United States v. Jerry Harris

Opinion

In the

United States Court of Appeals For the Seventh Circuit

____________________

Nos. 20-1236 & 20-2234 UNITED STATES OF AMERICA, Plaintiff-Appellee,

v.

DAVID GIBSON and JERRY HARRIS, Defendants-Appellants.

____________________

Appeal from the United States District Court for the Northern District of Indiana, South Bend Division.

No. 3:18-cr-33 — Jon E. DeGuilio, Judge.

____________________

ARGUED JANUARY 21, 2021 — DECIDED APRIL 30, 2021

____________________

Before SYKES, Chief Judge, and MANION and ST. EVE, Circuit Judges.

ST. EVE, Circuit Judge. An informant gave South Bend police the number to a phone that drug dealers in the South Bend area were supposedly using to sell drugs. To confirm this tip, officers carried out a series of controlled buys in which confidential informants or undercover officers called the number and followed instructions to buy heroin. Relying on the controlled buys, officers submitted an affidavit to a 2 Nos. 20-1236 & 20-2234

state court judge requesting an order for the phone’s service provider to share 30 days of precise, real-time GPS location data for the phone. The state court judge issued a “court order ” granting the request. Relying on similar affidavits, offic- ers later obtained two more court orders authorizing an additional 60 days of real-time tracking.

The investigation ultimately led officers to two men at the top of the drug-trafficking conspiracy: David Gibson and Jerry Harris. Both defendants were federally indicted for conspiring to distribute heroin. Before trial, the district court denied their motion to suppress evidence obtained through the cellphone tracking. The court treated the state court orders as valid search warrants for the tracking. At trial, officers and cooperators testified to the large-scale drug-trafficking scheme that the defendants had overseen. The jury ultimately convicted both defendants of conspiring to distribute one kilogram or more of heroin. At sentencing, the district court found that the defendants had conspired to distribute a total of 10.5 kilograms of heroin. The defendants now appeal the court’s denial of their motion to suppress. Harris also challenges the drug-quantity calculations at trial and sentencing, the court’s limits on his cross-examination of the cooperators at trial, and his sentence. We affirm the district court’s well-reasoned rulings across the board.

I. Background

A. The Investigation In late 2016 Ryan Williams was charged in Indiana state court for selling drugs. In March 2017, as an act of cooperation , Williams provided South Bend police the number to a phone (ending in -5822) that, he claimed, drug dealers in the

Nos. 20-1236 & 20-2234 3

South Bend area used to sell drugs. Over the next several months, officers made 28 controlled buys using the phone number. In these controlled buys, a confidential informant or undercover officer would call the number and follow instructions to buy heroin. Each controlled buy involved half-gram increments of heroin. Different dealers, including Williams himself on occasion, would show up to deliver the drugs. The dealers would sell indiscriminately to anyone who called the number. At meeting spots, cars lined up to buy drugs.

A few months into the investigation, officers sought to establish surveillance of the phone’s location. In July 2017, offic- ers submitted an affidavit to an Indiana state court judge requesting an order for Sprint, the phone’s service provider, to supply 30 days of precise, real-time GPS location data for the phone. The affidavit described two separate controlled buys in which a confidential informant had called the phone and met someone who sold him heroin. One of the controlled buys had occurred two and a half months earlier; the other had occurred the day before. The affidavit did not cite Federal Rule of Criminal Procedure 41 (which governs search warrants based on probable cause). Indeed, the affidavit did not mention probable cause at all. Instead, it cited federal statutes governing the installation of mobile tracking devices, pen registers , and trap and trace devices. See 18 U.S.C. §§ 3117, 3124.

Based on the affidavit, an Indiana judge signed a “court order” finding probable cause to believe that the user of the -5822 phone had engaged in illegal drug possession and trafficking , and that precise tracking of the phone’s location would facilitate the user’s apprehension. Thus, the judge ordered Sprint to supply 30 days of precise, real-time GPS location data for the phone. As authority for the order, the judge 4 Nos. 20-1236 & 20-2234

cited Rule 41, the Stored Communications Act, see 18 U.S.C. § 2703, and the federal statutes governing mobile tracking devices , pen registers, and trap and trace devices, see 18 U.S.C. §§ 3117, 3123, 3124. Per the order, Sprint gave officers 24-hour access to the phone’s precise location for 30 days.

At the end of the 30-day period, officers submitted a second affidavit, requesting 30 more days of real-time cellphone tracking. The affidavit explained that, since obtaining the first court order, officers had carried out several more controlled buys using the same phone number. It described one of them in detail. The affidavit added that “this is a very complex organization with approximately fifteen members who utilize the [phone] to facilitate drug trafficking.” In all other respects, the second affidavit mirrored the first. Based on the affidavit, the state court judge signed another order, essentially identical to the first, authorizing 30 more days of real-time cellphone tracking.

The same series of events happened one more time. At the end of the second 30-day period, officers submitted a third affidavit , requesting 30 more days of GPS tracking data for the phone. This affidavit closely resembled the second one. It described in detail “one of several undercover officer buys” made in the previous 30-day period. The state court judge signed another materially identical order authorizing 30 more days of real-time cellphone tracking.

Officers eventually recovered the -5822 phone in October 2017 when they pulled over a man named Raymond Love for a traffic violation. Love had two “flip phones” on him, including the -5822 phone. Throughout the traffic stop, both phones rang nonstop.

