I concur in the majority opinion. There was probable cause to search Tyrone Christian's house, and, at the very least, the officers executed that search in good faith. But because of our precedent, we must ignore critical evidence of which the officers undisputedly knew and isolate the good-faith analysis to the four corners of the affidavit. See United States v. Laughton , 409 F.3d 744, 751-52 (6th Cir. 2005). I write separately to explain why Laughton 's limit on the good-faith exception conflicts with Supreme Court precedent and should be overruled.
I.
Officer Thomas Bush's affidavit included a number of facts linking Christian and his house to drug trafficking: (1) Christian had four drug-crime convictions in the past nineteen years (two of which involved conduct that occurred at his house); (2) a confidential informant had purchased drugs from Christian at his house nine months earlier; (3) within the past four months, several "subjects" told the officers that they had purchased "large quantities" of drugs from Christian at his house; and, finally, (4) on the day of the search, officers stopped Rueben Thomas after they saw him leave the "area of" Christian's house and discovered 20 grams of heroin in Thomas's car. R. 42-1, Pg. ID 114-15. Critically, Thomas's heroin showed current drug dealing at Christian's house, supplementing the older information in Bush's affidavit. But the link between Thomas's heroin and Christian's house was blurry because the affidavit was vague. The affidavit did not say that the officers saw Thomas interact with Christian or that they saw Thomas inside Christian's house-only that they saw him "walk away from the area of" Christian's house. Id. at 115.
Still, the magistrate believed the affidavit was good enough and granted the officers' request for a search warrant. After obtaining the warrant, the officers *315searched Christian's house and uncovered extensive evidence of drug dealing: marijuana, cocaine, heroin, drug packaging materials, and two guns. Based on this evidence, Christian was convicted of possessing a controlled substance with intent to distribute, possessing a firearm in furtherance of drug trafficking, and being a felon in possession of a firearm.
Christian claims the evidence against him should have been suppressed, arguing that the officers lacked probable cause to search his house and that the good-faith exception to the exclusionary rule does not apply. Because of Laughton , the parties' good-faith arguments are restricted to the language of the affidavit. And because that language is vague on a critical point-the link between Thomas's heroin and Christian's house-the parties parse through the affidavit and debate the best interpretation of its language (almost as if they were interpreting a statute).
But uncontroverted evidence shows that on the day of the search, surveilling officers twice observed Thomas interacting with Christian at Christian's house. First, Thomas met with Christian for "approximately five minutes" in the driveway of his house. R. 152, Pg. ID 1131-32. Then, later that afternoon, Thomas returned and went inside for about two hours. After he left, the officers stopped him and discovered the heroin. These facts link Thomas and his heroin to Christian and his house. But, unfortunately, they were left out of the affidavit. The first encounter did not make it into the affidavit at all, and the second one did only in the vague terms described above.
Laughton confines us to the words of that vague affidavit in evaluating whether the good-faith exception applies. We cannot consider the officers' actual observations or determine the reason those observations did not make it into the affidavit.
II.
Laughton is wrong. To see why, we need to start with first principles. The Fourth Amendment protects "[t]he right of the people to be secure ... against unreasonable searches and seizures." U.S. Const. amend. IV. But it does not spell out how we are to protect that right. When the Fourth Amendment was ratified, the only way to enforce its protections was through private tort suits against officers-the exclusionary rule, Section 1983, and Bivens actions did not yet exist. See Collins v. Virginia , --- U.S. ----, 138 S. Ct. 1663, 1676, 201 L.Ed.2d 9 (2018) (Thomas, J., concurring) ("Historically, the only remedies for unconstitutional searches and seizures were 'tort suits' and 'self-help.' "); Gardner v. Neil , 4 N.C. 104, 104 (1814) (stating that "the action of trespass is the only proper form of action" for a Fourth Amendment violation); Akhil R. Amar, The Bill of Rights as a Constitution , 100 Yale L.J. 1131, 1176-78 (1991) ; William Baude & James Y. Stern, The Positive Law Model of the Fourth Amendment , 129 Harv. L. Rev. 1821, 1840 (2016).
That changed in 1914 when the Supreme Court first excluded evidence obtained in violation of the Fourth Amendment. Weeks v. United States , 232 U.S. 383, 34 S.Ct. 341, 58 L.Ed. 652 (1914). The facts in Weeks were extreme: officers, lacking any particularized information or a warrant, broke into the defendant's home, took incriminating documents, then returned and took even more. See id. at 386, 393-94, 34 S.Ct. 341. To deter such flagrant misconduct by law enforcement, the Supreme Court created the exclusionary rule. Id. at 393-94, 34 S.Ct. 341 ; see also United States v. Leon , 468 U.S. 897, 906, 908, 916-17, 104 S.Ct. 3405, 82 L.Ed.2d 677 (1984). The underlying premise is that police are less likely to engage in misconduct if they *316know that any evidence obtained thereby will be inadmissible at trial.
