United States v. Rollerson, No. 20-2258 (7th Cir. 2021)
Annotate this Case
Justia Opinion Summary
Want to stay in the know about new opinions from the Seventh Circuit US Court of Appeals?
Sign up for free summaries delivered directly to your inbox. Learn More ›
You already receive new opinion summaries from Seventh Circuit US Court of Appeals. Did you know we offer summary newsletters for even more practice areas and jurisdictions? Explore them here.
Officers used a confidential informant to arrange four controlled drug buys from Rollerson. Officers obtained search warrants for an apartment and for Rollerson’s home, then stopped Rollerson for speeding. They recovered a gun and marijuana from his car. Rollerson, a convicted felon, said that he “was the only one who had something to do with the drug sales.” In Rollerson's home, police found $150,000 in cash that Rollerson admitted were drug proceeds. Rollerson had a key to the apartment, which contained four kilograms of fentanyl, 52 grams of heroin, 97 grams of cocaine, and 236 grams of tramadol, plus digital scales, firearms, and mail addressed to Rollerson.
Rollerson was convicted on certain drug and firearm charges but acquitted on other drug charges. The Seventh Circuit affirmed his 276-month sentence. Rollerson claimed that the prosecution did not present sufficiently reliable information that he sold heroin and fentanyl during four controlled drug buys for which he was not charged and that those uncharged controlled buys and other drugs for which he was acquitted were not “part of the same course of conduct … scheme or plan” as his offenses of convictions. The record at sentencing on the controlled buys was sparse but absent contradictory evidence, a police officer’s affidavit attesting that the buys occurred provided the “modicum of reliability” needed to find by a preponderance of the evidence that Rollerson committed those additional crimes.
Download PDF
In the United States Court of Appeals For the Seventh Circuit ____________________ No. 20-2258 UNITED STATES OF AMERICA, Plaintiff-Appellee, v. KEENAN ROLLERSON, Defendant-Appellant. ____________________ Appeal from the United States District Court for the Southern District of Indiana, Indianapolis Division. No. 1:17-cr-00101-JPH-DML-1 — James P. Hanlon, Judge. ____________________ ARGUED MAY 26, 2021 — DECIDED JULY 30, 2021 ____________________ Before EASTERBROOK, ROVNER, and HAMILTON, Circuit Judges. HAMILTON, Circuit Judge. A jury convicted defendant Keenan Rollerson on drug and rearm charges but acquitted him on other drug charges. He appeals only his sentence, arguing that the district court erred by increasing his Sentencing Guideline range based on drug activity for which he was either acquitted or never charged. Speci cally, Rollerson claims that the prosecution did not present su ciently reliable 2 No. 20-2258 information that he sold heroin and fentanyl to an informant during four controlled drug buys for which he was not charged. He also asserts that those uncharged controlled buys and other drugs for which he was acquitted were not “part of the same course of conduct … scheme or plan” as his o enses of conviction. U.S.S.G. § 1B1.3(a)(2). We a rm Rollerson’s sentence because the conduct at issue was supported by suf ciently reliable information and was relevant to his convictions. To be sure, the record at sentencing on the controlled buys was sparse. But at least in the absence of contradictory evidence, a police o cer’s a davit attesting that the buys actually occurred provided the “modicum of reliability” that is needed to nd by a preponderance of the evidence that Rollerson committed those additional crimes. See United States v. Helding, 948 F.3d 864, 871 (7th Cir. 2020). I. Factual and Procedural Background In July 2016, the Drug Enforcement Administration (DEA) began investigating Rollerson’s drug dealing activities in Indianapolis. The following spring, the DEA and local law enforcement used a con dential informant to arrange controlled drug buys from Rollerson. The government asserts that on four occasions, Rollerson sold drugs to the informant at an Indianapolis apartment that he controlled: twenty- ve grams of heroin on March 31, twenty- ve grams of heroin on April 3, twenty-four grams of fentanyl on April 17, and thirty grams of fentanyl on April 24. The police used these controlled buys to secure search warrants for the apartment and for Rollerson’s home. On April 27, 2017, after setting up surveillance at both addresses, the police stopped Rollerson for speeding. They recovered a gun and marijuana from his car. Rollerson admitted No. 20-2258 3 the gun and marijuana were his. He also acknowledged that he was a convicted felon. When told that police were about to search his residences, Rollerson said he would cooperate and that, although his son was home, Rollerson himself “was the only one who had something to do with