No. 24-3729

Appeal from the United States District Court for the Northern District of Ohio at Cleveland. No. 1:22-cr-00679-2—Pamela A. Barker, District Judge.

Argued: June 12, 2025

Decided and Filed: September 4, 2025

Before: THAPAR, READLER, and BLOOMEKATZ, Circuit Judges.

ARGUED: Lori Beth Riga, OFFICE OF THE FEDERAL PUBLIC DEFENDER, Cleveland, Ohio, for Appellant. Colleen Egan, UNITED STATES ATTORNEY'S OFFICE, Cleveland, Ohio, for Appellee. ON BRIEF: Lori Beth Riga, OFFICE OF THE FEDERAL PUBLIC DEFENDER, Cleveland, Ohio, for Appellant. Colleen Egan, UNITED STATES ATTORNEY’S OFFICE, Cleveland, Ohio, for Appellee.

THAPAR, J., delivered the opinion of the court in which READLER, J., concurred and BLOOMEKATZ, J., concurred in the result. BLOOMEKATZ, J. (pp. 15-25), delivered a separate concurring opinion.

Page 2THAPAR, Circuit Judge. If federal agents observe someone leave a house, travel to an agreed-upon location, sell drugs to an undercover officer, and then immediately return to the house, is there a fair probability that the agents will find contraband at the house? And if the agents then obtain a search warrant for the house and find nearly $60,000 worth of drug proceeds, a money counter, firearms, a pill press and scales, and body armor (as well as the dealer himself), is it clear error for the district court to give the defendant a sentencing enhancement for maintaining a drug premises? Because the answer to the first question is “yes,” and the answer to the second question is “no,” we affirm.

Federal agents from the Bureau of Alcohol, Tobacco, Firearms and Explosives (“ATF”) began investigating Jayon Florence due to his connections to a known drug dealer. Using an undercover officer, the agents conducted a series of controlled purchases of fentanyl and cocaine from Florence during late 2022 and early 2023. After the third controlled purchase, agents obtained a federal search warrant to track Florence’s cell phone. So, while arranging a fourth and final purchase, agents monitored Florence’s cell-phone pings. Those pings placed Florence near a particular intersection in East Cleveland.

As agents drove around that area, the undercover officer called Florence to set up the fourth purchase. Florence said he needed to get dressed first. He added that the would-be purchaser (the undercover officer) should relax because “the s*** isn’t going anywhere.” R. 31, Pg. ID 113. As agents continued surveilling the area in which the cell-phone monitoring placed Florence, they then saw Florence's car (the same car in which Florence had arrived for a previous controlled buy) parked outside a residence on Melbourne Road.

As the agents surveilled the Melbourne Road residence, Florence texted the undercover officer to meet Florence at the same spot as two previous sales. The agents observed Florence exit the Melbourne Road residence, get in his car, and drive to the location. They tailed Page 3Florence. Florence then sold the drugs to the undercover officer in Florence’s car. Followed by the agents, Florence drove back to the Melbourne Road residence, parked, and entered the house.

Four days later, a task force officer with the ATF relayed this information to a magistrate judge, who issued a search warrant for the Melbourne Road residence.

Two days after receiving the warrant, the agents searched the Melbourne Road residence. Florence and his aunt were there. Law enforcement recovered nine firearms, over 300 rounds of ammunition, body armor, a money counter, digital scales with residue, a pill press with powder residue, over thirty grams of a fentanyl mixture, four pounds of suspected marijuana, and $59,507 in cash.

A grand jury indicted Florence on nine counts of various drug and firearm offenses. Florence moved to suppress the evidence found from the search. After the district court denied his suppression motion, Florence pled guilty to four counts related to the distribution of drugs and illegal possession of firearms. In his plea agreement, Florence admitted to possessing all the firearms and drugs, but he preserved his right to appeal the district court's denial of his suppression motion.

At sentencing, the district court applied a two-level enhancement to Florence’s sentence for maintaining a premises for the manufacture or distribution of a controlled substance. See U.S.S.G. § 2D1.1(b)(12). The court sentenced Florence to 106 months in prison.

On appeal, Florence challenges the district court’s denial of his suppression motion and its application of the drug-premises enhancement.

Florence’s appeal of the denial of his suppression motion fails. The search warrant was supported by probable cause.

Probable cause is “not a difficult standard to meet.” United States v. Whitlow , 134 F.4th 914, 919 (6th Cir. 2025). The critical question is whether there was a “fair probability” that the officers would find evidence of criminal wrongdoing in the location to be searched. Illinois v. Gates , 462 U.S. 213, 238 (1983). And the warrant affidavit had to establish “a nexus between Page 4the place to be searched and the evidence sought.” United States v. Ellison , 632 F.3d 347, 349 (6th Cir. 2011). At bottom, this court's job 'is simply to ensure that the magistrate had a substantial basis for concluding that probable cause existed” when he issued the warrant. Gates , 462 U.S. at 238–39 (cleaned up).

Here, the probable-cause analysis is straightforward: This court’s recent en banc decision in United States v. Sanders squarely forecloses Florence’s challenge. 106 F.4th 455 (6th Cir. 2024) (en banc). In that case, officers had received a tip from an informant that Sanders was dealing drugs. Id. at 459. The police then set up two controlled buys from Sanders. Id. On the first buy, the officers surveilled the transaction and then followed Sanders back to an apartment. Id. On the second buy, the officers observed Sanders leave that same apartment, travel elsewhere to complete the transaction, and then return to the apartment. Id. at 460. Officers relied on this evidence to obtain a search warrant for the apartment. Id. Sanders moved to suppress the evidence found during the search. Id. The district court denied his motion, and the en banc court affirmed. Id. at 460, 467, 476.