But the Supreme Court has recognized that suppression often comes with its own "substantial" costs-both to the criminal justice system (letting the guilty (and possibly dangerous) go free) and to the truth-seeking process. Leon , 468 U.S. at 907-08, 104 S.Ct. 3405. So the Supreme Court has repeatedly reminded us that suppression should always be "our last resort, not our first impulse." Herring v. United States , 555 U.S. 135, 140, 129 S.Ct. 695, 172 L.Ed.2d 496 (2009) (quoting Hudson v. Michigan , 547 U.S. 586, 591, 126 S.Ct. 2159, 165 L.Ed.2d 56 (2006) ) (internal quotation mark omitted). In turn, several principles constrain the application of the exclusionary rule. First, exclusion is not an individual right but a rule aimed at deterrence. Id. at 141, 129 S.Ct. 695. Second, a Fourth Amendment violation is a necessary-but not a sufficient-ground for exclusion. Id. Third, and perhaps most importantly, the value of any future police deterrence must outweigh suppression's "substantial social costs." Hudson , 547 U.S. at 596, 126 S.Ct. 2159.
Assuming there is a Fourth Amendment violation, how exactly should courts balance the costs versus the benefits of suppression? Again, the Supreme Court tells us: look at the misconduct. Exclusion must deter egregious misconduct-misconduct "sufficiently deliberate that exclusion can meaningfully deter it, and sufficiently culpable that such deterrence is worth the price paid by the justice system." Herring , 555 U.S. at 144, 129 S.Ct. 695. In contrast, when officers act in an objectively reasonable but mistaken manner, exclusion serves no purpose. Leon , 468 U.S. at 919, 104 S.Ct. 3405. That is so even if that mistake violated a suspect's Fourth Amendment rights. See Hudson , 547 U.S. at 596, 126 S.Ct. 2159.
In short, the ultimate focus must be on the nature of police misconduct. That conduct must exhibit "deliberate, reckless, or grossly negligent disregard for Fourth Amendment rights" to trigger the exclusionary rule. Davis v. United States , 564 U.S. 229, 238, 131 S.Ct. 2419, 180 L.Ed.2d 285 (2011) (internal quotation marks omitted); Herring , 555 U.S. at 144, 129 S.Ct. 695 (adding systemic negligence to the list).
III.
Of course, focusing on police misconduct does not excuse courts from looking at the affidavit. Indeed, the affidavit is where courts must start. Leon , 468 U.S. at 915, 104 S.Ct. 3405. But the ultimate inquiry is whether, considering "all of the circumstances," the officers acted reasonably when relying on the blessing of the judge. Herring , 555 U.S. at 145, 129 S.Ct. 695 (quoting Leon , 468 U.S. at 922 n.23, 104 S.Ct. 3405 ). And to make this determination, "we must consider the actions of all the police officers involved." Id. at 140, 129 S.Ct. 695 (citing Leon , 468 U.S. at 923 n.24, 104 S.Ct. 3405 ).
This is where Laughton went astray. Laughton 's refusal to look beyond the affidavit is, in effect, a judgment that factual omissions are always culpable misconduct. To start, Leon precludes such an all-or-nothing approach to the exclusionary rule. 468 U.S. at 922 n.23, 923 n.24, 104 S.Ct. 3405 (explaining that good faith depends on "all of the circumstances"). But more importantly, the underlying premise is not true-omitted facts usually do not stem from misconduct at all but from isolated negligence or the time pressures that officers often face during investigations. Indeed, an officer would have practically no incentive to leave favorable information out of an affidavit. See Hudson , 547 U.S. at 596, 126 S.Ct. 2159 (stating "the value of *317deterrence depends upon the strength of the incentive to commit the forbidden act"). Doing so would only increase the chance that a magistrate may reject the warrant application and "leav[e] the officer empty-handed." United States v. Thomas , 908 F.3d 68, 74-75 & n.3 (4th Cir. 2018).
Because there is no nefarious conduct to deter, the best that excluding evidence may do in this scenario is encourage more careful affidavit drafting. While that is a laudable goal, it is not worth the substantial costs of exclusion. See Herring , 555 U.S. at 141, 129 S.Ct. 695 ; Hudson , 547 U.S. at 596, 126 S.Ct. 2159. Those costs are particularly high in an omitted-facts situation: when an officer in fact reasonably relied on the magistrate's warrant for a search, and that search yielded evidence proving that the defendant is in fact guilty. Under that scenario, the exclusionary rule cannot "pay its way." Davis , 564 U.S. at 238, 131 S.Ct. 2419 (quoting Leon , 468 U.S. at 919, 104 S.Ct. 3405 ).