Nos. 20-1236 & 20-2234 5

While tracking the -5822 phone, officers observed that it was located at various times in houses that they later connected to Gibson and Harris. Officers executed a search warrant on the home associated with Harris, where they found a digital scale and almost $4,000 cash. B. Charges and Motion to Suppress A federal grand jury indicted Gibson and Harris with one count of conspiring to distribute more than one kilogram of heroin between March and October 2017. See 21 U.S.C. §§ 841(b)(1)(A), 846. Before trial, the defendants moved to suppress all evidence obtained through the phone tracking. They maintained that officers could not track the phone without a search warrant. Following a hearing, the district court denied the defendants’ motion to suppress. It ruled that the state court orders were valid warrants for the phone tracking. C. Trial The cases against Gibson and Harris were consolidated for trial. Various law enforcement officers testified, as did a few cooperating witnesses who had participated in the drug-trafficking conspiracy. Two of these cooperators were Williams and Loveless Daniel Naylor. Like Williams, Naylor was one of the street-level dealers who sold drugs directly to calling customers. Before the defendants went to trial, Williams pled guilty to conspiring to distribute one kilogram or more of heroin . Naylor pled guilty to possession with intent to distribute heroin.

Williams testified to the details of the drug-trafficking operation . Gibson ran the operation and Harris was his “righthand man.” There were 20 to 25 dealers who worked in shifts. Gibson or Harris gave the dealers phones and “packs” of 6 Nos. 20-1236 & 20-2234

drugs to sell. The packs contained 22 to 24 individually packaged half-gram bags of heroin. A single dealer could go through as many as five or six packs on a weekend day. Most customers bought between one and ten half-gram bags at a time, but some bought up to fifteen. There were two phone numbers that customers could call: the -5822 number and another number ending in -9243. Williams testified that a transaction required, at most, “probably about three” phone calls. An officer involved in the controlled buys testified that a successful buy generally required between one and four calls, though sometimes more were necessary. Naylor testified similarly regarding the details of the drug-trafficking operation. He added that, between March and October 2017, Gibson gave him between 600 and 700 grams of heroin to sell.

Over Harris’s objection, the court forbade defense counsel from cross-examining the cooperating witnesses about the specific sentences they hoped to avoid by testifying for the government. The court explained that defense counsel could ask about mandatory minimums and “substantial sentences” but could not reference specific terms of imprisonment. Otherwise the jury might infer what sentences the defendants themselves would receive if convicted.

The government introduced only 6.5 grams of heroin at trial. To prove the full quantity of drugs involved in the conspiracy , the government called DEA Task Force Officer Joseph Focosi. Officer Focosi used two different formulas to calculate drug quantity. First, he relied on Williams’s testimony that he alone could sell up to six “packs” per day, with each pack containing 22 to 24 half-gram bags. Assuming that the dealers sold only one pack per day, and rounding the amount of heroin in a pack down to 10 grams, Officer Focosi testified that

Nos. 20-1236 & 20-2234 7

the dealers would have sold 2.1 kilograms of heroin over a seven-month period (10 grams x 210 days = 2,100 grams). Defense counsel did not object to this testimony.

Officer Focosi’s second formula extrapolated drug quantity from phone calls. A DEA analyst had testified that, from the end of March 2017 through mid-October 2017, there were about 50,000 successful calls to the -5822 phone and 34,000 successful calls to the -9243 phone. Conservatively estimating that each transaction involved six phone calls and half a gram of heroin, Officer Focosi testified that the conspiracy involved 7 kilograms of heroin (84,000 total calls / 6 calls per transaction = 14,000 transactions x .5 grams per transaction = 7,000 grams).

Harris’s counsel objected that the underlying testimony about calls per transaction had “not been that precise” and that Officer Focosi was offering an opinion based on his “experience and expertise” in deciding what number of calls to use. The district court overruled the objection. The court explained that whether six calls was a precise estimate went to the weight of the testimony, and that Officer Focosi was relying on what he had learned from the investigation rather than “exercising any experience.”

Defense counsel also objected to Officer Focosi’s assumption that all 84,000 calls were between dealers and customers, when “presumably, there would be calls between a dealer and other people as well.” The court overruled this objection, too, explaining: “I think that’s true, but that, again, goes to weight, and I think that’s something you can explore on cross examination .”

On cross-examination, Officer Focosi testified that he chose six calls per transaction because “it was more than one 8 Nos. 20-1236 & 20-2234

through four” (the range that witnesses had testified to). He admitted that he could have chosen a larger number, anywhere from 5 to 24. He also admitted that he did not know the content of specific calls and conceded that some calls “might have had nothing to do with heroin deals.”

The jury found both defendants guilty of conspiring to distribute one kilogram or more of heroin. D. Sentencings The probation officer calculated each defendant’s advisory Guideline range as 360 months to life. Both defendants had a criminal history category of VI, but Gibson’s offense level (38) was one point higher than Harris’s (37) because Gibson was an “organizer or leader” of the drug-trafficking operation whereas Harris was a “manager or supervisor.” See USSG § 3B1.1.

