United States v. Lance Dillard

In the United States Court of Appeals For the Seventh Circuit ____________________ No. 15‐3343 UNITED STATES OF AMERICA, Plaintiff‐Appellee, v. LANCE DILLARD, also known as DOUBLE, Defendant‐Appellant. ____________________ Appeal from the United States District Court for the Northern District of Illinois, Eastern Division. No. 1:13‐cr‐00288‐1 — Gary Feinerman, Judge. ____________________ ARGUED APRIL 7, 2017 — DECIDED MARCH 12, 2018 ____________________ Before POSNER, RIPPLE, and SYKES, Circuit Judges. RIPPLE, Circuit Judge. With the assistance of a cooperating informant, law enforcement surveilled multiple heroin sales involving Lance Dillard and Gregory Chester. On the basis of the evidence obtained in that surveillance, Mr. Dillard and  Circuit Judge Posner retired on September 2, 2017, and did not partici‐ pate in the decision of this case, which is being resolved by a quorum of the panel under 28 U.S.C. § 46(d). 2  No. 15‐3343 Chester were arrested and charged in a three‐count indict‐ ment: two counts of distribution of heroin, in violation of 21 U.S.C. § 841(b)(1)(C), and one count of conspiracy to distrib‐ ute heroin, in violation of 21 U.S.C. §§ 846 and 841(a)(1). Both 1 men were convicted on all applicable counts. Mr. Dillard then moved for judgment of acquittal and for a new trial. The district court denied the motions and sentenced him to 10 years’ imprisonment on each of the first two counts and 70 months on the third count, all to run concurrently. Mr. Dillard now appeals. He concedes that the district court generally expressed unwillingness to allow testimony that related to his alleged membership in the Hobos, a partic‐ ularly notorious gang. Nevertheless, he argues, the court al‐ lowed numerous law enforcement officials to describe their positions in terms that strongly suggested that Mr. Dillard was a member of a gang and allowed one reference to the Ho‐ bos in cross‐examination of a lay witness. He also contends that, after a single juror was exposed to inflammatory press coverage, the court’s decision to dismiss only that single juror was not sufficient remedial action. We conclude that the district court did not err in its evi‐ dentiary rulings and that the jury was not exposed to signifi‐ cant prejudicial testimony. Furthermore, the district court took adequate steps to protect against any further potential juror bias. Accordingly, we affirm the judgment of the district court. 1  Chester  was  charged  only  in  Counts  One  and  Three  and  convicted  of both. Mr. Dillard was charged and convicted on all three counts. No. 15‐3343  3 I BACKGROUND A. In mid‐2011, Mr. Dillard and Chester were engaged in a conspiracy to distribute heroin. On three occasions during the course of the conspiracy, the pair sold heroin to a man named Keith Daniels. Daniels was, at the time of the sales, a confi‐ dential informant to law enforcement. On April 8, 2013, Mr. Dillard and Chester were the subjects of separate federal criminal complaints charging distribution of heroin. In the affidavit attached to each complaint, an FBI agent detailed several recorded or monitored transactions with Daniels, although the affidavit referred to him only as 2 the “cooperating source.” On April 9 and 10, 2013, law en‐ forcement arrested Mr. Dillard and Chester, respectively. On April 14, 2013, while the defendants were both in federal cus‐ tody on the drug charges, Daniels was murdered outside of his home. A grand jury later charged Mr. Dillard and Chester in a three‐count indictment alleging a conspiracy to distribute heroin as well as two counts relating to specific sales to Dan‐ iels, which occurred on June 9, 2011, and September 1, 2011. The Government introduced substantial evidence in the course of several days of trial testimony. The evidence in‐ cluded: audio recordings of calls and meetings between Dan‐ iels and the defendants to arrange heroin sales; voice identifi‐ cation of the defendants on the calls by law enforcement of‐ ficers; a separate audio recording of Chester in prison in 2 R.1 at 3. 