The Supreme Court's instruction to focus on culpability is enough to show that the good-faith analysis must consider facts that are not included in the affidavit. But the Supreme Court has been even more explicit. In Sheppard , an officer under severe time pressure used the wrong warrant application form for his search (a form for drugs rather than murder). Massachusetts v. Sheppard , 468 U.S. 981, 986, 104 S.Ct. 3424, 82 L.Ed.2d 737 (1984). The magistrate judge explained that edits were necessary but only made some of them; as a result, the warrant still authorized only a search for drugs. Id. at 986-87, 104 S.Ct. 3424. Despite the obvious error, the Court held that the good-faith exception applied. In doing so, it rejected the argument that the officers' reliance on a facially invalid warrant undermined good faith. Given the circumstances, "[t]he officers ... took every step that could reasonably be expected of them." Id. at 987-89, 104 S.Ct. 3424. Among other things, they thoroughly investigated the suspect in a short amount of time, sought the advice of a district attorney, presented the warrant application to a judge, and trusted that he had fixed it. Id. at 984, 988-89, 104 S.Ct. 3424. Those facts were not in the affidavit but still were relevant to the Sheppard court. Thus, Sheppard "forecloses ... a categorical rule" that the good-faith exception depends entirely on the face of the warrant itself. United States v. Franz , 772 F.3d 134, 146 (3d Cir. 2014) ; accord United States v. Frazier , 423 F.3d 526, 534-35 (6th Cir. 2005). And Sheppard 's logic extends to affidavits and any other documents in a warrant application.
Indeed, our sister circuits have applied the good-faith exception when affidavits (often prepared under time pressure) omitted a few words that were needed to establish probable cause. See, e.g. , United States v. McKenzie-Gude , 671 F.3d 452, 456-57, 460 (4th Cir. 2011) ; United States v. Martin , 297 F.3d 1308, 1320 (11th Cir. 2002). Even the Tenth Circuit, which purportedly follows a "four corners" rule, still considers (1) additional information presented to the issuing judge, (2) "information relating to the warrant application process," and (3) "testimony illuminating how a reasonable officer would interpret factual information contained in an affidavit." See United States v. Knox , 883 F.3d 1262, 1272 & n.9 (10th Cir. 2018).
In addition, we already allow courts to look outside facially valid documents to see if there was a Fourth Amendment violation that compels suppression. For example, if a facially valid warrant was rooted in culpable misconduct, then the good-faith exception does not apply. See Herring , 555 U.S. at 146, 129 S.Ct. 695 ("If the police have been shown to be reckless in maintaining a warrant system, or to have *318knowingly made false entries to lay the groundwork for future false arrests, exclusion would certainly be justified ...."); see also Franks v. Delaware , 438 U.S. 154, 155-56, 98 S.Ct. 2674, 57 L.Ed.2d 667 (1978). Likewise, a facially valid warrant cannot support good faith if the officers purposely withheld damaging information from it to present an "incomplete and misleading" picture to the magistrate. United States v. West , 520 F.3d 604, 611-12 (6th Cir. 2008). In other words, we already consider facts outside the affidavit when evaluating good faith-we just consider facts that undermine probable cause and ignore facts that support it. Under Laughton , outside facts are a one-way ratchet in favor of criminals. This disparity upsets the cost-benefit balance at the heart of the good-faith exception: we should only undermine the truth-finding function of the criminal justice system when necessary to deter culpable misconduct.
IV.
What would a world without Laughton look like in practice? No court can envision every situation in which good faith does or does not apply. But a few things are clear. Whenever an affidavit's four corners are thorough enough to satisfy Leon , the good-faith exception applies (barring some sort of culpable misconduct by the police). Indeed, Leon is a low bar. An affidavit exceeds the Leon bar when it contains "some connection" between "the illegal activity and the place to be searched," even if that connection is "remote" and supported by only a slight "modicum of evidence." United States v. White , 874 F.3d 490, 496-97 (6th Cir. 2017) (quoting United States v. Carpenter , 360 F.3d 591, 596 (6th Cir. 2004) (en banc) (internal quotation marks omitted)). Too often courts raise the Leon bar, making it practically indistinguishable from the probable cause standard itself.1 Doing so effectively eliminates the good-faith exception, or at the very least reduces it to a limited, narrow role. But when properly applied, Leon means that in most cases an affidavit will satisfy the good-faith exception. In the instances when an affidavit is vague on a critical point, however, courts must go further and ask whether the officers were objectively culpable in relying on that affidavit. To answer that question, courts must consider all the circumstances that bear on the officers' culpability, including any time pressure that the officers were under, what facts were known to the officers but omitted from the affidavit, and how defective the affidavit was without that omitted information.