Harris objected to the probation officer’s conclusion that he had conspired to distribute 10.5 kilograms of heroin. See USSG § 2D1.1(b)(18)(3) (increasing offense level for crimes involving 10 to 30 kilograms of heroin). To reach that number, the probation officer used Officer Focosi’s second formula from trial. Unlike Officer Focosi, however, the probation officer assumed, based on the “most conservative” estimate at trial, that every drug sale required four, rather than six, calls (84,000 total calls / 4 calls per transaction = 21,000 transactions x .5 grams per transaction = 10,500 grams). Harris objected to this methodology as “speculative and not reasonable.” The court overruled his objection. It found, based on the evidence at trial, that the proposed drug-quantity calculation was “both certain and reasonable.” There was no evidence that any dealers had used the shared drug phones to make

Nos. 20-1236 & 20-2234 9

personal calls, and even if they had, the number of personal calls “would have to be dramatic to defeat the finding of ten or more kilograms” given how conservative the four-call and half-gram numbers were.

The court sentenced both defendants well below the low end of the Guidelines range. Specifically, it sentenced Gibson to 240 months’ imprisonment and Harris to 262 months. The court explained the discrepancy at Harris’s sentencing. At 47, Gibson was a bit older than Harris, who was 38. Moreover, Gibson had recently received a 144-month consecutive sentence in another case for carrying a firearm in relation to a drug-trafficking crime. See 18 U.S.C. § 924(c). In the court’s view, Gibson’s older age and consecutive sentence influenced his risk of reoffending upon release. The same calculus did not apply to Harris, who was younger and not facing a consecutive sentence. The court went on to determine that Harris warranted a three-level reduction, which corresponded to a sentencing range of 262 to 327 months. It found that “a sentence in this range will satisfy the purposes of the sentencing statute” and further “note[d] that even if I calculated the Guideline range differently in the first instance, I would vary to this same range based on my consideration of the 3553(a) factors as a whole.” A review of the § 3553(a) factors “persuade [d] the Court that a sentence at the low end of the now revised recommended sentencing range is appropriate.” The defendants timely appealed.

II. Discussion

On appeal the defendants challenge the district court’s admission of evidence obtained through the cellphone tracking. Harris also challenges the drug-quantity calculations at trial and sentencing, the district court’s limits on his cross-

10 Nos. 20-1236 & 20-2234

examination of cooperating witnesses, and the reasonableness of his sentence. A. Cellphone Tracking The defendants’ first contention is that officers violated the defendants’ Fourth Amendment rights by tracking the location of the -5822 phone without a warrant based on probable cause. They submit that the “court orders” authorizing the tracking were not valid warrants because they cited the Stored Communications Act, which requires a lesser showing than probable cause. See 18 U.S.C. § 2703(d) (requiring only “specific and articulable facts showing that there are reasonable grounds to believe that the [information sought is] relevant and material to an ongoing criminal investigation”); see also United States v. Castro-Aguirre, 983 F.3d 927, 934 (7th Cir. 2020) (describing this standard as “significantly lower than the probable-cause requirement for a warrant”). They contend as well that the underlying affidavits did not supply probable cause for a search warrant.

A Fourth Amendment “search” generally requires a warrant based on probable cause. Carpenter v. United States, 138 S. Ct. 2206, 2221 (2018); Katz v. United States, 389 U.S. 347, 357 (1967). The government does not dispute that the cellphone tracking in this case amounted to a “search” requiring a warrant , and we assume for purposes of this appeal that it did. Nonetheless, the government maintains that the search was lawful because a valid warrant authorized it. The district court agreed and denied the defendants’ motion to suppress. We review the district court’s legal conclusions de novo and its factual findings for clear error. United States v. Jackson, 962 F.3d 353, 357 (7th Cir. 2020).

Nos. 20-1236 & 20-2234 11

A valid search warrant “require[s] only three things”: (1)

an independent magistrate issuing it; (2) a showing of probable cause “that the evidence sought will aid in a particular apprehension or conviction for a particular offense”; and (3) a particular description of “the things to be seized, as well as the place to be searched.” Dalia v. United States, 441 U.S. 238, 255 (1979) (internal quotations and citations omitted); accord United States v. Brewer, 915 F.3d 408, 414 (7th Cir. 2019). Substance matters more than form in this context. Thus, a court order can satisfy the warrant requirement even if it is not labeled a “warrant.” Dalia, 441 U.S. at 256 (wiretap order was a valid warrant); United States v. Ning Wen, 477 F.3d 896, 898 (7th Cir. 2007) (same). And a warrant that finds probable cause and cites Rule 41 satisfies the Fourth Amendment even if it also recites the lower standard of the Stored Communications Act. United States v. Sanchez-Jara, 889 F.3d 418, 421 (7th Cir. 2018).

The court orders in this case satisfy the requirements for a search warrant. First, the defendants do not contend, and there is no reason to believe, that the state court judge who issued the orders was anything but neutral and detached. Next, the orders cited Rule 41 and found probable cause to believe that the cellphone tracking would lead to the apprehension of drug traffickers. Last, the orders particularly described the object of the search: the location of the -5822 phone. With respect to the particularity requirement, we have held that “a warrant authorizing police to follow an identified phone, to see where it goes and what numbers it calls, particularly describes the evidence to be acquired.” Sanchez-Jara, 889 F.3d at 421; accord Brewer, 915 F.3d at 414 (“Judges must describe the specific person, phone, or vehicle to be tracked to satisfy the Fourth Amendment’s particularity requirement.”)