4  No. 15‐3343 3 which he stated that someone had “wor[n] a wire on” him; testimony by officers who surveilled the transactions; still photos from videos of certain transactions which showed Mr. Dillard’s distinctive forearm tattoo; testimony by Mr. Dillard’s girlfriend identifying his nickname as “Double,” which could be heard on the recordings, as well as identifying his tattoo, stating that he had multiple cell phones, and fur‐ ther stating that he had rented and purchased cars in her name, including cars used in two meetings relevant to the charged offenses; cell phone records showing calls between the defendants at times relevant to the conspiracy, specifi‐ cally, before and after transactions; cell phone location data showing their movements on September 1, 2011, which matched the locations they described in substance during rec‐ orded calls on that day; and evidence concerning Chester’s gambling of several thousands of dollars in the days follow‐ ing the sales to Daniels, including on a trip to Las Vegas with Mr. Dillard. In accordance with pretrial rulings by the district court, the jury was told only that Daniels was deceased and therefore unavailable as a witness; his murder was not men‐ tioned. B. Prior to trial, Chester filed a motion in limine to prohibit the Government from introducing testimony regarding the defendants’ membership in, or association with, a gang, in‐ cluding the Hobos gang. The district court stated that its ten‐ 3 R.283 at 237 (Trial Tr. 729). No. 15‐3343  5 dency was “to err, if at all, on the side of not permitting” evi‐ 4 dence of gang affiliation. Specifically, the court noted that in applying the balancing test under Federal Rule of Evidence 403, it “almost always … ha[s] found the danger of unfair prej‐ udice, when it is there, outweighs what is most frequently a 5 kind of modest probative value.” The Government asserted that both defendants were members of the Hobos street gang and that it was through this gang membership that Daniels knew them and was able to contact them regarding heroin sales. Nevertheless, the court determined that it did not have sufficient information based on the parties’ pretrial submis‐ sions about how any testimony regarding gang membership would be introduced and for what purpose. The district court denied the motion and determined that it would rule on any gang‐related evidence through the course of trial. Scattered references to gangs occurred throughout trial, sometimes over Mr. Dillard’s or Chester’s objections and sometimes not. The lion’s share of references came from law enforcement officers describing their assignments or their units.6 After one of these exchanges on the third day of trial, 4 R.287 at 16. 5 Id. 6 Examples of this testimony include one FBI agent noting that he “typi‐ cally [does] gang, drug investigations,” R.280 at 6; that he uses a particular technique in investigations “centered on drugs and gangs,” id. at 67; and that he met with one of the defendants at the Chicago Police Department’s “gang intelligence unit,” id. at 72. A Chicago Police officer likewise noted that he worked for “the Organized Crime Division, gang investigations,” R.281 at 159 (Trial Tr. 249); that “as a gang investigator … I investigated 6  No. 15‐3343 Mr. Dillard’s attorney asked for a sidebar and objected that a particular law enforcement witness had used the term “gang” four or five times in just a few sentences of testimony. The court responded that it understood the objection and was “al‐ lergic” to “references to gangs” “unless there is something 7 that really hooks it up.” The prosecutor then stated that he understood the concern, did not anticipate that particular re‐ 8 sponse by the witness, and would like to “just move on.” The court indicated that it did not want to draw attention to the gang references with an instruction to ignore, and Mr. Dillard’s attorney agreed. The court added that counsel should “be aware and … caution whatever other witnesses 9 they are going to have to stay away from the terminology.” Again, Mr. Dillard’s attorney stated that this approach was 10 “fine.” On the fifth day of trial, a lay witness, Dennis Myers, was called by Mr. Dillard’s attorney. Myers testified that one of the voices on a particular recorded call with Daniels was not criminal gangs, investigated gang shootings, gang homicides, gang‐re‐ lated incidents dealing with narcotics as well as weapons,” id.; and, in a prior position, he had been responsible for “gang activity and the gang crime” that happened in a particular area, id. at 163 (Trial Tr. 253). Another Chicago Police officer described being in the “gang investigations unit” as a “gang task force officer” and being “specially trained … for the gangs, gang enterprise, the cities, the City gang crime areas.” R.282 at 37–38. 7 R.282 at 38–39 (Trial Tr. 338–39). 8 Id. at 39 (Trial Tr. 339). 