In so doing, the en banc court stated that there is probable cause when officers see somebody 'leave a location' and engage 'in a sale involving . . . contraband.' Id. at 462. During the second controlled buy, officers observed Sanders leave the apartment, get in his car, sell drugs, and then return to the apartment. That alone “plainly” sufficed for probable cause. Id. at 463 (“This evidence alone would end the matter.”).¹ To be sure, the court then noted the Page 5first controlled buy and the initial tip as well. Id. at 463-64. But the second controlled buy alone—namely, the surveillance of Sanders leaving the apartment, completing the second drug transaction, and returning to the apartment—sufficed. Id. at 463; see also id. at 464 ('[T]he tip was not a vital component of the probable cause equation.'). Further, the probable-cause determination did not hinge on whether Sanders resided at the apartment. See id. at 466 (“Nothing in the affidavit suggests that Sanders lived at the apartment.”).

Just like in Sanders , law enforcement observed Florence leave a residence, complete a drug transaction, and then return to that residence. And just like in Sanders , the agents' observations established a fair probability that Florence was storing illegal drugs in the residence or, at the very least, 'evidence of drug trafficking,' including from the controlled buy that officers had just conducted. Id. at 463–64.

Florence fails to distinguish Sanders . First, Florence is wrong when he implies the affidavit here must be identical to the one at issue in Sanders . See id. at 462 (recognizing there is “no model fact pattern for establishing a fair probability that contraband will be found”). After all, Sanders was not a particularly close call on the probable cause question. The en banc court held that the second controlled buy 'plainly demonstrate[d]' probable cause to search the residence: The additional evidence referenced in the affidavit only 'add[ed] to what [was] already obvious.” Id. at 463–64. So even assuming the warrant affidavit here lacks what the one in Sanders contained, that says nothing about whether the search warrant here lacked probable cause. Put another way, the facts that established probable cause in Sanders don’t set a floor for establishing probable cause.

In any event, the specific distinctions Florence draws between this case and Sanders are irrelevant. He contends that there was a tighter nexus between Sanders and the apartment in that case because of the informant's tip plus the two controlled buys connected to the apartment. Here, by contrast, there was no tip, and only one controlled buy. But not all factual differences create legal distinctions. Again, the tip in Sanders “was not a vital component of the probable cause equation.' Id. at 464. And the officers' observation of Sanders leaving his residence, completing a single drug transaction, and returning to a residence was enough for the Sanders Page 6court to find probable cause to search that residence. Id. at 463. So, Sanders covers Florence’s conduct.

If anything, the evidence here is stronger than in Sanders. There, the probable cause determination hinged on circumstantial evidence that a person leaving a location, engaging in a drug deal, and immediately returning to the same location allowed a reasonable inference that evidence of contraband would be found at that place. Id. at 466 (finding direct evidence of drug dealing in the location to be searched unnecessary and holding that “circumstantial evidence that contraband might be found at the place to be searched suffices'). That same coming-andreturning evidence was present here, but there were additional indicators that contraband would be found in the apartment. Recall that Florence, likely while in the Melbourne Road residence, spoke to an undercover officer on the phone shortly before the buy. Florence told the officer that he would be ready soon and assured him that “the s*** isn’t going anywhere.” R. 31, Pg. ID 113. Then, after agents had located his car in the driveway, Florence texted the officer that he would be “ready in a minute” and instructed him to meet at the spot of two prior sales. Id. As compared to Sanders , Florence vouching for his control over the drugs while likely in the residence provides further indication that there was contraband in that residence. Although not legally necessary to establish probable cause given agents’ observation of Florence coming and going from the residence for the controlled buy, this evidence bolsters the conclusion that it was fairly probable that officers would find drugs there.²

Finally, Florence's reliance on United States v. McPhearson , 469 F.3d 518 (6th Cir. 2006), is misplaced. There, police had a warrant for McPhearson’s arrest on assault charges. Id. at 520. They arrested McPhearson outside his home and found cocaine on his person during a search incident to the arrest. Id. The police obtained a warrant to search McPhearson's home based on that cocaine, relying on the “known drug dealer inference.” Id. at 521. The district court granted McPhearson’s motion to suppress because there was no “evidence presented” to Page 7the issuing judge “that there were drugs in the house.” Id. at 522. This court affirmed. The affidavit 'did no more than state' that the defendant who resided at the house to be searched “was arrested for a non-drug offense with a quantity of crack cocaine on his person.” Id. at 524. Critically, the affidavit did not establish that McPhearson was a 'known drug dealer,' as 'his prior convictions were for property crimes.” Id. at 525. And probable cause could not otherwise be established because a suspect’s “mere presence” outside a residence doesn’t create a sufficient “nexus between the place to be searched and the evidence to be sought.” Id. (citation omitted). In short, there were no “facts suggesting that [the defendant’s] home would contain evidence of crime.' Id.

Probable cause based on controlled buys, in contrast, does not depend on the 'known drug dealer” inference. Rather, law enforcement had sufficient proof that the Melbourne Road residence would “contain evidence of crime” (namely, illegal drugs). Id. Florence exited that residence before getting in his car and driving to complete a drug transaction. See Sanders , 106 F.4th at 463; id. at 466 (distinguishing McPhearson when probable cause depends on a controlled buy and not the known drug dealer inference). Of course, maybe Florence had been storing the drugs in the car alone, but that possibility doesn’t diminish the “fair probability” that Florence stored drugs in the Melbourne Road residence. Id. at 466–67. Thus, the search was supported by probable cause.

Florence next challenges the district court's application of a two-level sentencing enhancement for maintaining a drug premises. See U.S.S.G. § 2D1.1(b)(12).