12 Nos. 20-1236 & 20-2234

(emphasis added). It makes no difference that the court orders were not labeled “warrants,” or that they cited, in addition to Rule 41, other statutes including the Stored Communications Act. See Dalia, 441 U.S. at 256; Ning Wen, 477 F.3d at 898; Sanchez-Jara, 889 F.3d at 421.

We reject the defendants’ contention that the underlying affidavits did not supply probable cause. Probable cause for issuance of a search warrant exists if there is “a fair probability that contraband or evidence of a crime will be found in a particular place.” Illinois v. Gates, 462 U.S. 213, 238 (1983). The state court judge’s finding of probable cause “carries a strong presumption of correctness.” Sanchez-Jara, 889 F.3d at 421. Our task as “a reviewing court is simply to ensure that the [state court judge] had a substantial basis for concluding that probable cause existed.” Gates, 462 U.S. at 238–39 (internal quotations, citation, and alterations omitted).

We have held that a properly executed controlled buy is generally “a reliable indicator as to the presence of illegal drug activity.” United States v. Sidwell, 440 F.3d 865, 869 (7th Cir. 2006); see also United States v. Bacon, 991 F.3d 835, 839 (7th Cir. 2021) (“[C]ontrolled buys ordinarily go a long way toward establishing probable cause.”). Here, the affidavits described several controlled buys in detail, including the arrangement of the buys through calls to the -5822 phone. The first affidavit described two controlled buys involving the same confidential informant, whom officers searched, wired, and closely monitored. On two separate occasions, the confidential informant called the -5822 phone and successfully bought heroin, which he then turned over to the police. The second and third affidavits referenced the earlier orders and described additional, more recent controlled buys in which

Nos. 20-1236 & 20-2234 13

undercover officers called the same phone and bought heroin. We have little trouble concluding that these controlled buys gave the state court judge a substantial basis for a finding of probable cause to track the location of the phone sufficient to support each of its orders. See Gates, 462 U.S. at 238–39.

The defendants do not contest the method or execution of the controlled buys. Instead, they complain that one of the controlled buys in the first affidavit was two and a half months old. That is true, but the other controlled buy in the first affidavit took place the day before the state court judge signed the first order. So, to the extent that the first controlled buy might have been “stale,” the “more recent” controlled buy mitigated that issue by showing that “the same kind of criminal activity continued” through the present day. United States v. Rees, 957 F.3d 761, 769–70 (7th Cir. 2020).

The defendants also suggest that the affidavits did not supply probable cause to track them, given that the affidavits did not mention them by name. But probable cause for a search warrant need not be tied to any particular person. See Gates, 462 U.S. at 238 (requiring “a fair probability that contraband or evidence of a crime will be found in a particular place”) (emphasis added); see also Bacon, 991 F.3d at 841. In this case, officers knew that the users of the -5822 phone were selling drugs, but they did not know who the users were. That is why they wanted to track the phone—to apprehend the users for drug trafficking. In line with that mission, the state court judge found probable cause to believe that the cellphone tracking would facilitate the apprehension of drug traffickers. This finding was enough for probable cause, even though the judge did not identify Gibson or Harris by name. See Dalia, 441 U.S. at 255 (“[T]hose seeking the warrant must 14 Nos. 20-1236 & 20-2234

demonstrate to the magistrate their probable cause to believe that ‘the evidence sought will aid in a particular apprehension or conviction’ for a particular offense.” (quoting Warden, Md. Penitentiary v. Hayden, 387 U.S. 294, 307 (1967))).

We therefore conclude that officers had a valid warrant to track the -5822 phone. The district court correctly denied the defendants’ motion to suppress. We need not address the government ’s alternative arguments that the good-faith exception applies, see United States v. Leon, 468 U.S. 897, 922 (1984), and that the defendants lack “standing” to challenge the cellphone tracking, see Byrd v. United States, 138 S. Ct. 1518, 1530 (2018). We note, however, that an officer’s reliance on a warrant presumptively establishes that the officer acted in good faith in carrying out a search. Leon, 468 U.S. at 922; United States v. Lickers, 928 F.3d 609, 618 (7th Cir. 2019). Moreover, we are skeptical that either defendant has standing to challenge the cellphone tracking. There was no evidence at the suppression hearing that either defendant personally possessed or used the -5822 phone during the 90-day tracking period . The defendants maintain that the government conceded the defendants’ possession of the phone. But even if that is true, the evidence from trial suggests that the defendants had, at most, on-and-off possession of this cellphone that approximately two dozen drug dealers shared to sell drugs. And there is no evidence that either defendant ever used the phone for personal, rather than commercial, purposes. Just as drug dealers who briefly occupy stash houses while packaging drugs lack a legitimate expectation of privacy in the stash houses, the defendants here would seem to lack a legitimate expectation of privacy in the whereabouts of a shared drug phone. See Minnesota v. Carter, 525 U.S. 83, 90–91 (1998); United States v. Gray, 491 F.3d 138, 147 (4th Cir. 2007); H. Hunter

Nos. 20-1236 & 20-2234 15

Bruton, Note, The Shifting Nature of Stash-House Standing and Sentencing, 42 N.Y.U. Rev. L. & Soc. Change 351, 358 (2018). Ultimately, we need not resolve this issue. And given the parties ’ dispute as to whether the government conceded the defendants ’ possession of the phone, it is best to leave it for another day. B. Drug-Quantity Calculations Apart from the cellphone tracking, Harris takes issue with the drug-quantity calculations at trial and at his sentencing.