9 Id. 10 Id. No. 15‐3343  7 one of the defendants, but was his own son, Patrick Davis. He also testified that the phone number used on that call was for Davis’s phone, and that he had used that number in the past. Finally, he noted that Davis and Daniels were first cousins. On cross, the Government asked Myers if he recognized any other voices in the conversation and if he knew his son’s friends. Myers said that he did because “[t]hey all live to‐ 11 gether.” The Government then asked if he knew Chester. At that point, Chester’s counsel objected that it was be‐ yond the scope of direct. In a sidebar, the court disagreed, and the Government further clarified that, given the conspiracy charge, it was possible they were working with “other people 12 like Patrick Davis” to accomplish the sales. Indeed, the Gov‐ ernment continued, out of the earshot of the jury: [J]ust so I can be clear, Patrick Davis is a Hobo. He is in the gang. … Gregory Chester is a Hobo. That is how all of these guys know each affilia‐ tion. … [T]his witness is stepping right into [sic], and defense counsel is opening that door. And it is very troubling to us after all the work—we have stayed out of the gang side of the thing… . I cannot cross‐examine this person on how these people all know each other, his bias against the Government, his bias against— in favor of certain individuals, and trying to protect them because of their gang affiliations, 11 R.284 at 44 (Trial Tr. 791). 12 Id. at 46 (Trial Tr. 793). 8  No. 15‐3343 your Honor. I have to go there.13 Chester’s attorney objected that it was prejudicial to Chester and that Myers was not his witness but Mr. Dillard’s. The court responded that it “didn’t anticipate that we were going to be getting into this area at all,” but that “[t]he problem gets 14 manufactured by calling this witness.” The court noted that 15 the Government was “always entitled to bring up bias.” Mr. Dillard’s attorney seemed to accept the conclusion gener‐ ally as “fine,” but objected that “this is … not bias” and caused 16 “extreme prejudice” to his client. THE COURT: You know, when it comes to 403 balancing, I am as sensitive to that as any‐ body else. But when it has been raised by reason of calling a witness and there is this kind of con‐ nection—or arguable connection in terms of bias, I can’t somehow say, well, okay, we will take a piece of this fella’s evidence into mind without considering that component. If it is raised by calling the witness, it has been raised. And I didn’t do it. You did. [Mr. Dillard’s counsel]: I understand, your Honor. This is going to become a RICO trial if he knows they are Hobos. Is the ruling—I just want to be clear on how far this is going to go. 13 Id. at 46–47 (Trial Tr. 793–94). 14 Id. at 47 (Trial Tr. 794). 15 Id. at 48 (Trial Tr. 795). 16 Id. No. 15‐3343  9 THE COURT: Is what? [Mr. Dillard’s counsel]: Is the ruling—what I am concerned, is this going to basically become the RICO case? It is going to be limited to, as I understand, which I still object to—but as I un‐ derstand but that whether he know these guys are Hobos— THE COURT: He is not going to enlarge the scope of this case. We are dealing with this con‐ spiracy. But what you have done, as I under‐ stand, by calling this fella and getting him to make this identification of his son, is to put into the case an issue that I had not anticipated was going to be in the case. That is—I cannot stuff that omelet back into the eggshell. You did it. And what the consequences are, I regret to say, they are going to be. I didn’t make it. You did.17 Following the exchange, the Government attorney continued to ask about Davis’s acquaintances and about the defendants. He asked how they knew each other, and Myers again stated 18 that they all “came from the same project.” The prosecutor 19 then asked, “Do you know what they call themselves?” My‐ ers said no, and then the prosecutor asked, “Have you ever heard the word ‘Hobo’?,” to which Myers responded that he 17 Id. at 48–49 (Trial Tr. 795–96). 18 Id. at 51 (Trial Tr. 798). 19 Id. 10  No. 15‐3343 20 “heard of it in the paper.” Although the prosecutor contin‐ ued to ask about the relationships between the defendants and others, no other references to gangs or the Hobos specifi‐ cally were made. C. On the first day of trial, the Chicago Sun‐Times had pub‐ lished both in print and online an article referencing the trial, a larger RICO trial against Hobo members, and the murder of Daniels following the defendants’ arrest. At the start of the second day of trial, the prosecutor arrived and alerted the court and the attorneys for the defendants to the article’s pub‐ lication the day before. The court indicated that it would in‐ quire of the jurors whether anyone had seen the article and agreed with the prosecutor that it would be “appropriate … to continue to admonish the jury on a daily basis to avoid the 21 media.” The court apparently alerted the court librarian to concerns about media coverage and asked for