The government bears the burden of proving the applicability of a particular sentencing enhancement by a preponderance of the evidence. United States v. Stubblefield , 682 F.3d 502, 510 (6th Cir. 2012). Section 2D1.1(b)(12) provides for a two-level enhancement '[i]f the defendant maintained a premises for the purpose of manufacturing or distributing a controlled substance.” So this enhancement applies to “anyone who (1) knowingly (2) opens or maintains any place (3) for the purpose of manufacturing or distributing a controlled substance.” United States v. Johnson , 737 F.3d 444, 447 (6th Cir. 2013). There’s no dispute that the first and third Page 8prongs apply to Florence. The only question is whether the second prong (the maintenance prong) does. When the defendant doesn't formally own the premises, the touchstone for the maintenance prong is de facto (actual) control. United States v. Taylor , 85 F.4th 386, 390 (6th Cir. 2023).

The district court concluded that the second prong applied. The court found sufficient evidence of control largely based on the sheer quantity of guns, drugs, and drug paraphernalia that Florence stored at the Melbourne Road residence. The court first reasoned that those drugs and guns “show[ed]” that Florence enjoyed “access” to the residence. R. 76, Pg. ID 556. And since Florence could come and go from the residence and bring such items into it, he must have “had some control over the[] premises.” Id. In the end, the court found it 'a close call' but overruled Florence’s objection to the enhancement’s application. Id.

When the district court finds the matter a 'close call,' the standard of review often matters. United States v. Uminn , 820 F. App’x 353, 356 (6th Cir. 2020).

This court applies clear-error review to the district court’s findings of fact and de novo review to the district court’s interpretation of the Guidelines. United States v. Abdalla , 972 F.3d 838, 850 (6th Cir. 2020). But the appropriate standard of review for the district court's application of the maintenance prong to a particular set of facts is murky. Compare United States v. Broadnax , 777 F. App'x 137, 141 (6th Cir. 2019) (applying clear-error review), with United States v. Hernandez , 721 F. App’x 479, 482 (6th Cir. 2018) (applying de novo review).

The right answer hinges on which “judicial actor is better positioned” to make the call. Miller v. Fenton , 474 U.S. 104, 114 (1985). Under Supreme Court precedent, district courts are best positioned to make fact-bound calls in the sentencing context, so deferential, clear-error review is the appropriate standard.

1.

Start with the Court’s unanimous opinion in Buford v. United States , 532 U.S. 59 (2001). The Buford Court assessed the appropriate standard of review for a district court’s determination Page 9that a defendant’s prior felony convictions were “related” to one another—and thus deserved to be treated as a single felony for purposes of the “career offender” provision. 532 U.S. at 60–61. An application note to the Guidelines at the time stated that prior convictions were “related” to each other when, among other things, they “were consolidated . . . for sentencing.” Id. at 61 (citation omitted). Relevant circuit precedent, in turn, provided that the convictions could be considered “consolidated” even if they had not been formally consolidated for sentencing. Id. The question thus became whether the defendant’s convictions were “functionally consolidated”—in other words, whether they were “factually or logically related.” Id. (citation omitted).

Buford held that the district court was best positioned to make this determination, so a deferential standard of review was appropriate. Id. at 64. To support this conclusion, the Court pointed to many factors. For example, district courts are “more familiar with trial and sentencing practices in general, including consolidation procedures.” Id. 'In addition, factual nuance may closely guide the legal decision, with legal results depending heavily upon an understanding of the significance of case-specific details.” Id. at 65. That fact-bound decision also “limit[ed] the value of appellate court precedent.' Id. at 65-66. The Court further pointed out that the 'functional consolidation' question boiled down to interpreting 'a minor, detailed, interstitial question of sentencing law'-that is, a guideline's application note. Id. at 65. The Court contrasted that question with a 'purely legal matter' like interpreting the words of a given guideline. Id.

Those same considerations apply here. In particular, “factual nuance” is front and center when determining whether the defendant controlled the premises in question. Id. As a result, whether the enhancement applies “depend[s] heavily upon an understanding of the significance of case-specific details.' Id. Indeed, the guideline's application note itself calls on courts to assess fact-intensive questions like “the extent to which the defendant controlled access to, or activities at, the premises.' U.S.S.G. § 2D1.1(b)(12) cmt. n.17. And that involves weighing many facts, including “duration, acquisition of the site, renting or furnishing the site, supervising, protecting, supplying food to those at the site, and maintaining continuity.' Broadnax , 777 F. App’x at 141.

Page 10The analysis in a more recent Supreme Court decision likewise cuts in favor of deference to the district court. See U.S. Bank Nat’l Ass’n ex rel. CWCapital Asset Mgmt. LLC v. Vill. at Lakeridge, LLC , 583 U.S. 387, 396 (2018). The Court noted that 'appellate courts should usually review a [lower court's] decision with deference' when a particular mixed question 'immerse[s] courts in case-specific factual issues.' Id. Appellate courts aren't in a good position to second-guess a district court's weighing of 'multifarious, fleeting, special, narrow facts that utterly resist generalization.” Id. (citation omitted).

That logic applies with full force here. Determining the applicability of the drugpremises enhancement entails “tak[ing] a raft of case-specific historical facts, consider[ing] them as a whole,' and 'balanc[ing] them one against another' to figure out the extent of the defendant’s control over the premises in question. Id. at 397. Under Supreme Court precedent, that is a fact-bound determination that warrants clear-error review. Id. at 399.