The dissent claims that this culpability-focused approach would entail a subjective "expedition into the minds of police officers." Dissenting Op. Part II (quoting Leon , 468 U.S. at 922 n.23, 104 S.Ct. 3405 ). Yet the Supreme Court already explained that looking at an "officer's knowledge and experience ... does not make the test ... subjective." Herring , 555 U.S. at 145-46, 129 S.Ct. 695 (emphasis added). Courts are not "inquiring into the subjective beliefs of law enforcement officers" when they consider *319"actual uncontroverted facts" known to them. McKenzie-Gude , 671 F.3d at 460-61 (citing Herring , 555 U.S. at 145, 129 S.Ct. 695 ). Neither the good-faith exception nor Supreme Court precedent require that we bury our heads in the sand and ignore uncontroverted evidence. We should follow the Supreme Court's lead and consider such evidence when determining the officers' culpability. See Herring , 555 U.S. at 146-47, 129 S.Ct. 695 ; Sheppard , 468 U.S. at 987-89, 104 S.Ct. 3424. To do anything less is to ignore the very purpose for the exclusionary rule in the first place.
Indeed, this case proves the point. Surveilling officers twice observed Thomas interacting with Christian at Christian's house. That is an uncontroverted fact, not a subjective belief. And that fact, had it been included in the affidavit, would have at the very least established good faith under Leon . A tight time constraint, not culpable conduct, is the most likely reason that this information was left out. See Majority Op. Part II.A. In contrast, if officers saw Thomas at a different house meeting with someone other than Christian, this would be a different case. In that light, the vague language in the affidavit ("the area of" Christian's house) would objectively appear to be intentional obfuscation rather than negligent oversight. R. 42-1, Pg. ID 115. And under those circumstances, the outside facts would support suppression. But with the facts we have, suppressing the evidence serves no societal purpose.
In short, courts can only apply the good-faith exception by evaluating officer conduct and can only evaluate officer conduct by looking beyond the four corners of the affidavit. The time has come for us to get in line with the Supreme Court's good-faith doctrine. We should overrule Laughton .
CONCURRING IN THE JUDGMENT
HELENE N. WHITE, Circuit Judge, concurring in the judgment.
I join in Part I of Judge Gilman's opinion. However, because I conclude that the search-warrant affidavit was sufficient to justify a reasonably well-trained officer's good-faith reliance on the magistrate's finding of probable cause, United States v. White , 874 F.3d 490, 496 (6th Cir. 2017), I concur in the result of the majority opinion.
DISSENT
RONALD LEE GILMAN, Circuit Judge, dissenting.
Considering the totality of the circumstances, I believe that the facts set forth in the affidavit fail to establish a "fair probability" that drug activity was occurring at Christian's residence (the Residence) at the time the search warrant was executed. See United States v. Brooks , 594 F.3d 488, 492 (6th Cir. 2010). I also do not think that the exception established in United States v. Leon , 468 U.S. 897, 104 S.Ct. 3405, 82 L.Ed.2d 677 (1984), applies in the present case because the affidavit does not establish a sufficient nexus between the Residence and drug activity at the time of the search.
The majority's conclusion that the issue of probable cause is "really not even close" strikes me as totally unsupportable. See Maj. Op. 309. Unlike the majority, I acknowledge that whether there was probable cause and whether the good-faith exception is met are close calls. But I ultimately conclude that the affidavit falls short because it does not provide any "particularized facts" connecting the Residence to drug activity at the time that the search warrant was executed. See *320United States v. McPhearson , 469 F.3d 518, 524 (6th Cir. 2006). I therefore respectfully dissent.
I. PROBABLE CAUSE
"To establish probable cause adequate to justify issuance of a search warrant, the governmental entity or agent seeking the warrant must submit to the magistrate an affidavit that establishes 'a fair probability that contraband or evidence of a crime will be found in a particular place.' " Brooks , 594 F.3d at 492 (quoting United States v. Berry , 565 F.3d 332, 338 (6th Cir. 2009) ). This requires "a nexus between the place to be searched and the evidence sought," McPhearson , 469 F.3d at 524 (quoting United States v. Carpenter , 360 F.3d 591, 594 (6th Cir. 2004) (en banc)), at the time the warrant is issued, United States v. Hython , 443 F.3d 480, 485 (6th Cir. 2006). The probable-cause standard is practical and nontechnical. United States v. Frazier , 423 F.3d 526, 531 (6th Cir. 2005). In other words, a reviewing court should consider the "totality of the circumstances" rather than "engage in line-by-line scrutiny of the warrant application's affidavit." United States v. Williams , 544 F.3d 683, 686 (6th Cir. 2008). But the court must limit its "review of the sufficiency of the evidence supporting probable cause ... to the information presented in the four-corners of the affidavit." Frazier , 423 F.3d at 531.