updates. On the fourth day of trial, after receiving an update, the court spoke with the attorneys about the fact that the Sun‐Times piece had been picked up and published online by CBS. To guard against the possibility of influence, the attorneys agreed that the court would inquire of the jury whether anyone uses the internet as a news source and then follow up any positive an‐ swers with individual inquiries with jurors about what web‐ sites they read. The court also said that it was not inclined to repeat its initial admonition about news to the jury because it 20 Id. 21 R.281 at 4 (Trial Tr. 94). No. 15‐3343  11 believed the jury was capable of following the instruction. The court accordingly questioned the jurors and none of their answers indicated that they had used the CBS site or had seen media coverage of the trial. The following day, a single juror, David Moskowitz, con‐ tacted the court and asked to speak with the judge. The judge then called the juror during the instruction conference with the attorneys present. Juror Moskowitz informed the judge that his wife had seen the Sun‐Times article and out of con‐ cern for his safety had brought it to his attention. The juror did not believe he could set it aside and wanted to alert the court. The court excused him from further service and in‐ quired of the juror, at the request of Mr. Dillard’s attorney, whether he had spoken to anyone about the article. Ju‐ ror Moskowitz indicated that he had not. No party suggested any further questioning of any of the jurors. D. Mr. Dillard moved for a new trial and judgment of acquit‐ tal at the close of the evidence. The court denied his motion, and the jury returned a verdict of guilty on all counts. Mr. Dillard was sentenced to the 10‐year mandatory mini‐ mum on Counts One and Two, and 70 months on Count Three, each to run concurrently. II DISCUSSION Mr. Dillard now claims that the court erred in its handling of the gang references and of the prejudicial media coverage 12  No. 15‐3343 and that the prosecution compounded these errors. The cu‐ mulative effect, he contends, is the denial of his right to a fair trial before an impartial jury. A. We first examine whether the court erred in allowing ref‐ erences to the term “gang” or “Hobos” at trial. We review the district court’s rulings on evidentiary issues for an abuse of discretion. See United States v. Ozuna, 674 F.3d 677, 681 (7th Cir. 2012). A district court considering such evidence applies Federal Rule of Evidence 403, determining whether the “pro‐ bative value is substantially outweighed by the danger of … unfair prejudice” to the defendant. Id. at 682. Our cases have “long recognized the substantial risk of un‐ fair prejudice attached to gang affiliation evidence, noting such evidence is likely to be damaging to a defendant in the eyes of the jury and that gangs suffer from poor public rela‐ tions.” United States v. Irvin, 87 F.3d 860, 864 (7th Cir. 1996) (internal quotation marks omitted). “We therefore require careful consideration by district courts in determining the ad‐ missibility of gang membership and gang activity evidence.” Id.; see also United States v. Montgomery, 390 F.3d 1013, 1018 (7th Cir. 2004) (“[I]n our review we examine the care and thor‐ oughness with which a district judge considered the admis‐ sion or exclusion of gang‐involvement evidence.”). Mr. Dillard claims that our decision in Irvin, 87 F.3d 860, counsels that reversal is appropriate here. In Irvin, two de‐ fendants were charged with and convicted of possession with intent to distribute methamphetamine and of a firearm of‐ fense. At trial, over the defendants’ objections, the jury heard No. 15‐3343  13 various pieces of evidence linking the defendants to the Dia‐ blos, a motorcycle gang. The district court had instructed the Government to use the term “club,” rather than “gang”; the Government “then asked several questions mocking the use of the term motorcycle ‘club,’ before resuming the use of the word ‘gang.’” Id. at 863. On appeal, we determined that the majority of references should have been excluded under the familiar balancing test of Federal Rule of Evidence 403. In Irvin, we acknowledged that “[w]e have consistently held that, under appropriate circumstances, gang evidence has probative value warranting its admission over claims of prejudice.” Id. at 864. We determined, however, that in Irvin, the probative value of the evidence was minimal. The Gov‐ ernment had argued that the evidence established a “joint venture,” essentially “that the defendants’ common member‐ ship in the motorcycle gang made it more likely that they were involved in the drug