1. Trial Harris submits that Officer Focosi’s trial testimony was unreliable and ran afoul of the rules governing expert testimony . See Fed. R. Evid. 702; Fed. R. Crim. P. 16(a)(1)(G); see also United States v. Gaytan, 649 F.3d 573, 582 (7th Cir. 2011). As such, he challenges the sufficiency of the drug-quantity evidence and seeks to overturn the jury’s finding that he conspired to distribute at least one kilogram of heroin. See 21 U.S.C. § 841(b)(1)(A) (providing a mandatory-minimum sentence for offenses involving one kilogram or more of heroin).

Harris did not move for a judgment of acquittal, so we review the jury’s drug-quantity determination only for a “manifest miscarriage of justice.” United States v. Chaparro, 956 F.3d 462, 468 (7th Cir. 2020) (internal quotations and citation omitted ). This means that the jury’s verdict stands unless “the record is devoid of evidence pointing to guilt, or … the evidence on a key element of the offense was so tenuous that a conviction would be shocking.” Id. (internal quotations and citation omitted).

Harris’s challenge to the jury’s drug-quantity determination goes nowhere because the government offered the jury 16 Nos. 20-1236 & 20-2234

two independent ways to find that Harris conspired to distribute more than one kilogram of heroin and Harris challenges only one of these methods. First, Officer Focosi testified , based on Williams’s testimony that he could sell up to six “packs” of heroin on a weekend day, the conspiracy involved 2.1 kilograms of heroin. To reach that number, Officer Focosi conservatively estimated that the dealers sold only one 10-gram pack per day during the seven-month conspiracy (10 grams x 210 days = 2,100 grams). Officer Focosi’s second formula extrapolated drug quantity from phone calls. A DEA analyst had testified that there were about 84,000 successful calls to the two drug phones during the conspiracy. Conservatively estimating that each transaction involved six phone calls and half a gram of heroin, Officer Focosi testified that the conspiracy involved 7 kilograms of heroin (84,000 total calls / 6 calls per transaction = 14,000 transactions x .5 grams per transaction = 7,000 grams). Harris claims that Officer Focosi’s second formula was inadmissible expert testimony that the government failed to timely disclose. But even if that is true, Harris has never challenged Officer Focosi’s first formula. Officer Focosi’s first formula independently sustains the jury’s drug- quantity finding and renders harmless any error in the district court’s admission of Officer Focosi’s second formula. See Fed. R. Crim. P. 52(a); United States v. Jett, 908 F.3d 252, 265 (7th Cir. 2018). Thus, Harris has not identified a prejudicial error in the drug-quantity calculation at trial—much less a manifest miscarriage of justice. Chaparro, 956 F.3d at 468.

2. Sentencing Harris also challenges the district court’s calculation of the drug quantity at sentencing. Whereas the government asked the jury to find that Harris conspired to distribute one

Nos. 20-1236 & 20-2234 17

kilogram or more of heroin, at sentencing the district court had to calculate the actual amount of heroin involved in the conspiracy and reasonably foreseeable to Harris. See USSG § 2D1.1, comment. (n.5); USSG § 1B1.3(a)(1)(B). Relying on the probation officer’s recommendations and the evidence at trial, the court found that the government had established by a preponderance of the evidence that Harris conspired to distribute 10.5 kilograms of heroin. The court reached this number by applying Officer Focosi’s second formula and assuming that each drug transaction required four, rather than six, calls (84,000 total calls / 4 calls per transaction = 21,000 transactions x .5 grams per transaction = 10,500 grams). Harris maintains that this calculation rested on speculative and unreliable data. More specifically, he claims that 4 calls per transaction is an arbitrary estimate, and that there is no evidence that all 84,000 calls were drug related.

A defendant attacking a district court’s factual findings at sentencing has “a steep hill to climb.” United States v. Ranjel, 872 F.3d 815, 818 (7th Cir. 2017). We “will not disturb a sentencing court’s factual findings unless they are clearly erroneous .” Id. The government must prove drug quantity by a preponderance of the evidence. United States v. Medina, 728 F.3d 701, 705 (7th Cir. 2013).

The Sentencing Guidelines tie a defendant’s offense level to the quantity of drugs involved in the offense. See USSG § 2D1.1(c). In a drug conspiracy, “each conspirator is responsible for both the drug quantities directly attributable to him and amounts involved in reasonably foreseeable dealings by co-conspirators.” United States v. Austin, 806 F.3d 425, 431 (7th Cir. 2015); see USSG § 1B1.3(a)(1)(B). If the quantity of drugs seized “does not reflect the scale of the offense,” a court must 18 Nos. 20-1236 & 20-2234

“approximate” the total drug quantity. USSG § 2D1.1, comment . (n.5). To make such an approximation, the court “may consider, for example, the price generally obtained for the controlled substance, financial or other records, similar transactions in controlled substances by the defendant, and the size or capability of any laboratory involved.” Id.

Drug-quantity calculations are “not an exact science.”