the drug sales.” Police then took Rollerson to his home, where they found over $150,000 in cash that Rollerson admitted were drug proceeds. Rollerson was also in possession of a key to the stash house apartment. He told police that they would nd multiple kilograms of heroin hidden there. The search of the apartment actually uncovered four kilograms of fentanyl, fty-two grams of heroin, ninety-seven grams of cocaine, and two hundred thirty-six grams of tramadol, as well as digital scales, multiple rearms, and mail addressed to Rollerson at that address. On May 2, 2017, the government led a criminal complaint against Rollerson based on an a davit by DEA Task Force O cer Marc Campbell. In the a davit, O cer Campbell described the course of the investigation and the contraband found at Rollerson’s home and apartment. The a davit also included a brief description of the controlled buys, attesting that “Between March 2017 and April 2017, DEA/IMPD utilized an IMPD Con dential Source (CS) to conduct multiple controlled purchases of heroin from ROLLERSON in Indianapolis, Indiana. Each of these controlled purchases resulted in the seizure of heroin and fentanyl.” These controlled buys, however, were not included in the government’s charges— which focused instead on the drugs and guns uncovered at the stash house apartment. A grand jury indicted Rollerson on eight charges: Possession with Intent to Distribute Fentanyl (Count 1), Heroin 4 No. 20-2258 (Count 2), Cocaine (Count 3), and Tramadol (Count 4), as well as Unlawful Possession of Firearms by a Convicted Felon (Counts 5–8). The case went to trial, where the jury convicted Rollerson on Counts 2 and 5–8 (heroin and rearm o enses) but acquitted him on Counts 1, 3, and 4 (the fentanyl, cocaine, and tramadol). The Presentence Investigation Report (PSR) recommended a Sentencing Guideline range of 262 to 327 months for Count 2. This range was based upon a quantity of drugs that included not only the heroin in the o ense of conviction but also the fentanyl, cocaine, and tramadol from the acquitted counts and the four controlled buys used to obtain the search warrants. (The di erent types and amounts of drugs were all converted to a total “Converted Drug Weight” of 10,445.75 kilograms using the drug conversion tables in § 2D1.1, Note 8.) At sentencing, Rollerson objected to the PSR’s drug quantity, arguing (1) that the record contained no reliable information supporting the uncharged controlled-buy amounts, and (2) that neither the controlled buys nor the acquitted fentanyl, cocaine, and tramadol were relevant to his conviction for heroin. Without these uncharged and acquitted drug quantities, Rollerson’s guideline range for Count 2 would have been a much lower 110 to 137 months. The district court overruled Rollerson’s objections and explained its reasons for including both the uncharged and acquitted drug amounts in calculating the guideline range. As to the controlled buys, Judge Hanlon said: [I]n Paragraph 12 of the presentence report there is some detailed information about the No. 20-2258 5 controlled purchases. It lays out that there were four controlled purchases involving Fentanyl and heroin. It lays out the quantities involved. We also have the a davit that was led in support of the criminal complaint, which also discusses the controlled purchases. On that basis, the court concluded “that the information … relating to controlled buys is not unsupported or naked allegations” but rather “reliable” and “established by a preponderance of the evidence.” The judge added that “there isn’t any evidence that I have been made aware of from [Rollerson’s] counsel that would cause me to call into question the reliability of what’s contained in the report.” The judge then explained that these controlled buys also constituted “relevant conduct” under U.S.S.G. § 1B1.3(a)(2) because (1) they involved heroin, the same controlled substance as the o ense of conviction; (2) they occurred at the same stash house as the convicted o ense; and (3) they occurred within one month of the o ense of conviction. As for the fentanyl, cocaine, and tramadol for which the jury acquitted Rollerson, the judge explained that the trial evidence connecting Rollerson to the stash house—including but not limited to the mail addressed to him and the key in his possession—established his possession of those drugs by a preponderance of the evidence. The judge also found that this acquitted conduct was still relevant conduct for sentencing because it was all part of the same course of drug tra cking at the stash house. This was especially true, the judge said, because two of the controlled buys involved fentanyl, one of the substances found at the same stash house soon after that. 6 No. 20-2258 Accordingly, the judge sentenced Rollerson within the recommended guideline range: 276 months for Count 2, as well as 120 months for each gun charge (Counts 