And that’s all true even when the precise facts are undisputed. Buford expressly rejected the notion that a lack of factual dispute deprives its arguments about the institutional advantages of district courts of their force. The losing party in Buford argued for de novo review on the grounds that 'applying a Sentencing Guidelines term to undisputed facts . . . demands no deference at all.' Buford , 532 U.S. at 64 (emphasis added). The Court disagreed. Id. at 64–66.³

In sum, the Supreme Court has instructed us “to review fact-bound mixed questions of law and fact for clear error.” United States v. O’Lear , 90 F.4th 519, 536 (6th Cir. 2024). The application of the maintenance prong of the drug-premises enhancement to a particular set of facts is such a fact-bound question.⁴

Further, Sixth Circuit precedent on the proper standard of review for a different prong of the drug-premises enhancement—the primary use prong—doesn’t stand in the way of applying clear-error review to the maintenance prong.

In United States v. Tripplet , this court applied de novo review to the district court's application of the primary use prong of the drug-premises enhancement. 112 F.4th 428, 432 (6th Cir. 2024). The court didn’t reach the maintenance prong because the defendant didn’t challenge its application. Id. at 430, 432 n.1. Because Tripplet grappled only with the primary use prong, it didn’t hold—or even say—anything about the proper standard for reviewing application of the maintenance prong. See Wright v. Spaulding , 939 F.3d 695, 701-02 (6th Cir. 2019). And the standards of review for the two prongs need not be the same. The appropriate standard of review hinges on “the nature of the question presented,” like how a particular “Guidelines term ” applies to a set of facts. Buford , 532 U.S. at 63-64 (emphasis added). Indeed, this court has applied different standards of review to different provisions of a single guideline before. See, e.g. , United States v. Brown , 131 F.4th 337, 343–44 (6th Cir. 2025) (applying different standards of review to two prongs of a guideline involving possession of a firearm during an offense). Thus, Tripplet ’s conclusion about the standard of review on one prong doesn’t dictate the standard of review on another.

B.

Applying clear-error review here makes this an easy case. To be clearly erroneous, the district court’s decision must “strike us as wrong with the force of a five-week-old, unrefrigerated dead fish.' Taglieri v. Monasky , 907 F.3d 404, 409 (6th Cir. 2018) (en banc) (citation omitted), aff’d , 589 U.S. 68 (2020). The district court's decision here is no dead fish.

To be sure, the district court was correct that this was a close case on the merits. On the one hand, it strains credulity to believe that Florence lacked control over a residence from which he freely came and went and in which he stored tens of thousands of dollars in cash, sizable amounts of drugs, drug paraphernalia, firearms, and body armor. Most people wouldn’t leave their Porsche in a garage that they had no control over for weeks at a time. Nor would they leave Page 12their valuable jewelry in a safe without knowing the code. The same holds true for the drugs, cash, and other items at issue here.

On the other hand, the fact that Florence’s father owned the residence makes this case more difficult. Familial trust might have served as an effective substitute for actual control: Florence might have been able to freely access and store things at his father’s home by virtue of familial ties. That doesn't necessarily mean that Florence exercised de facto control over his father’s home.⁵

But this isn’t the first time we have seen a case like this. See United States v. Bennett , No. 22-5142, 2024 WL 966367 (6th Cir. Mar. 6, 2024). In Bennett , law enforcement searched a restaurant from which the defendant sold drugs, his girlfriend’s residence, and his home. Id. at *1. The police found 418.65 grams of heroin at his girlfriend’s house, but they didn’t find any drugs at the other locations. Id . The defendant didn’t have a legal or possessory interest in his girlfriend’s house. Id. at *5. But in determining that the enhancement applied, the Bennett panel emphasized that the government had proved by a preponderance of the evidence that the defendant possessed the heroin there, and that alongside the drugs, law enforcement found “additional indicia of drug trafficking at the house” like “a duffel bag with a press and mixer in it.” Id. at *6. Based on these considerations, the panel concluded that the defendant exercised de facto control over his girlfriend’s residence such that the enhancement could apply. Id.

Just like the defendant in Bennett , Florence didn’t have a legal or possessory interest in the Melbourne Road residence. But the drugs found there were his—he expressly admitted as much. Further, there were “additional indicia of drug trafficking” at the residence: a pill press, scales, firearms, body armor, a money counter, and almost sixty thousand dollars in cash. That drug paraphernalia provided circumstantial evidence that the defendant did in fact control the premises. And as Bennett makes clear, the district court didn’t clearly err in applying the drugpremises enhancement.

Page 13Of course, unpublished opinions like Bennett don't bind district courts. Fonseca v. Consol. Rail Corp. , 246 F.3d 585, 591 (6th Cir. 2001). But as noted above, the presence of 'additional indicia of drug trafficking' like in Bennett provided circumstantial evidence that Florence did in fact control the relevant premises. And reaching the same conclusion as a Sixth Circuit panel will rarely, if ever, amount to clear error on the district court’s part.

What’s more, even the precedent that cuts in Florence’s favor doesn’t establish clear error by the district court.

Consider, for example, this court's unpublished opinion in United States v. Whiteside , 747 F. App’x 387 (6th Cir. 2018). In that case, law enforcement had conducted three controlled buys from the defendant at a residence. Id. at 389. The defendant answered the door for the first and second buys; it is unclear whether he answered the door for the third. Id. at 389–90. After the third controlled buy, law enforcement executed a search warrant at the residence. Id. at 390. The defendant and three other people were present in the kitchen at the time of the search. Id. Drug paraphernalia and firearms were found in the kitchen, and a baggie of crack cocaine was found at the defendant's feet. Id. Eventually, the defendant was convicted of, among other things, distributing cocaine base, although he was acquitted of the charges related to the first controlled buy. Id. at 393.