United States v. Sewell, 780 F.3d 839, 849 (7th Cir. 2015). “Determining drug quantities under the Sentencing Guidelines is often difficult, and district courts may make reasonable though imprecise estimates based on information that has indicia of reliability.” United States v. Bozovich, 782 F.3d 814, 818 (7th Cir. 2015). The Guidelines do not permit “‘nebulous eyeballing ,’” but “some amount of reasoned ‘speculation and reasonable estimation’” is permissible. United States v. Hollins, 498 F.3d 622, 631 (7th Cir. 2007) (quoting United States v. Jarrett , 133 F.3d 519, 530 (7th Cir. 1998)).

Here, the district court properly relied on the evidence in the record to conservatively calculate the drug quantity. Its calculation was not clear error. All three numbers that the court plugged into Officer Focosi’s second formula (84,000 total calls / 4 calls per transaction = 21,000 transactions x .5 grams per transaction = 10,500 grams) had a firm basis in the record. Harris does not challenge the half-gram-per-transaction figure, nor could he. Williams, Naylor, and an officer involved in the controlled buys testified that the heroin was prepackaged and sold in half-gram increments. Most customers bought between one and ten half-gram bags at a time, though some bought as many as fifteen. One half-gram bag per transaction was thus an extremely conservative estimate.

Nos. 20-1236 & 20-2234 19

Harris describes the four-calls-per-transaction estimate as arbitrary. But Williams, who regularly sold heroin to buyers who called the drug phones, testified that the transactions required , at most, three calls. Further, an officer who participated in the controlled buys testified that the transactions usually required between one and four calls. Given this evidence , the four-call estimate was reliable and indeed very conservative. Harris counters that the discrepancy between the calls-per-transaction figure at trial (six calls) and sentencing (four calls) demonstrates the arbitrariness of both numbers . But at trial, the government sought to prove a minimum drug quantity beyond a reasonable doubt. See 21 U.S.C. § 841(b)(1)(A) (providing a mandatory-minimum sentence for offenses involving one kilogram or more of heroin); see also Alleyne v. United States, 570 U.S. 99, 103 (2013) (holding any fact that increases the penalty for a crime must be proven to a jury beyond a reasonable doubt). The actual quantity of drugs was not an element of the offense. See United States v. Abdulahi, 523 F.3d 757, 760 (7th Cir. 2008). At sentencing, by contrast, the government sought to prove the actual drug quantity by a preponderance of the evidence. See id. at 760–61. Given the different burdens of proof and required showings, there was nothing inconsistent about using different estimates of calls per transaction at trial and sentencing.

Harris directs most of his criticism to the third figure—

84,000 total calls. But this number, too, had a firm basis in the record. A DEA analyst reviewed phone records and testified that there were 84,000 successful calls to the two drug phones during the conspiracy. Harris insists that any number of these calls could have been unrelated to drugs. And to be sure, Officer Focosi conceded at trial that he could not rule out the possibility that some of the calls were unrelated to drug 20 Nos. 20-1236 & 20-2234

transactions. But, as the district court recognized, there is no evidentiary support for Harris’s speculative hypothesis that the two dozen drug dealers who shared the drug phones were also receiving personal calls on the phones. All evidence points to the opposite conclusion—these were drug phones that on-duty dealers used to continuously make drug sales. When Williams was asked at trial who called the phones, he responded, “people who wanted to buy drugs.” Naylor testified that “customers” called the phones. The government had to prove drug quantity by a preponderance of the evidence— not to an absolute certainty. On these facts, the district court did not err in calculating the heroin quantity.

Harris relies heavily on United States v. Howard, 80 F.3d 1194 (7th Cir. 1996). But the problem in Howard was the probation officer’s failure to explain the evidentiary basis for the drug-quantity calculation. As we said: “Where either the probation officer or the prosecution offers an estimate of the drug quantities for which the defendant should be held responsible , the defendant ought to be on notice of all assumptions, rationale, and methodology underlying the calculation.” Id. at 1204. Here, the government’s methodology was transparent. It derived from specific testimony that the government offered at trial. Harris tested the reliability of that testimony through cross-examination. Indeed, he bases his challenge to the court’s drug-quantity calculation primarily on concessions that he obtained from Officer Focosi at trial. Unlike the defendant in Howard, Harris could, and did, challenge the government’s methodology because he knew the evidentiary basis for it. A disputed evidentiary basis is not the same thing as no evidentiary basis. This distinction renders Harris’s outof -circuit citations equally inapposite. See United States v. Sepulveda, 15 F.3d 1161, 1198 (1st Cir. 1993); United States v.

Nos. 20-1236 & 20-2234 21

Shonubi, 998 F.2d 84, 90 (2d Cir. 1993); United States v. Collado, 975 F.2d 985, 998 (3d Cir. 1992).

Harris also complains that the government’s use of phone calls to extrapolate drug quantity was unprecedented. That is not entirely true, as other circuits have upheld drug-quantity calculations extrapolated, at least in part, from phone calls. See, e.g., United States v. Green, 40 F.3d 1167, 1175 (11th Cir. 1994). More importantly, though, the district court’s job was to make a reasonable estimate of drug quantity on the record before it. Austin, 806 F.3d at 431. This was a unique case. The evidence showed that Gibson and Harris oversaw a large drug-trafficking operation in which approximately two dozen drug dealers sold drugs indiscriminately to anyone who called either of two shared drug phones. The court had to approximate the quantity of drugs involved in this conspiracy, and it did so reasonably. Nothing required the court to locate a prior case approving of its methodology for calculating drug quantity.