transaction together, and thus that they were both in constructive possession of the drugs and both intended to distribute the drugs.” Id. The trial court, ref‐ erencing a “conspiracy,” had accepted that reasoning. Id. We noted, however, that there was no conspiracy charge at issue; the defendants had been charged individually with posses‐ sion with intent to distribute. Therefore, any supposed agree‐ ment was not an element of any charged offense. We acknowledged that a joint venture was still a legitimate sub‐ ject for the Government to explore, but concluded that the gang evidence was not probative: It may be true that common membership in any group, whether it be a gang or a church group, makes it more likely that two people are involved in a given activity together, illegal or 14  No. 15‐3343 not. The inference is certainly weak, however, if the group itself is not somehow connected to the activity at issue. In other words, the fact that [the defendants] are members of a motorcycle club is not especially probative of whether they jointly ventured to distribute drugs, unless the motorcycle club is shown to be involved with drugs. In this case, we are missing that critical connection linking the motorcycle gang with drug trafficking, or any criminal activity for that matter. The government admitted at oral argu‐ ment that it did not present any evidence demonstrating that the Diablos was a criminal organization involved in the distribution of drugs. Id. In the face of the extremely low probative value of gang membership on the charged conduct, we noted “substantial” danger of unfair prejudice. Id. at 865. We reasoned that gangs evoke “images of criminal activity and deviant behavior” and that evidence of gang membership therefore risked “[g]uilt by association.” Id. The risk was heightened by evidence specific to that case involving “devil’s heads and demonic insignia,” which was “clearly inflammatory.” Id. at 866. We therefore concluded that the admission of the evidence was error. With respect to one of the two defendants for whom the Govern‐ ment’s evidence of guilt “was far from overwhelming,” we concluded that the error was not harmless. Id. We affirmed nevertheless the conviction of another defendant against whom the Government had a stronger case and who had in‐ troduced certain gang evidence in his own defense. No. 15‐3343  15 We revisited the issue of gang evidence in United States v. McMahan, 495 F.3d 410 (7th Cir. 2007), judgment vacated in part on other grounds sub nom. Smith v. United States, 552 U.S. 1091 (2008). In McMahan, “[t]wo law enforcement officers were asked what sort of work they did”; the first replied that “he primarily investigated violent street gangs,” and the second that he was “a gang crime specialist” who “participated in gang and narcotics investigations.” Id. at 421–22. In finding no reversible error, we stated, Irvin does not forbid all gang evidence. But, more importantly, nothing about Irvin is partic‐ ularly relevant in the present case. Irvin dealt, as it says, with “gang evidence”—that is, the de‐ fendant’s membership in a gang, the Diablos motorcycle gang. The evidence included testi‐ mony about a large gang tattoo on a defendant’s back, two rings (one which said Diablos and one with a devil’s head), a picture of a vest with gang insignia, a Diablos “greeting card,” a Dia‐ blos wallet, and other gang‐related personal ef‐ fects. One defendant testified that a motorcycle gang would kill him if he testified. The evidence in Irvin was found to be prejudicial and not harmless. The evidence in the present case can‐ not be compared in any meaningful way with that in Irvin. Here, the officers said they did gang investigations; they did not say that is all they did. They did not say that the defendants were members of a gang. In total, the testimony lasted a few minutes in a two‐week trial. The ev‐ idence was not unfairly prejudicial, but even were we somehow to find prejudice, we would 16  No. 15‐3343 also find the admission of the evidence harm‐ less. Id. at 422. Although each case must be assessed on its own facts, Irvin and McMahan are helpful in our review of the district court’s ex‐ ercise of discretion. Mr. Dillard notes, correctly, that the num‐ ber of references in his case exceed those in McMahan. In char‐ acter, however, they are virtually identical. The Government introduced no direct evidence of Mr. Dillard’s alleged gang affiliation. No extraordinarily prejudicial evidence, such as the satanic symbolism in Irvin, ever reached the jury. When its witnesses mentioned gangs generically as part of a discus‐ sion of their law enforcement