But the government conceded, and the Whiteside panel agreed, that the government failed to prove that the defendant had exercised control over the residence. Id. at 394. Why? There was conflicting trial evidence regarding the defendant’s role in the first controlled buy. Id. Also, the defendant had been acquitted on the count related to the first controlled buy, and there was uncertain trial testimony about whether the defendant actually opened the door during another controlled buy. Id. Finally, several people were present at the residence when law enforcement executed the warrant. Id. Taken together, all these facts “call[ed] into question ‘the extent to which [the defendant] controlled access to, or activities at, the premises.’” Id. (quoting U.S.S.G. § 2D1.1 cmt. n.17). Thus, the panel concluded that the drug-premises enhancement didn’t apply. Id. at 395.

Page 14In reaching that conclusion, the Whiteside panel merely held that facts like the jury acquittal and unclear testimony “call[ed] into question” the extent of the defendant’s control over the premises. Id. at 394. But even if the facts of Whiteside are similar to the facts of Florence’s case, clear-error review requires more than just questions about the applicability of the enhancement. Indeed, even if the district court reached an arguably wrong conclusion, that wouldn’t be enough to vacate Florence’s sentence.

At the end of the day, Whiteside ’s reasoning supports a conclusion that a different district court could have reasonably come out differently than the district court did here. But the fact that a different district court could come out differently does not amount to clear error by this district court. Broadnax , 777 F. App'x at 141. And, as discussed, our court previously applied the drug-premises enhancement based on just as much evidence of the defendant’s control. In short, the district court didn’t clearly err.

* * *

We affirm.

Page 15BLOOMEKATZ, Circuit Judge, concurring in the judgment. I agree that we should affirm the district court on the two issues in this appeal. But I depart from the majority opinion’s reasoning as to both the motion to suppress and the sentencing enhancement. Therefore, I respectfully concur in the judgment and explain why I would decide this appeal on alternative grounds.

We can affirm the district court’s denial of Florence’s suppression motion because the officers relied in “good faith” on the warrant here. United States v. Leon , 468 U.S. 897, 920–22 (1984). So, unlike the majority opinion, I would not reach the more difficult question of whether the affidavit demonstrated probable cause for the search. United States v. Ardd , 911 F.3d 348, 351 (6th Cir. 2018).

Nexus Evidence. As the majority opinion articulates, Florence’s motion to suppress turns on the evidence linking his drug dealing activities to the Melbourne residence. See Maj. Op. at 4–5. The fact that officers knew that Florence was dealing drugs is insufficient; officers had to connect his illicit activities to the target of their search—the Melbourne residence. We call this evidence of a “nexus.” United States v. Reed , 993 F.3d 441, 447 (6th Cir. 2021).

The main nexus evidence here comes from police surveillance on a single day, related to a single drug sale. While driving around the neighborhood, police officers noticed a vehicle that Florence had used in previous drug deals parked in the driveway of the Melbourne residence. Shortly thereafter, they observed Florence leave the Melbourne residence, drive away in the vehicle, sell drugs to an undercover officer, and then return to the same house. I'll adopt the majority opinion's helpful shorthand and refer to this surveillance evidence as 'coming-andreturning evidence.” Maj. Op. at 6.

Page 16In addition to the coming-and-returning evidence, the officers connected Florence’s drug dealing activities to the residence through text messages. Just before the officers located the familiar vehicle in the driveway, Florence texted the undercover officer about their imminent drug transaction. He texted that he needed to get dressed before leaving and that the drugs '[weren't] going anywhere.' Warrant, R. 31, PageID 113. And after the officers began surveilling the vehicle but before they observed Florence leave the residence, Florence messaged the undercover officer the meeting location and said that he'd be 'ready in a minute.' Id. Although Florence texted about 'get[ting] dressed' while presumably in the house, id. , the officers did not know how much time Florence had spent at the Melbourne residence on that day or any other day. Indeed, police databases indicated that Florence had a different residential address.

The officers also relied on some cell phone location data from the week leading up to the sale to support their application for a warrant to search the Melbourne residence. But that data only placed Florence in a several-block radius that included the residence and did not correspond to any dates on which officers observed drug activity. So that data does not support a nexus, and the majority opinion likewise does not rely on it.

Together, then, the evidence of a nexus suggesting that officers would find contraband at the Melbourne residence rested on a single observation of Florence leaving and returning to the residence before and after completing a drug transaction, and text messages Florence sent while presumably inside the residence in which he “vouch[ed] for his control over the drugs” and said he needed to get dressed. Maj. Op. at 6.

Probable Cause . Was this sufficient nexus evidence for probable cause to search the house? To answer that question, we evaluate whether there was a “fair probability” that police would find evidence of Florence’s drug dealing in the Melbourne residence (not in the car, not on Florence, but in the house itself). Reed , 993 F.3d at 447. The probable cause inquiry is “factintensive' and eschews a 'neat set of legal rules' or 'precise definition.' United States v. Sanders , 106 F.4th 455, 461 (6th Cir. 2024) (en banc) (cleaned up). Instead, in evaluating whether there was a 'fair probability that the specific place that officers want[ed] to search Page 17w[ould] contain the specific things that they [we]re looking for,” Reed , 993 F.3d at 447, courts look at the “totality of the circumstances,” Florida v. Harris , 568 U.S. 237, 244 (2013).