For these reasons, the district court did not clearly err in calculating the drug quantity at Harris’s sentencing. And even if it did, any error was harmless. At Harris’s sentencing, the court said on the record that, even if it had calculated the Guidelines range differently, it would vary to the same range and impose the same sentence based on its independent consideration of the sentencing factors under 18 U.S.C. § 3553(a). When a sentencing court bases its sentence on the § 3553(a) factors and says it would have imposed the same sentence regardless of the Guidelines range, an error in calculating the Guidelines range may be harmless. See, e.g., United States v. Snyder, 865 F.3d 490, 500–01 (7th Cir. 2017); Molina-Martinez v. United States, 136 S. Ct. 1338, 1346–47 (2016). That happened 22 Nos. 20-1236 & 20-2234

here. The district judge presided over the trial and knew the facts of the conspiracy inside and out. He thoroughly explained his sentence and anchored it in the § 3553(a) factors. Even if his drug-quantity calculation was wrong, the district judge knew the scale of the conspiracy and Harris’s role in it. He explained that he would have imposed the same sentence regardless of the Guidelines range. On these facts, any error in calculating drug quantity was harmless. C. Cross-Examination of Cooperators Harris argues next that the district court improperly constrained his cross-examination of cooperating witnesses. He says he should have been allowed to question Williams and Naylor about the specific sentences they hoped to avoid by testifying for the government.

The Sixth Amendment guarantees a criminal defendant the opportunity to effectively cross-examine witnesses against him. United States v. Trent, 863 F.3d 699, 704 (7th Cir. 2017). But this right is not absolute: “a district court has discretion to place reasonable limits on cross-examination, especially when necessary to prevent irrelevant or confusing evidence from being presented to the jury.” Id. Such irrelevant or confusing evidence includes “information from which [the jury] could infer defendants’ potential sentences.” Id. at 705.

Federal juries do not decide sentences in noncapital cases, so specific sentencing information “might confuse or mislead the juries in their true task: deciding defendants’ guilt or innocence .” Id. The risk is that “the reality of a serious sentence could prejudice the jury and cause it to acquit the defendants of crimes they actually committed.” United States v. Hunter, 932 F.3d 610, 619 (7th Cir. 2019). For these reasons, we have

Nos. 20-1236 & 20-2234 23

held that district courts may in some circumstances bar defense counsel from cross-examining cooperating witnesses about “the exact length” of the witness’s potential sentence. Trent, 863 F.3d at 706; Hunter, 932 F.3d at 620.

Our standard for reviewing a district court’s limit on cross-examination “depends on whether the court’s limit ‘directly implicates the core values of the Confrontation Clause.’” Trent, 863 F.3d at 704 (quoting United States v. Recendiz , 557 F.3d 511, 530 (7th Cir. 2009)). If it does, our review is de novo; if it does not, we review for abuse of discretion. Id. The core values of the Confrontation Clause include allowing the defendant “to expose a witness’s motivation for testifying, his bias, or his possible incentives to lie.” Id. at 705 (quoting Recendiz, 557 F.3d at 530). “But that value is only offended when ‘the defense is completely forbidden from exposing the witness’s bias.’” Id. (quoting United States v. Sanders, 708 F.3d 976, 990 (7th Cir. 2013)). When the defense has “a reasonable opportunity to question witnesses about their biases, the Sixth Amendment is not implicated.” Id.

The district court’s limits on Harris’s cross-examination of Williams and Naylor did not implicate the core values of the Sixth Amendment. The court imposed a limited restriction on Harris’s cross-examination. It only forbade Harris from questioning Williams and Naylor about the exact sentences they hoped to avoid. It otherwise gave Harris free reign to ask the witnesses about mandatory minimums and “substantial sentences .” See Trent, 863 F.3d at 705–06.

And Harris took full advantage. Williams testified on cross-examination that, although he was a felon with two guns on him at the time of his arrest, the government never charged him with being a felon in possession of a firearm. See 24 Nos. 20-1236 & 20-2234

18 U.S.C. § 922(g)(1). The government also never charged him with carrying a firearm in relation to a drug-trafficking crime—an offense that, he admitted, carried “a substantial mandatory minimum.” See 18 U.S.C. § 924(c). Williams further conceded that he was testifying for the government to get his “time cut” on the charge that he pled guilty to—conspiracy to distribute at least a kilogram of heroin—which carried “substantial penalties” and a “substantial mandatory minimum .” On top of that, Williams admitted that the government never filed a sentencing enhancement against him even though he had two qualifying prior convictions. See 21 U.S.C. § 851.

For his part, Naylor testified on cross-examination that the government agreed to dismiss a § 924(c) count against him carrying a “substantial mandatory minimum.” The government also agreed not to file an applicable sentencing enhancement on the charge that Naylor pled guilty to—possession with intent to distribute heroin—which would have resulted in “even longer punishment in prison.” Naylor added that he was hoping to receive “a further sentence reduction” for testifying against the defendants. He understood that the government could move for such a reduction based on its evaluation of his trial testimony.