The totality-of-the-circumstances approach requires us to examine each piece of evidence in the affidavit to assess its probative value and then determine whether those pieces of evidence are as a whole sufficient to establish probable cause. Gardenhire v. Schubert , 205 F.3d 303, 315 (6th Cir. 2000) (explaining that, in the context of an arrest, "[p]robable cause determinations involve an examination of all facts and circumstances within an officer's knowledge at the time of an arrest" (quoting Estate of Dietrich v. Burrows , 167 F.3d 1007, 1012 (6th Cir. 1999) )); United States v. Valenzuela , 365 F.3d 892, 897 (10th Cir. 2004) ("[C]ourts may not engage in a 'divide-and-conquer' analysis of facts to determine whether probable cause existed. However, neither may a court arrive at probable cause simply by piling hunch upon hunch. Thus, in assessing the totality of the circumstances, a reviewing court 'must examine the facts individually in their context to determine whether rational inferences can be drawn from them' that support a probable cause determination." (emphasis added) (citations omitted)).
A. Observations of Thomas
I will begin by analyzing the probative value of the evidence presented in the four corners of the affidavit, starting with the officers' observations of Thomas. Then, as the Supreme Court instructed in District of Columbia v. Wesby , --- U.S. ----, 138 S. Ct. 577, 199 L.Ed.2d 453 (2018), I will consider each piece of the evidence "as a factor in the totality of the circumstances." See id. at 588 (citations omitted).
According to the affidavit, law-enforcement officers observed Thomas "walk away from the area" of the Residence and leave in a vehicle on the day that the search warrant was issued. They followed Thomas and stopped him after an unknown period of time for a driving infraction. During the stop, the officers found approximately 20 grams of heroin in Thomas's vehicle. Crucially, the affidavit does not state that the officers saw Thomas entering or leaving the Residence, even though their surveillance was targeted specifically at that property. Nor does it say that Thomas was seen with Christian. In fact, the affidavit does not assert any connection at all between Christian and Thomas.
True enough, the affidavit states that, during the traffic stop, "Rueben Thomas *321admitted that he had recently been at an address on Grandville Avenue in the City of Grand Rapids but denied being at 618 Grandville[,] contrary to observations of the law enforcement officers." But I decline to interpret this "contrary to observations" statement as an indication that the officers saw Thomas actually entering or leaving the Residence itself. Officer Bush was undoubtedly aware that any evidence of Thomas being at the Residence would be highly relevant to the probable-cause determination, but chose instead to state simply that Thomas was seen walking in "the area of 618 Grandville"-a vague description that does not place Thomas at the Residence. Absent a direct statement that Thomas was seen entering or leaving the Residence, or even at the Residence in any sense, I find no basis to read such a factual assertion into the affidavit.
The majority, on the other hand, contends that the affidavit's lack of a direct statement that Thomas was at the Residence is attributable to the "haste of a criminal investigation," Maj. Op. 310 (quoting Illinois v. Gates , 462 U.S. 213, 235, 103 S.Ct. 2317, 76 L.Ed.2d 527 (1983) ), and that no "magic words" are required, Maj. Op. 310 (quoting United States v. Allen , 211 F.3d 970, 975 (6th Cir. 2000) (en banc)). But the affidavit's inclusion of the specific, nontechnical language "from the area" appears to me more consistent with an honest acknowledgement that the officers did not observe Thomas at the Residence itself. In fact, the common-sense meaning of the language "from the area" suggests that Thomas was near but not at the Residence when observed by the officers.
The majority also notes that the affidavit "need only have 'allege[d] facts that create a reasonable probability" that "Thomas was seen leaving 618 Grandville." Maj. Op. 310 (alteration in original) (quoting United States v. Tagg , 886 F.3d 579, 589 (6th Cir. 2018) ). This statement reflects a subtle but crucial error. The affidavit must contain facts establishing that probable cause exists to believe that evidence of drug activity will be present in the Residence at the time of the search . Stated differently, probable cause must be established in relation to whether there is evidence of drug activity in the Residence, not in relation to whether Thomas was seen leaving the Residence. See United States v. Brooks , 594 F.3d 488, 492 (6th Cir. 2010) ("To establish probable cause adequate to justify issuance of a search warrant, the governmental entity or agent seeking the warrant must submit to the magistrate an affidavit that establishes 'a fair probability that contraband or evidence of a crime will be found in a particular place.' " (quoting United States v. Berry , 565 F.3d 332, 338 (6th Cir. 2009) )).
This leaves us to consider the significance of the following: A single individual with no known connection to Christian was seen walking away from the area of the Residence and then leaving that area in a car. He was followed by officers for an unknown amount of time to a subsequent location where a traffic stop was conducted, during which heroin was found in the vehicle. If this provides any nexus at all between evidence of drug trafficking and the Residence, that nexus is so speculative and attenuated that it cannot, without more, support a finding of probable cause. See United States v. Arvizu , 534 U.S. 266, 274, 122 S.Ct. 744, 151 L.Ed.2d 740 (2002) (noting that a hunch is insufficient to support a finding of reasonable suspicion for a Terry stop, and that the reasonable-suspicion standard is easier to satisfy than the probable-cause standard).