The majority opinion treats the probable cause question here as an easy one under our caselaw, but I don’t think it is so “straightforward.” Maj. Op. at 4. As far as I’m aware, our court has never held there was probable cause to search a residence in a case that rested solely on evidence of just one occasion when the defendant left a residence to sell drugs then returned to it afterwards. To be sure, our cases have uniformly concluded that coming-and-returning evidence weighs in favor of probable cause. But in each of our cases affirming probable cause that has involved coming-and-returning evidence, there has been additional nexus evidence. In some cases, there was more than one drug transaction involving travel to or from the residence. See, e.g. , Sanders , 106 F.4th at 463; United States v. White , 990 F.3d 488, 489-90 (6th Cir. 2021); United States v. Burrell , 114 F.4th 537, 552 (6th Cir. 2024). In others, there was a tip from a confidential informant. See, e.g. , Sanders , 106 F.4th at 464; United States v. Jones , 817 F.3d 489, 491 (6th Cir. 2016). In yet others, the premises were the defendant’s own residence, which this court has recognized carries a strong inference that evidence of drug trafficking will be found there. See, e.g. , United States v. Coleman , 923 F.3d 450, 457 (6th Cir. 2019); White , 990 F.3d at 492. Or, even in cases where the residence in question was not the defendant’s, officers had observed drug transactions right outside the residence. See, e.g. , United States v. Ellison , 632 F.3d 347, 349-50 (6th Cir. 2011). In each case cited in the majority opinion or by the government affirming probable cause, there was more evidence of a nexus than surveillance of a single instance of the defendant going to and returning from a drug deal.

Here, as the majority opinion emphasizes, we do have additional indicia of a nexus to the Melbourne residence apart from the coming-and-returning evidence. We have the text messages. But is that enough? They do not place Florence at the house for very long. And, unlike a tip from a reliable confidential informant, the texts do not provide independent corroboration that there are drugs in the house. I think the probable cause inquiry is difficult here, as it is not clearly encompassed within the fact patterns from our existing precedent.

The majority opinion, by contrast, says that this case is “squarely” decided by the court’s en banc decision in Sanders . Maj. Op. at 4. It points to language in Sanders saying that coming- Page 18and-returning evidence “alone would end” the probable cause inquiry. Id. (quoting Sanders , 106 F.4th at 463). While Sanders included this broad language, its disposition did not rest on this single round-trip surveillance alone; instead, it declared, “there [wa]s more.” Sanders , 106 F.4th at 463. The court emphasized the myriad nexus evidence, including additional surveillance of the defendant returning to the house after another drug sale to an undercover officer, and a tip from a confidential informant. Id. at 463–64. Indeed, the opinion’s opening line highlighted the critical fact that a “confidential informant notified officers that [the defendant] was dealing drugs from a nearby apartment' that was the subject of the search. Id. at 459. That informant, the court later explained, was known to the officers, and other information the informant had provided was “proven right.” Id. at 464.

Each of these facts is important. As Sanders reminds us, the probable cause inquiry, “at its core, depends on the totality of the circumstances'-it's a 'constellation' in which each of these facts is but one “star.” Id. at 461, 463 (cleaned up). Recognizing that probable cause is “a fluid concept” that “turn[s] on the assessment of probabilities in particular factual contexts,” the Supreme Court “ha[s] rejected rigid rules, bright-line tests, and mechanistic inquiries in favor of a more flexible, all-things-considered approach.' Harris , 568 U.S. at 244 (quoting Illinois v. Gates , 462 U.S. 213, 232 (1983)).

So, despite some broad language in Sanders that was unnecessary given the facts of that case, I would not read our precedent as drawing a bright-line rule that there is probable cause every time officers observe a defendant leaving from a residence and then returning to it after a drug deal. Nor would I read the majority’s opinion in this case as establishing a bright-line rule that coming-and-returning evidence from a single drug sale suffices for probable cause. That language in the majority’s opinion here is unnecessary, like it was in Sanders , because there is additional nexus evidence—the text messages. Accordingly, I would not subject probable cause to the type of formalistic inquiry that-as even Sanders recognized-the Supreme Court has rejected. 106 F.4th at 461 (citing Maryland v. Pringle , 540 U.S. 366, 371 (2003)).

Even apart from precedent, it makes no sense to have a bright-line rule that coming-andreturning evidence from one drug sale suffices for probable cause. If we were to embrace such a rule, then in all cases, a defendant’s leaving a house, selling drugs, and returning to the house Page 19would necessarily raise a “fair probability” that the house contains drugs, regardless of context. A few examples demonstrate why, depending on context, that might not always be true. Consider a contractor who also deals drugs; they leave a client’s home where they’re working for the day, conduct a single drug sale, and then return to continue the job. Assume, as here, that the police do not know whether the drug dealer lives at the house. Automatic probable cause? I wouldn’t think so. What about a grandson who leaves his grandma’s house to sell drugs during an extended holiday celebration, and then returns to the party immediately after? These hypotheticals show that the majority opinion’s purported bright-line rule does not work.

Because there is no such bright-line rule, I would not call the probable cause question here 'straightforward.' Maj. Op. at 4. And we need not answer it because, as I explain, the government met the good-faith exception to the warrant requirement anyway. See Leon , 468 U.S. at 920-22; Ardd , 911 F.3d at 351. I would have affirmed the district court's denial of Florence’s motion to suppress on that much simpler ground.

Good Faith . We need not decide whether the warrant was supported by probable cause because, either way, we can affirm the district court’s denial of Florence’s motion to suppress based on the “good-faith” exception. As the district court explained, courts will not suppress the fruits of a search that lacked probable cause if the officer who executed the search could have relied on the warrant in good faith. Leon , 468 U.S. at 920–22. And I agree with the district court that this good-faith exception makes suppression inappropriate here.

To fall within the good-faith exception, the officer’s reliance on the warrant must have been in 'objective' good faith. Id . at 920. We have held that an officer does not rely on a warrant in objective good faith when the warrant is “so lacking in indicia of probable cause as to render official belief in its existence entirely unreasonable.' Id . at 923 (citation omitted). A warrant this deficient is also known as a 'bare bones affidavit.' Burrell , 114 F.4th at 554 (internal quotations omitted).