assignments, the Government did not belabor the point or introduce any evidence about the 22 Hobo gang itself. Moreover, after an objection by defense 22 The Government also notes that, like United States v. McMahan, 495 F.3d 410 (7th Cir. 2007), judgment vacated in part on other grounds sub nom. Smith v. United States, 552 U.S. 1091 (2008), and unlike United States v. Irvin, 87 F.3d 860 (7th Cir. 1996), Mr. Dillard was charged with conspiracy, and we routinely have held that gang evidence may be admitted, as it was in McMahan, “to demonstrate the existence of a joint venture or conspiracy and a relationship among its members.” United States v. King, 627 F.3d 641, 649 (7th Cir. 2010) (internal quotation marks omitted); see also United States v. Santiago, 643 F.3d 1007, 1011 (7th Cir. 2011) (noting that, because the charges include conspiracy, gang membership could be probative of rele‐ vant issues, including “level of mutual trust between the individuals”). Although the Government is correct, and indeed we might have approved of such references in this case had they been evidence essential to the de‐ fendants’ relationship with each other, the references under review were not offered for that purpose and we will not consider their probative value on that basis post hoc. No. 15‐3343  17 counsel to repeated use of the term “gang,” the court in‐ structed the prosecutor to inform future witnesses to avoid the terminology and the issue did not resurface. Although hindsight counsels that it might have been better to have placed this restriction on the Government prior to the presen‐ tation of its case, we certainly cannot say that the district court’s handling of the matter came even close to constituting an abuse of discretion. The majority of the gang references fall squarely within McMahan and are not “unfairly prejudicial.” 23 Id. at 422. Our chief concern is the Government’s questioning of My‐ ers and the Government attorney’s use of the term “Hobo.” Mr. Dillard called Myers to provide an alternate voice identi‐ fication for one of the recorded calls involving Daniels. He claimed that it was the voice of his own son, Davis, rather than Mr. Dillard. The court thereafter ruled that it would allow the Government to question Myers about the relationship of Da‐ vis and the defendants, including common gang membership. The Government asked how the men knew each other, and also asked Myers if he had heard the term “Hobo.” Myers’s answers did not show common gang membership and said nothing about the Hobos; he said only that the two had grown up together and lived together. The court committed no error in permitting the line of questioning. The testimony the Gov‐ ernment attempted to elicit would have been probative of bias 23 Our determination that the district court committed no error in permit‐ ting these references, in context and without a tie to Mr. Dillard or any other particular aggravating circumstances, does not suggest that inci‐ dental references to gang membership are always admissible. The test re‐ mains a balance of the probative value against the conceded prejudice at‐ tached to gang affiliation, consistent with Federal Rule of Evidence 403. 18  No. 15‐3343 as a general matter and generally admissible. See Ozuna, 674 F.3d at 682. More importantly, as the Government notes, the defense offered Myers’s testimony in an attempt to exculpate Mr. Dillard by showing that it was Davis, not Mr. Dillard, on the phone. The Government’s effort to expose the relationship between Davis and the other defendants effectively under‐ mined that attempt, by suggesting that Davis could have been with Mr. Dillard or could have been an additional member of the Dillard‐Chester conspiracy. Balanced against this proba‐ tive value is a very low possibility of prejudice. The Govern‐ ment was unsuccessful in eliciting testimony from Myers as to gang membership and, to its credit, despite the court clearly ruling that the door had been opened to such testi‐ mony, the Government abandoned the line of questioning quite quickly. In certain cases, the district courts have taken even more cautious approaches to gang references, for example by al‐ lowing or instructing counsel to lead witnesses “during the testimony that might mention gangs.” See, e.g., United States v. 24 Harris, 587 F.3d 861, 868 (7th Cir. 2009). However, even in 24 In United States v. Harris, 587 F.3d 861 (7th Cir. 2009), we considered a challenge under Rule 403 to testimony by a law enforcement officer that he had interviewed the defendant regarding “gang and drug activity” and “gang information.” Id. at 867. We acknowledged that “[e]vidence of gang