To conclude otherwise would allow officers seeking a search warrant to rely on speculation that drug activity near a residence *322is related to that residence, significantly lowering the burden for the government to show probable cause in areas where drugs are prevalent. Because the government cites no case that would support such a proximity test for establishing probable cause, I believe that the officers' observation of Thomas has little value on its own. But that does not end the inquiry. We must consider, as I do below, whether other evidence in the record bolsters or corroborates a connection between Thomas's alleged drug activity and the Residence, such that the magistrate could have found a fair probability that evidence of drug trafficking would be found at the Residence at the time of the search.
B. Tips from unidentified informants
The affidavit further states:
Within the last four months, your affiant has been involved in or received information from several debriefs of subjects who have stated that Tyrone Christian is a large scale drug dealer. These subjects further stated that they have purchased large quantities of heroin and crack cocaine from Christian at 618 Grandville Avenue ... in the last four to five months.
Officer Bush's assertion that he received information from unidentified "subjects" omits critical particulars. Among other things, the affidavit does not identify the number of individuals who made the statements, explain what constituted a "debrief," identify the contexts in which the debriefs occurred, or even specify the date that the information was received (all of the information could have been received four months before the search).
More importantly, Officer Bush's statement gives no indication as to the veracity or reliability of the information obtained. He did not assert any belief concerning the reliability or veracity of the subjects' comments, let alone provide any factual basis by which the magistrate could assess their reliability or veracity. See United States v. Helton , 314 F.3d 812, 822 (6th Cir. 2003) (explaining that, under Sixth Circuit precedent, an affidavit "must contain a statement about some of the underlying circumstances indicating the informant was credible or that his information was reliable" (quoting United States v. Smith , 182 F.3d 473, 477 (6th Cir. 1999) )).
The affidavit's complete failure to address the credibility and reliability of the information provided by the subjects is even more glaring when juxtaposed with Officer Bush's inclusion of a paragraph supporting the credibility and reliability of the confidential informant who conducted the controlled buy in January 2015. With regard to this latter informant, Officer Bush stated that "[y]our affiant was able to confirm much of the information provided by the credible and reliable informant through information maintained by the Grand Rapids Police Department, other credible and reliable informants, public information sources and other law enforcement agencies." This statement indicates that Officer Bush was well aware that hearsay statements from informants should be accompanied by an explanation of their credibility and reliability. Accordingly, his failure to do so with respect to information obtained from the unidentified subjects implies the absence of any such indicia.
An affidavit establishing probable cause based on an informant's tip must also provide facts identifying the basis of the informant's knowledge. United States v. Frazier , 423 F.3d 526, 532 (6th Cir. 2005). "The 'basis for knowledge' factor uses the degree of detail in a tip to infer whether the tipster 'had a reliable basis for making his statements.' " Helton , 314 F.3d at 822 *323(quoting Smith , 182 F.3d at 477 ). Although Officer Bush's affidavit states that the basis of the subjects' knowledge was that they had each purchased drugs from Christian at the Residence, the affidavit provides almost no details about the purchases beyond identifying the types of drugs involved. The unidentified subjects did not state exactly when they had purchased drugs from Christian, the amount of the drugs purchased, or whether they entered the Residence and saw any controlled substances or other evidence of drug trafficking inside. This lack of detail further reduces the probative value of the information obtained from these sources.
"[I]n the absence of any indicia of the informants' reliability, courts insist that the affidavit contain substantial independent police corroboration." Frazier , 423 F.3d at 532. There is no evidence in the present case that the police corroborated any of the information obtained from the unidentified subjects. The affidavit does not indicate that the police engaged in any ongoing surveillance of the Residence, conducted subsequent controlled purchases, or otherwise tried to verify that Christian was currently using the Residence to sell drugs. And although the police established surveillance of the Residence on the very day that the affidavit was executed, the affidavit contains no observations by the police suggesting that Christian was then using the Residence as a base of operations.
Because the information from these unidentified subjects lacks any indicia of veracity or reliability and was not corroborated by subsequent police investigation, it should be accorded little weight in determining whether there was probable cause to search the Residence. See United States v. McPhearson , 469 F.3d 518, 524 n.3 (6th Cir. 2006) ("Thus, an allegation of drug dealing based on information from an untested confidential informant is insufficient to establish probable cause to search the alleged drug dealer's home. However, where the allegation of drug dealing is coupled with independently corroborated information from police officers, it may be sufficient to establish probable cause."); Helton , 314 F.3d at 822 (concluding that little weight should be given to statements from an informant whose reliability has not been determined); see also United States v. Allen , 211 F.3d 970, 976 (6th Cir. 2000) (en banc) (noting that an anonymous tip, even one that is "rich in particulars," will not be enough to establish probable cause if only innocent details are corroborated by the police, but holding that a magistrate may find probable cause to search a residence when "a known person, named to the magistrate, to whose reliability an officer attests with some detail, states that he has seen a particular crime and particular evidence, in the recent past").
