s acts of negligence and, during summation, explicitly relied on the consent order to link prior allegations of defendant’s negligence with plaintiff’s current claims. In light of the prejudicial nature of the consent order and its repeated use at trial we cannot say that its admission did not have a substantial impact on the verdict.

Here the evidence portrayed defendant as a serial pill pusher, oblivious to the health and safety of those in his care, and a danger to patients. Since the evidence could have induced the jury to punish him for his unrelated misdeeds, admission into evidence of the consent order was sufficiently prejudicial to defendant so as to require a new trial (see Badr, 75 NY2d at 637 [cross-examination of a witness with prior bad acts was “sufficiently prejudicial” to require a new trial]; compare with Geary v Church of St. Thomas Aquinas, 98 AD3d 646, 647 [2d Dept 2012], lv denied 20 NY3d 860 [2013] [no new trial required when the court improperly precluded evidence since “there is no indication that the evidence would have had a substantial influence on the result of the trial”]).

C. The General Verdict Sheet

Given our determination that defendant is entitled to a new trial, we briefly address his claims that use of a general verdict sheet was improper. Defendant argued that the court should have provided the jury with a special verdict sheet with individual interrogatories because plaintiff relied on three different theories of liability. However, there was a single theory of liability presented to the jury based on the defendant’s continuum of negligent treatment.

Although defendant did not propose to the trial court the specific type of special verdict he now advocates, such a special verdict sheet itemizing the subcategories of damages may assist a court’s review of the jury’s monetary award (CPLR 4111 [d]; see Killon v Parrotta, 125 AD3d 1220, 1223 [3d Dept 2015]). On retrial, defendant should be afforded the opportunity to argue in support of a special verdict sheet on damages.

Accordingly, the order should be reversed, with costs, and a new trial ordered.

Chief Judge DiFiore and Judges Pigott, Abdus-Salaam, Stein and Garcia concur; Judge Fahey taking no part.

Order reversed, with costs, and a new trial ordered.