5–8), all to run concurrently. On appeal, Rollerson renews the arguments he raised at sentencing, that the district court erred by including both the uncharged and acquitted drug amounts in his guideline calculation. II. Reliability “A criminal defendant has a due process right to be sentenced based on accurate information…. [W]here the district court sentences a defendant based on the drug-quantity guidelines, it must nd the government’s information su ciently reliable to determine drug quantity by a preponderance of the evidence.” Helding, 948 F.3d at 870, citing United States v. Tucker, 404 U.S. 443, 447 (1972), and United States v. Lister, 432 F.3d 754, 762 (7th Cir. 2005). Rollerson claims that the uncharged and acquitted drug amounts should not count toward his guideline range because the prosecution failed to prove them with reliable information by a preponderance of the evidence. We review the substantive reasonableness of a sentence under an abuse-of-discretion standard, but we “must rst ensure that the district court committed no signi cant procedural error, such as … selecting a sentence based on clearly erroneous facts….” Gall v. United States, 552 U.S. 38, 51 (2007). We have thus said that “whether the district court followed the proper procedures in imposing sentence is a question of law that we review de novo.” United States v. Young, 863 F.3d 685, 688 (7th Cir. 2017), citing United States v. Mendoza, 510 F.3d 749, 754 (7th Cir. 2007). Here, though, the “signi cant procedural error” that Rollerson points to is the enhancement No. 20-2258 7 of his sentence “based on clearly erroneous facts.” Gall, 552 U.S. at 51 (emphasis added). So ultimately we ask whether the district judge clearly erred in nding that the government proved Rollerson’s conduct by a preponderance of the evidence. See United States v. Sandidge, 784 F.3d 1055, 1061–63 (7th Cir. 2015) (a rming sentence enhancement; district court “did not commit clear error” in nding “su ciently reliable” information proving defendant’s conduct by preponderance of evidence). “A sentencing court acts within its discretion when it credits con dential informants’ statements about drug quantity, but when a defendant objects, the evidence supporting that quantity must be found to be reliable.” Helding, 948 F.3d at 866. The “threshold for a su cient reliability nding” is “low.” Id. at 871. But if the PSR “asserts ‘nothing but a naked or unsupported charge,’ the defendant’s denial of that information su ces to cast doubt on its accuracy.” Id. at 870, quoting United States v. Marks, 864 F.3d 575, 580 (7th Cir. 2017). A truly bare allegation and bare denial would be in equipoise, unable to meet the prosecution’s burden of proof by a preponderance of the evidence. Once the prosecution presents su ciently reliable evidence, however, it will meet its burden unless the defense can muster evidence in the other direction. Here, the prosecution provided su ciently reliable evidence. As for Rollerson’s acquitted conduct, the evidence at trial amply supported a nding by a preponderance of the evidence that Rollerson possessed the fentanyl, cocaine, and tramadol found in the stash house. Rollerson had the key to the stash house. Mail was addressed to him there. An o cer testi ed to seeing him enter or exit the apartment on at least six occasions over the course of the investigation. To top those 8 No. 20-2258 o , when o cers told Rollerson that the apartment would be searched, he admitted that multiple kilograms of drugs were hidden there and that he—not his son—was responsible for them. 1 The uncharged drug buys present a di erent problem. The PSR described these four controlled buys in detail, saying that, on each occasion, Rollerson directed the informant to meet him at the stash house, drove there, and then sold the informant either heroin or fentanyl. The PSR stated the drug amount recovered in each buy. The PSR said that its source for all these details was the police a davit used to obtain the search warrant for the stash house. That a davit, however, was not attached to the PSR and was not o ered at sentencing by the government or the defense. It is not in the record at all, even though both sides admit it is the key piece of evidence bearing on the reliability of the uncharged drug amounts. As a result, the only statement of the drug amounts in our record came from the PSR’s summary of that police document that the district judge and we have not seen.2 The record establishing the reliability of the controlled buys was sparse, but it was su cient as a matter of due process, at least in the absence of con icting evidence. In addition to the PSR’s summary of the search warrant a davit, the judge also relied upon O cer Campbell’s a davit attached 1 The practice of considering acquitted conduct at sentencing is controversial but is clearly allowed if the conduct is proven by a preponderance of the evidence. United States v. Watts, 519 U.S. 148, 149 (1997); United States v. Waltower, 643 F.3d 572, 577 (7th Cir. 2011). 