C. The January controlled buy
Christian also contends that the evidence of the January 2015 controlled buy was stale when the affidavit was executed eight months later, and thus could not have supported a finding of probable cause to search the Residence. The government disputes this contention, arguing that because the officers sought records and indicia of continuous drug trafficking, the evidence was not stale.
"[S]tale information cannot be used in a probable cause determination." United States v. Perry , 864 F.3d 412, 414 (6th Cir. 2017) (quoting United States v. Frechette , 583 F.3d 374, 377 (6th Cir. 2009) ); see also United States v. Harris , 255 F.3d 288, 299 (6th Cir. 2001) ("Because probable cause to search is concerned with facts relating to a presently existing condition, ... there arises the unique problem of whether the *324probable cause which once existed has grown stale." (quoting United States v. Spikes , 158 F.3d 913, 923 (6th Cir. 1998) )). Whether evidence is stale is a flexible inquiry that does not "create an arbitrary time limitation within which discovered facts must be presented to a magistrate." United States v. Greene , 250 F.3d 471, 480 (6th Cir. 2001) (quoting Spikes , 158 F.3d at 923 ). In considering the length of time between the events listed in the affidavit and the application for the warrant, a court should consider several factors, including:
[1] the character of the crime (chance encounter in the night or regenerating conspiracy?), [2] the criminal (nomadic or entrenched?), [3] the thing to be seized (perishable and easily transferable or of enduring utility to its holder?), [and 4] the place to be searched (mere criminal forum of convenience or secure operational base?) ....
Spikes , 158 F.3d at 923 (quoting Andresen v. State , 24 Md.App. 128, 331 A.2d 78, 106 (Md. Ct. Spec. App. 1975) ).
1. Second and fourth Spikes factors
There is little question that the second and fourth factors weigh in favor of finding that the evidence of the January controlled buy was not stale. The affidavit supports the conclusion that Christian had been occupying the Residence in Grand Rapids since at least 2009 and was thus "entrenched" in the community. See Frechette , 583 F.3d at 379 (finding that the defendant was entrenched when evidence in the affidavit indicated that he had lived in the residence in question for 16 months). Moreover, courts have repeatedly held that a defendant's residence "is clearly a 'secure operational base.' " Id. (quoting United States v. Paull , 551 F.3d 516, 522 (6th Cir. 2009) ); see also United States v. Powell , 603 F. App'x 475, 478 (6th Cir. 2015) (concluding that an individual's home "is more like a secure operational base than a mere forum of convenience").
2. First Spikes factor
The first and third factors, however, weigh in favor of finding that the evidence of the controlled buy was stale. With regard to the first factor, "[i]f an affidavit recites activity indicating protracted or continuous conduct, time is of less significance." United States v. Henson , 848 F.2d 1374, 1382 (6th Cir. 1988) (quoting United States v. Haimowitz , 706 F.2d 1549, 1554-55 (11th Cir. 1983) ). This court has used both the terms "protracted and continuous" and "ongoing and continuous." Compare Perry , 864 F.3d at 415, with United States v. Hython , 443 F.3d 480, 485 (6th Cir. 2006). Both variations appear to encompass two principles: that the conduct extended over a significant period of time and that it continued up to (or close to) the time of the search. (For clarity, I will use "protracted" for the first principle and "continuous" for the latter.) The key question, then, is whether the affidavit contains facts supporting an inference that Christian was engaged in recurrent or sustained drug-trafficking activity up to the time of the search.
As this court has pointed out:
The crime at issue in this case-the sale of drugs out of a residence-is not inherently ongoing. Rather, it exists upon a continuum ranging from an individual who effectuates the occasional sale from his or her personal holdings of drugs to known acquaintances, to an organized group operating an established and notorious drug den. The inclusion of outdated information has been insufficient to render an entire affidavit stale when the affidavit as a whole establishes that the criminal activity in question is ongoing *325and continuous, or closer to the "drug den" end of the continuum.
Hython , 443 F.3d at 485. But if the affidavit, taken as a whole, suggests that the defendant is engaged in something closer to the "occasional sale from ... personal holdings," id. , then "information goes stale very quickly 'because drugs are usually sold and consumed in a prompt fashion,' " United States v. Brooks , 594 F.3d 488, 493 (6th Cir. 2010) (quoting Frechette , 583 F.3d at 378 ).