membership can be inflammatory” but found that the district court had taken “several steps … to reduce the prejudicial impact of [the defend‐ ant’s] admitted gang membership.” Id. Among those steps were preclud‐ ing more inflammatory statements, including the defendant’s admission to gang membership, and allowing the Government to lead its witnesses during testimony that might mention gangs. We did not require such measures, but we noted that their use had resulted in “testimony at trial No. 15‐3343  19 those cases, we have stated that our standard simply “ask[s] district courts to consider carefully whether to admit evidence of gang membership and gang activity in criminal prosecu‐ tions.” Id. at 867 (emphasis added). As in Harris, here, the court expressed its desire to err on the side of caution with the evidence, and after counsel objected to the limited law en‐ forcement testimony using the word “gang,” none of which was directly tied to the defendants, the court instructed the prosecutor to inform future witnesses to avoid the terminol‐ ogy. The district court was plainly sensitive to this issue, hav‐ ing discussed it before and during trial on several occasions, and, as is evident from the transcript, was very dismayed by its eventual determination that counsel had opened the door to more gang testimony. We see no abuse of discretion on this record. B. Mr. Dillard also claims that the district court erred in its handling of the risks of exposure to prejudicial media cover‐ age. We review a district court’s handling of the risk of preju‐ dice posed by trial publicity for an abuse of discretion. United States v. Berry, 92 F.3d 597, 600 (7th Cir. 1996). After the jurors had been admonished once by the district court to avoid any media coverage which might expose them to extra‐record evidence in the case, the Sun‐Times published a brief article about the trial, noting its connection to a much larger RICO trial, identifying the defendants as members of that was far less inflammatory than it” might have otherwise been. Id. at 868. 20  No. 15‐3343 the Hobo gang, and stating that the confidential informant in this case, Daniels, was shot more than a dozen times in the face in retaliation for “snitching.” After the story was republished online by another national media outlet, the court met with counsel and decided that the appropriate course would be to question the jury collectively about whether they receive news from online sources. If some jurors did receive news from such sources, those jurors would be questioned individually about what sources they used to ascertain whether any juror had been exposed to the story. In the collective questioning of the jury, several jurors admitted to generally reading online news, but, in individual question‐ ing, none of the jurors indicated that they routinely read the sources that had published the article. Several indicated that they had avoided local media during the trial. At that point, no party indicated that further questioning or action was re‐ quired. Days later, Juror Moskowitz contacted the court to in‐ dicate that the piece had been sent to him, and he also denied speaking to anyone else about the article. Again, no further action was requested. Mr. Dillard now claims that the article posed a danger of extreme prejudice. In his view, the likelihood of that prejudice was so great, given one juror’s admitted exposure, that the 25 court’s treatment of the issue was inadequate. 25 The court and the prosecutor brought the issues to the parties’ attention, and there was no contemporaneous disagreement with the court’s ap‐ proach to it. The Government therefore argues that the issue was waived or forfeited. We are inclined to agree that the issue was waived by Mr. Dillard. The question of how to handle publicity was raised and dis‐ No. 15‐3343  21 A district court’s obligation in cases involving prejudicial publicity is longstanding and straightforward: [T]he procedure required by this Circuit where prejudicial publicity is brought to the court’s at‐ tention during a trial is that the court must as‐ certain if any jurors who had been exposed to such publicity had read or heard the same. Such jurors who respond affirmatively must then be examined, individually and outside the pres‐ ence of the other jurors, to determine the effect of the publicity. However, if no juror indicates, upon inquiry made to the jury collectively, that he has read or heard any of the publicity in question, the judge is not required to proceed further. Margoles v. United States, 407 F.2d 727, 735 (7th Cir. 1969). There is no general presumption of unfairness caused by me‐ dia exposure; instead the danger turns on “the severity of the threat” posed by the particular publicity, specifically, “the na‐ ture of the information so publicized and the degree of juror exposure to it.” United States v. Thomas, 463 F.2d 1061, 1063 (7th Cir. 1972). cussed on multiple occasions in the trial court. At defense counsel’s re‐ quest, the court asked Juror Moskowitz whether he had spoken to anyone else. Mr. Dillard’s attorney never suggested a more aggressive response in the course of trial. Even if we were to consider the issue not waived by counsel’s apparent assent to the district court’s approach because no ob‐ jection was made, we would review for plain error. See United States v. Olano, 507 U.S. 725, 731–32 (1993). 22  No. 15‐3343 In assessing the severity of the threat, it is noteworthy that this was not a case involving the kind of widespread prejudi‐ cial publicity that has concerned our court in other cases. See, e.g., United States v. Malsom, 779 F.2d 1228, 1240 (7th Cir. 1985) (admonishing the jurors to disregard general publicity re‐ garding Libya in a trial involving unlawful exports of imple‐ ments of war); United States v. Solomon, 422 F.2d 1110, 1119 (7th Cir. 1970) (“We cannot say the * * * publicity was so inten‐ sive and extensive or the examination of the entire panel re‐ vealed such prejudice that a court could not believe the an‐ swers of the jurors and would be compelled to find bias or preformed opinion as a matter of law.” (emphasis added) (quoting Beck v. Washington, 369 U.S. 541, 557 (1962)). The only issue in this case was a single article in a print media source, also available online. Although the damaging nature of the article is beyond dispute, its reach was not wide. Mr. Dillard presents two concrete objections to the district court’s handling of the publicity issue: the court failed to strongly and repeatedly admonish the jury to avoid media cov‐ erage and failed to individually question the jurors again after Juror Moskowitz called to report his own exposure and was excused. With respect to frequency, the court instructed the jurors a total of three times to avoid the media or extra‐record mate‐ rial. Certainly, there are cases in which courts have gone fur‐ ther. Mr. Dillard points to United States v. Daddano, 432 F.2d 1119, 1127 (7th Cir. 1970). In that case, the court instructed the jury forty‐six times in fifteen days. We have never required such a high level of precaution, and the facts of Daddano, in‐ volving more than fifty published newspaper articles, some from the front page, are not comparable. Moreover, the trial No. 15‐3343  23 in Daddano occurred before our decision in Margoles, which provided trial courts with a clear framework for addressing potentially prejudicial publicity. The court in this case declined to repeat the admonition after excusing Juror Moskowitz and specifically stated that it believed the jury capable of following the first instruction. This approach is consistent with our general presumption that jurors follow the court’s instructions. See, e.g., United States v. Rodriguez, 929 F.2d 1224, 1228 (7th Cir. 1991) (noting that a court’s striking of testimony and issuing a curative in‐ struction was presumed effective because, “[g]enerally, we assume the jury followed the court’s instruction”). Moreover, the court directed its own library staff to monitor the press, and, through that monitoring, the online publication came to the attention of defense counsel. When the issue of the online article arose, it questioned the jurors, consistent with Margo‐ les. In its final instructions, moreover, the court reminded the jury to consider only the record evidence. Importantly, there was consensus among counsel that this course of proceeding was sufficient, and we have no rule that would suggest other‐ wise. The district court made careful fact‐specific determina‐ tions grounded in prevailing standards. Its understanding of the law was correct; its factual assessment was reasonable; there was no abuse of discretion. Conclusion The district court did not err either in its treatment of the incidental references to gangs at trial or in its ruling that the 24  No. 15‐3343 defense questioning of Myers opened the door to the Govern‐ ment’s further inquiries about gangs. There was no abuse of discretion in the court’s balancing, under Rule 403, of the dan‐ ger of unfair prejudice of the statements against their proba‐ tive value. Furthermore, the court followed the procedures set forth in our cases to assure that Mr. Dillard faced a fair and impartial jury. Accordingly, the judgment of the district court is affirmed. AFFIRMED