As relevant here, we will not consider an affidavit 'bare bones' if it demonstrates a “minimally sufficient nexus” between the alleged criminal activity and the place to be searched. Sanders , 106 F.4th at 469 (citation omitted). While it need not meet the more demanding Page 20probable cause standard, the affidavit must still include 'particularized facts that indicate veracity, reliability, and basis of knowledge” for an officer to be able to rely on it in good faith. United States v. McCoy , 905 F.3d 409, 416 (6th Cir. 2018) (citation omitted). The affidavit must do more than 'nakedly assume or vaguely conclude, without attempting to demonstrate why, probable cause has been satisfied.” Sanders , 106 F.4th at 468.

Based on this standard, the affidavit here was not bare bones. The affidavit underlying the search warrant detailed the investigation into Florence's drug dealing. And, as described above, it connected that drug dealing to the Melbourne residence through the coming-andreturning evidence and the text messages. That evidence shows at least a “minimally sufficient nexus' between the Melbourne residence and Florence's drug trafficking activity. Id. at 469. The executing officers could therefore have relied on the warrant in good faith, which-even without wading into probable cause—defeats Florence’s motion to suppress.

I also agree with the majority opinion’s decision to uphold the district court’s application of the premises enhancement. But I would have approached the standard of review differently.

Premises Enhancement Background. The Sentencing Guidelines provide for a two-level enhancement '[i]f the defendant maintained a premises for the purpose of manufacturing or distributing a controlled substance.' U.S.S.G. § 2D1.1(b)(12). The enhancement has three elements. The government must prove, by a preponderance of the evidence, that the defendant “(1) knowingly (2) open[ed] or maintain[ed] any place (3) for the purpose of manufacturing or distributing a controlled substance.' United States v. Johnson , 737 F.3d 444, 447 (6th Cir. 2013). The second element—whether Florence “maintained” the Melbourne residence—is the only one at issue here.

To determine whether a defendant 'maintained' a premises under § 2D1.1(b)(12), the Guidelines instruct courts to consider both “whether the defendant held a possessory interest in (e.g., owned or rented) the premises,” and “the extent to which the defendant controlled access to, or activities at, the premises.” U.S.S.G. § 2D1.1 cmt. 17. The government did not contend that Florence held a formal possessory interest in the property. Instead, the government argued Page 21that Florence “maintained” the Melbourne residence because he had “de facto control” over it. See United States v. Broadnax , 777 F. App'x 137, 141 (6th Cir. 2019) (citation omitted). The district court agreed, concluding that Florence exercised sufficient control of the Melbourne residence for the court to apply the enhancement.

Standards of Review. In reviewing a district court's decision to apply a sentencing enhancement, we generally start with ascertaining the applicable standard of review. See United States v. Washington , 715 F.3d 975, 982–83 (6th Cir. 2013). A case can present three types of questions: legal questions, factual questions, and mixed questions of law and fact. See U.S. Bank Nat’l Ass’n ex rel. CWCapital Asset Mgmt. LLC v. Vill. at Lakeridge, LLC , 583 U.S. 387, 393 (2018). The type of question presented dictates the appropriate standard of review. Legal questions are those that involve, for example, defining a governing standard or test, see id. , or interpreting a written instrument like a statute or regulation, Padhiyar v. Holder , 560 F. App'x 514, 517 (6th Cir. 2014). They are subject to de novo review. Id. Factual findings, also referred to as determinations of “historical fact,” answer “questions of who did what, when or where, how or why.” U.S. Bank , 583 U.S. at 394. We review factual findings for clear error, which puts 'a serious thumb on the scale' in favor of the district court's disposition. Id. The standard of review for mixed questions, such as “whether the historical facts found satisfy the [applicable] legal test,” varies, and has caused confusion and inconsistency in our sentencing enhancement caselaw. Id. at 394-96; see, e.g. , United States v. Thomas , 933 F.3d 605, 608–10 (6th Cir. 2019). Compare United States v. Walters , 775 F.3d 778, 782 (6th Cir. 2015) (reviewing 'de novo the district court's application of enhancements'), with United States v. Kessinger , 641 F. App'x 500, 506 (6th Cir. 2016) (stating that 'where the defendant challenges the application of a sentencing enhancement, we review for clear error”).

In my view, a different standard of review applies to each of Florence's arguments on appeal. As I read Florence’s appeal, he raises two main challenges to the district court’s ruling. The first is about what facts a district court can consider in determining whether a defendant “maintained” the premises for purposes of the enhancement. The second is about whether the record supports the district court’s finding that Florence was able to freely come and go from the Page 22Melbourne residence. I consider Florence’s arguments—along with the appropriate standards of review—in turn.

Permissible Facts for the Maintenance Analysis. Florence first argues that the district court impermissibly considered the 'sheer volume' of drugs, firearms, cash, and other drug trafficking tools kept at the Melbourne residence in determining whether he 'maintained' it. Appellant Br. at 31 (quoting United States v. Tripplet , 112 F.4th 428, 432-33 (6th Cir. 2024)). The district court, as the majority opinion describes, centered its maintenance analysis on the volume of contraband found at the Melbourne residence. Maj. Op. at 8. Officers found distribution quantities of narcotics, nine firearms, hundreds of rounds of ammunition, body armor, two digital scales, a pill press, a money counter, and almost $60,000 in cash. In his plea agreement, Florence admitted that all these items belonged to him. The district court reasoned that the substantial quantity of contraband showed control over the Melbourne residence because Florence was able to bring those items in “and obviously nothing was done about” it. Sent’g Tr., R. 76, PageID 556.