This testimony shows that Harris could, and did, question the cooperators at length about their potential biases and motives to lie. As such, we review for abuse of discretion. Trent, 863 F.3d at 704. To decide if the court abused its discretion, we ask “whether the jury had sufficient information to make a discriminating appraisal of the witness’s motives and biases.” Id. at 706 (quoting Sanders, 708 F.3d at 991). The district court gave Harris wide latitude to cross-examine the cooperators

Nos. 20-1236 & 20-2234 25

about everything except the exact sentences they hoped to avoid by testifying for the government. As set forth above, Harris thoroughly probed the cooperators’ biases and motives to lie. Both witnesses admitted that they faced significant sentences that they hoped the government would ask the court to reduce in light of their testimony. As in Trent, the court “did not err, let alone abuse its discretion,” in keeping the cooperators exact sentences off limits. Id.

Harris counters that Trent and Hunter should not apply here because Williams and Naylor were not codefendants in the defendants’ case. But what matters under Trent and Hunter is that the jury might infer the defendants’ potential sentences from the cooperators’ potential sentences. See Trent, 863 F.3d at 705; Hunter, 932 F.3d at 619. Here, Williams was charged separately, but he faced the same exact charge as the defendants: conspiracy to distribute at least a kilogram of heroin . See 21 U.S.C. §§ 841(b)(1)(a), 846. Indeed, the district court instructed the jury that Williams “was involved in and has pled guilty to charges relating to the crime the defendants are charged with committing.” Given the identical nature of Williams ’s charges and the defendants’ charges, the court was well within its discretion to limit Harris’s cross-examination of Williams.

Naylor, on the other hand, pled guilty to possession with intent to distribute heroin, see 21 U.S.C. § 841(a)(1), a different crime corresponding to a lower mandatory minimum. The district court acknowledged this difference, but explained its decision to nonetheless limit Harris’s cross-examination of Naylor:

While it’s not the exact same charge, it is under the same statute. I believe the offense to which he pled 26 Nos. 20-1236 & 20-2234

guilty to constitutes one of the lesser includeds that the Court intends to give, and the plea agreement does, in fact, state that there’s a mandatory minimum of five years, which I think would give some insight into what these defendants might be looking at if convicted of these charges. The court carefully and permissibly exercised its discretion . We acknowledge, as we did in Hunter, that sharing specific sentencing information with the jury is of concern when the cooperators and defendants face the same or similar charges. Hunter, 932 F.3d at 619. Here, though, the court reasonably concluded that the charge to which Naylor pled guilty was similar enough to the defendants’ charges that the jury might try to “deduce or infer the sentences facing the similarly-charged defendants” from information about Naylor’s sentence. Id. On these facts, we see no abuse of discretion .

We thus affirm the district court’s limits on Harris’s cross-

examination of Williams and Naylor. Harris’s only other argument on this score is a perfunctory and conclusory plea for us to overrule Trent and Hunter. This argument is waived. Puffer v. Allstate Ins. Co., 675 F.3d 709, 718 (7th Cir. 2012) (“[E]ven arguments that have been raised may still be waived on appeal if they are underdeveloped, conclusory, or unsupported by law.”). D. Harris’s Sentence Finally, Harris argues that his sentence of 262 months’ imprisonment is substantively unreasonable because it is 22 months longer than Gibson’s, even though Gibson was an “organizer or leader” of the drug-trafficking scheme whereas

Nos. 20-1236 & 20-2234 27

Harris was only a “manager or supervisor.” See USSG § 3B1.1 (requiring a four-level enhancement for organizers or leaders and a three-level enhancement for managers or supervisors).

We review the reasonableness of a sentence for abuse of discretion. Castro-Aguirre, 983 F.3d at 943. A below-Guidelines sentence is presumptively reasonable. Id. at 944. “Indeed , we have never ‘deemed a below-range sentence to be unreasonably high.’” Id. (quoting United States v. Brown, 932 F.3d 1011, 1019 (7th Cir. 2019)).

The district court did not abuse its discretion in sentencing Harris to 262 months’ imprisonment—a sentence that fell nearly 100 months below the low end of the advisory Guidelines range of 360 months to life. As Harris points out, the district court had to consider, among other sentencing factors, “the need to avoid unwarranted sentence disparities among defendants with similar records who have been found guilty of similar conduct.” 18 U.S.C. § 3553(a)(6). And it did so. At Harris’s sentencing, the court expressly considered the need to avoid an unwarranted sentencing disparity between Harris and Gibson, who had similar roles in the conspiracy and identical Guidelines ranges. The court explained, however, why a sentencing disparity between Gibson and Harris was not, in fact, “unwarranted.” At 47, Gibson was nine years older than Harris. Gibson was also serving a 12-year consecutive sentence for a § 924(c) conviction in another jurisdiction. Those factors influenced Gibson’s likelihood of reoffending upon release , which is another sentencing factor. See § 3553(a)(2)(C); see also Dean v. United States, 137 S. Ct. 1170, 1176 (2017). The same recidivism calculus did not apply to Harris, who was younger and not facing a long consecutive sentence in another case.

28 Nos. 20-1236 & 20-2234

A district court has broad discretion to balance the § 3553(a) factors. United States v. Warner, 792 F.3d 847, 855 (7th Cir. 2015). The court’s reasoned decision to give Harris 22 months more than Gibson was well within the range of reasonable sentences.

III. Conclusion

For these reasons, we affirm the defendants’ convictions and sentences.