2 The absence of the search warrant affidavit presents something of a mystery. Both sides have it and both could easily have offered it at sentencing. For tactical reasons, however, both sides chose not to do so. No. 20-2258 9 to the criminal complaint. That a davit also attested to the occurrence of the controlled buys, albeit brie y: “Between March 2017 and April 2017, DEA/IMPD utilized an IMPD Con dential Source (CS) to conduct multiple controlled purchases of heroin from ROLLERSON in Indianapolis, Indiana. Each of these controlled purchases resulted in the seizure of heroin and fentanyl.” This police a davit added a “modicum of reliability” to the PSR’s description of the controlled-buy amounts. Helding, 948 F.3d at 871. It was sworn under penalty of perjury, a process meant to “impress upon the a ant ‘the solemnity and importance of his or her words and of the promise to be truthful, in moral, religious, or legal terms.’” 2 Wayne R. LaFave, Search & Seizure § 4.3(e) (6th ed.), quoting State v. GutierrezPerez, 337 P.3d 205, 210 (Utah 2014). Indeed, the Fourth Amendment’s express requirement that warrants may issue only if “supported by Oath or a rmation” re ects the idea that sworn a davits bear an added signal of reliability. Accordingly, the district judge’s explicit reliance on O cer Campbell’s a davit submitted with the criminal complaint was enough to establish the reliability of the PSR’s allegations. See United States v. Smith, 280 F.3d 807, 810–11 (7th Cir. 2002) (rejecting reliability challenge to gun enhancement where detective testi ed that, after a controlled drug buy, the informant told the detective that Smith pointed a gun at the informant). For purposes of argument, we assume that, absent O cer Campbell’s a davit, the prosecution might well have needed to come forward with the search warrant a davit if it wanted to rely on the four buys. Although the PSR contained speci c allegations, “speci city alone … does not make information 10 No. 20-2258 reliable.” Helding, 948 F.3d at 869. We say might, however, because there are important di erences between the PSR allegations here and those we held insu cient in Helding. In Helding, the defendant’s drug calculation skyrocketed based on a con dential informant’s unsubstantiated allegations that the defendant sold him methamphetamine a year before the actual police investigation that led to the defendant’s charges. Unlike this case, Helding “involved no controlled buys with any CI….” Id. at 868. The informant’s alleged purchase of methamphetamine in Helding was not initiated and observed by police—it was merely a story about Helding’s past activity that was relayed to police. Moreover, in Helding, “The district court saw no a davits … corroborating” the informant’s allegations. Id. at 871. “More to it, nowhere did the PSR contain any information—even a representation by law enforcement—that the informants’ statements were known to be reliable.” Id. at 869. The uncharged drug amounts in this case were not based on uncorroborated allegations by an informant whose trustworthiness was unknown. Rather, the PSR’s description came from police documents supplied to the probation o ce recounting the o cers’ roles in setting up and observing these controlled buys. Language in Helding suggested that this might be su cient: “It may be enough for the government to supply the probation o ce, and, in turn, for the PSR to include, some statement bearing on the reliability of information provided by a con dential source.” Id. at 872. We need not decide that, however, given the inclusion of O cer Campbell’s a davit in the sentencing record. At least in the absence of con icting evidence, the PSR’s assertions concerning the controlled-buy amounts were No. 20-2258 11 su ciently reliable to support a nding that they were proven by a preponderance of the evidence. In due process terms, the defense had the opportunity to present con icting evidence about the controlled buys. The defense also had the opportunity to present the search warrant a davit or to call O cer Campbell to testify. The defense chose not to exercise those options, which left the government’s sparse evidence unrebutted. We see no indication that the defense was denied a fair hearing on these subjects. Still, it is worth repeating our earlier advice: “While it’s not required that a judge hear personally from witnesses under oath at a sentencing hearing about drug quantities, we think it’s not a terribly bad idea to do so when the witness is going to provide the basis for … a defendant’s relevant conduct.” Helding, 948 F.3d at 871, quoting United States v. Robinson, 164 F.3d 1068, 1070 (7th Cir. 1999). III. Relevant Conduct Even if proven by a preponderance of the evidence, Rollerson’s uncharged and acquitted conduct must still be “relevant” to his o enses of conviction to be used in his guideline calculation. U.S.S.G. § 1B1.3(a)(2). Rollerson asserts that neither the controlled buys nor the fentanyl, cocaine, and tramadol found at the stash house were “part of the same course of conduct or common scheme” as his conviction for possessing heroin with intent to distribute. See id. We disagree. The district court did not clearly err in nding that Rollerson’s uncharged and acquitted conduct was relevant to his o ense of conviction. See United States v. Baines, 777 F.3d 959, 963 (7th Cir. 2015) (“Whether uncharged o enses amount to relevant conduct under the Sentencing Guidelines is a factual determination, which we review for clear error.”) (citation omitted). 12 No. 20-2258 The fact that a defendant engaged in other uncharged or acquitted drug transactions “is not su cient to justify treating those transactions as ‘relevant conduct’ for sentencing purposes.” United States v. Ortiz, 431 F.3d 1035, 1041 (7th Cir. 2005), quoting United States v. Crockett, 82 F.3d 722, 730 (7th Cir. 1996). In assessing relevant conduct, we look for “a strong relationship between the uncharged conduct and the convicted o ense, focusing on whether the government has demonstrated a signi cant ‘similarity, regularity and temporal proximity.’” United States v. McGowan, 478 F.3d 800, 802 (7th Cir. 2007), quoting Ortiz, 431 F.3d at 1040. These factors are derived from commentary in the Sentencing Guidelines explaining that, under U.S.S.G. § 1B1.3, conduct is part of a “common scheme or plan” if it is “substantially connected” to a convicted o ense “by at least one common factor, such as common victims, common accomplices, common purpose, or similar modus operandi.” § 1B1.3, comment 5(B)(i). And, separately, activity is “part of the same course of conduct” as a convicted o ense if it is “part of a single episode, spree, or ongoing series of o enses.” Id. at 5(B)(ii). What we know about Rollerson’s uncharged drug buys showed similarity, regularity, and temporal proximity to his o ense of conviction. Two of the buys involved heroin, the same drug for which Rollerson was convicted. The other two involved fentanyl, a drug that was also found in large quantities at the stash house. The buys took place at that same stash house. And they occurred regularly through the weeks leading to the search of the stash house in April 2017. These facts support the district court’s nding that the uncharged buys constituted relevant conduct. No. 20-2258 13 The same goes for the acquitted possession of fentanyl, cocaine, and tramadol. Rollerson sold fentanyl during two of the controlled buys at the stash house, and the police soon after those buys found a large amount of it, four kilograms. The fentanyl, cocaine, and tramadol were also found alongside the heroin for which Rollerson was convicted. Moreover, Rollerson’s controlled buys involved two separate drugs, further supporting the district court’s nding that the cocaine and tramadol were likely part of the same drug-dealing scheme as the heroin, even though those substances were not involved in the controlled buys. Rollerson says his case is like United States v. Ortiz, 431 F.3d 1035 (7th Cir. 2005), and United States v. Draheim, 958 F.3d 651 (7th Cir. 2020), where we rejected relevant conduct ndings. But those cases are not at all like Rollerson’s. In Ortiz, we found that a defendant’s alleged purchases of large amounts of cocaine across three states from 1997 to 1999 were not part of the same course of conduct or scheme as his convictions for selling much smaller amounts of marijuana and cocaine to a DEA informant in 2000 and 2001. 431 F.3d at 1041–42. We explained that the prosecution had failed to show temporal proximity (the alleged conduct occurred at least ten months before the o enses of conviction), regularity (the alleged conduct’s frequency di ered from that of the sales leading to Ortiz’s convictions), or similarity (the alleged conduct occurred in a di erent location). Id. Rollerson’s uncharged and acquitted drug activity, on the other hand, all occurred at the same location during the same month. Draheim is also readily distinguishable. That defendant’s lone “sale of two grams of street meth in a city” simply did not “match up” with a “collaborative bulk order from the other side of the nation for nearly fty grams of pure [meth].” 958 F.3d at 659–60 (emphases 14 No. 20-2258 added). So the conduct in Draheim occurred in a di erent location, involved much more drugs, and included other traf ckers. Because Rollerson’s uncharged buys and acquitted drug amounts were relevant to his heroin conviction and proven with su ciently reliable information, Rollerson’s sentence is AFFIRMED.
Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You should read the full case before relying on it for legal research purposes.
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