As Florence reads our caselaw, the district court was not permitted to consider this evidence in evaluating whether Florence 'maintained' the premises-i.e., exercised de facto control over it. According to Florence, our precedent segregates the type of evidence pertinent to the various prongs. Rehashing our cases, he argues that the presence of substantial quantities of contraband is relevant not to whether the defendant maintained the premises, but instead only to the third element of the enhancement: whether the premises were used 'for the purpose of manufacturing or distributing a controlled substance.” Johnson , 737 F.3d at 447. In support, he cites numerous precedents that have considered such evidence for the purpose element. By contrast, he emphasizes that our cases on the maintenance element involved different facts. Correctly, he states that facts this court has generally considered for discerning de facto control include the length of time over which the defendant frequented the site, whether the defendant’s presence was continuous, whether the defendant supervised the site, and whether the defendant took measures to protect the premises, among others. See Broadnax , 777 F. App'x at 141. Under his view, the relevant facts for the maintenance element are whether Florence had a key to the residence, how often he was found there, and whether he had to knock to enter.

Page 23Therefore, he argues, 'the district court erroneously substituted factors from this Court's precedent to establish the [purpose] prong of the enhancement to support the [maintenance] prong.” Appellant Br. at 25.

I view this argument as mainly a legal one that should be subject to de novo review. Florence is arguing that a category of facts cannot be used to prove maintenance; it may not be the strongest legal argument, but it asks us to define the scope of proof relevant to an element of the enhancement. In doing so, Florence essentially asks this court to place limits on 'the governing totality-of-the-circumstances standard' for discerning maintenance, which is a legal question. See Monasky v. Taglieri , 589 U.S. 68, 84 (2020). Therefore, our review of this issue should be de novo. Id. at 83; U.S. Bank , 583 U.S. at 393. And that de novo review is for good reason—clarifying the type of facts that district courts can consider in the maintenance analysis helps “‘unify precedent’ and ensure that like defendants are treated alike.” United States v. Hill , 963 F.3d 528, 531 (6th Cir. 2020) (quoting Ornelas v. United States , 517 U.S. 690, 697 (1996)).

To be sure, this question may look at first glance like a mixed question that involves applying the governing law to historical facts. One could recast Florence's argument as contending that, based on what he views as the permissibly considered facts, there was insufficient evidence in the record to support the enhancement. But many legal questions can be reframed as law-to-fact application questions, and what separates a legal question from a mixed question may sometimes be a fine line. However construed, crucial to the standard of review here is that Florence is arguing that the district court impermissibly relied on facts pertinent only to the purpose element. He does not ask this court to reweigh the totality of the circumstances that the district court considered. See Buford v. United States , 532 U.S. 59, 64-65 (2001); see also Monasky , 589 U.S. at 84. Instead, Florence asks this court to elucidate the facts that can be permissibly considered in the maintenance analysis in the first place. In my view, that question is subject to de novo review. Monasky , 589 U.S. at 84.

On de novo review, I would hold that district courts can consider the presence of substantial quantities of contraband in the maintenance analysis. This court has considered the volume of contraband as part of its maintenance analysis in several unpublished opinions. See, e.g. , United States v. Tippins , 630 F. App'x 501, 504 (6th Cir. 2015); United States v. Barker , Page 24No. 24-3492, 2025 WL 707870, at *2 (6th Cir. Mar. 5, 2025); United States v. Hill , No. 22-5274, 2023 WL 152474, at *1, *3 (6th Cir. Jan. 11, 2023). I see no reason to proceed differently here. While the quantity of contraband is perhaps more frequently discussed in analyzing the purpose prong, it can be relevant to showing de facto control too. As the majority opinion recognizes, the fact that Florence 'stored tens of thousands of dollars in cash, sizable amounts of drugs, drug paraphernalia, firearms, and body armor” at the Melbourne residence supports the inference that he had control over it, because it is unlikely he would have left such valuable items there otherwise. Maj. Op. at 11–12.¹ I therefore would not cabin the consideration of the presence of significant quantities of contraband to the “purpose” prong, as Florence asks this court to do.

Access to the Melbourne Residence. Even under Florence's narrow reading of the maintenance element, he recognizes that the district court could consider the amount of access he had to the Melbourne residence as part of its control analysis. But he next argues that the district court erred in finding that he could come and go freely from the house because, in his view, that level of access is not supported by the record. This quarrel is with the district court's factfinding—it disputes “who” (Florence) “did what” (access the house) “when” (whenever he wanted) and “why” (to access his drugs, money, and related property). U.S. Bank , 583 U.S. at 394. It is a “well-settled rule” that “such factual findings are reviewable only for clear error.” Id.

On clear error review, I see no reason to disturb the district court’s finding that Florence could come and go from the Melbourne residence freely. Florence admitted in his plea agreement that all the drugs and other contraband found at the residence were his. It was reasonable for the district court to infer that he would not have stored these items—especially almost $60,000 worth of cash—at the residence if he could not access them at will. Moreover, Florence's counsel conceded at sentencing that Florence could access the residence as he pleased. Sent’g Tr., R. 76, PageID 535–36 (responding “[r]ight” when the district court posited Page 25that Florence was “able to come and go” from the residence). The district court did not clearly err in inferring that Florence had free access to the residence.

Residual Arguments. To the extent that Florence argues that the evidence is insufficient to establish that he maintained the premises, that idea is predicated on his view that the district court could not consider the volume of drugs and other contraband in the Melbourne residence or his ability to freely access the house. His argument is that there was 'no evidence' demonstrating his control of the residence beyond the “scant evidence” that Florence was at the Melbourne residence twice, once surrounding the drug sale the officers surveilled and again when they executed the search warrant. Appellant Br. at 29. This argument is, for the reasons described above, based on an erroneous view of the facts the district court could consider, and an unsuccessful challenge to its factfinding. I would reject any residual mixed question about the application of the maintenance element to the facts on that ground.

Accordingly, I join the majority opinion in affirming the district court on both issues, but I would have done so based on the grounds I have described above. I thus respectfully concur in the judgment.