Here, Officer Bush explains that "a credible and reliable informant" engaged in a controlled purchase of drugs from Christian at the Residence in January 2015, eight months before Officer Bush sought the warrant at issue in this case. But the affidavit provides almost no detail regarding the controlled buy-it does not state whether the officers observed the buy, identify the type or amount of the controlled substance purchased, indicate how the purchase was initiated, or reveal if the informant had purchased drugs from Christian previously. Nor does the affidavit disclose whether the informant saw large quantities of drugs in Christian's possession or in the Residence. See United States v. Abernathy , 843 F.3d 243, 255 (6th Cir. 2016) (noting that a large quantity of drugs found in a trash can outside of a residence would suggest "repeated and ongoing drug activity in the residence"). Nothing about the January 2015 single controlled buy of an unknown quantity of an unknown drug by an informant with an unknown relationship to Christian suggests that Christian was engaged in protracted or continuous drug trafficking.
In sum, the affidavit reflects only a single purchase from a reliable informant eight months before the search and no other credible evidence of drug activity beyond four prior drug convictions ranging from 4 to 19 years old (the significance of these drug convictions for the probable-cause determination will be discussed in further detail below). I therefore conclude that the affidavit does not establish that Christian was engaged in protracted and continuous drug trafficking. Cf. United States v. Sinclair , 631 F. App'x 344, 348 (6th Cir. 2015) (evaluating the Spikes staleness factors and concluding that the crime at issue was "an ongoing drug trafficking conspiracy" when a confidential informant reported purchasing heroin from the defendant "for several years," and the officers observed the defendant engaging in activity consistent with drug trafficking over the most recent 12 months, with the last observation occurring just 15 days before the search warrant was executed at the defendant's residence); United States v. Greene , 250 F.3d 471, 481 (6th Cir. 2001) (finding protracted and continuous drug trafficking where a reliable confidential informant reported purchasing drugs from the defendant at his residence at least 12 times, the last purchase occurring 23 months before the search warrant executed, because the informant also stated that a package was sent from the residence to a known drug dealer less than one month prior to the execution of the warrant).
3. Third Spikes factor
With regard to the third factor-whether the evidence to be seized is of enduring utility to the holder-the government contends that the warrant sought not just controlled substances, but also records of drug trafficking and firearms used in drug trafficking. These latter two categories of evidence, it argues, are likely to endure, even if controlled substances themselves are not. To support this argument, the government relies on United States v. Burney , 778 F.3d 536 (6th Cir. 2015). But Burney is distinguishable from the present case because there was no dispute that the 17-page affidavit in Burney provided ample *326evidence that the property had been used as a stash house for "a large-scale drug trafficking and money laundering operation- ... a regenerating, enduring criminal enterprise that bears no resemblance to a 'chance encounter in the night.' " Id. at 538, 541-42. Such an extensive operation was likely to involve "scales, weapons, safes, bagging materials, and the like," evidence that was "not readily consumable" and thus unlikely to "be consumed or to disappear." Id. at 541.
In his affidavit in the present case, Officer Bush acknowledged the distinction between those who occasionally sell from their own supply-and thus produce little lasting evidence-and those who sell regularly for profit using extensive networks that likely involve durable evidence like records and firearms. True enough, Officer Bush stated in the affidavit that he was seeking records and firearms related to extensive drug-trafficking operations. But this statement assumes what the affidavit tried and, I believe, ultimately failed to prove by substantial evidence-that Christian was engaged in organized and extensive drug-trafficking operations likely to involve not just controlled substances, but also records and firearms.
Because the government has provided credible evidence of only a single sale of an unknown quantity of a controlled substance in January 2015, rather than "a large-scale drug trafficking and money laundering operation,"see id. , it failed to provide a reason to believe that records of drug trafficking and firearms would be found at the Residence. Whether such records are durable is thus irrelevant. And "because drugs are usually sold and consumed in a prompt fashion," evidence of a single drug sale became stale "very quickly," well before the search was executed eight months later. See United States v. Abernathy , 843 F.3d 243, 250 (6th Cir. 2016) (quoting United States v. Brooks , 594 F.3d 488, 493 (6th Cir. 2010) ); see also United States v. Hython , 443 F.3d 480, 486 (6th Cir. 2006) (noting the limited evidentiary value of an undated controlled buy absent evidence of any recent drug activity at the residence). The third factor thus weighs in favor of finding the evidence of the controlled buy to be stale.
4. Conclusion on staleness
I believe that the first and third Spikes factors control the determination of whether evidence of the controlled buy is stale in this case. Although Christian is entrenched in the community and his residence would be a secure base of operations, the key question is whether evidence of drug activity would be found there at the time of the search. With no reliable evidence of continuous and protracted drug activity, the eight-month-old controlled buy was stale.
This court's decision in Brooks offers strong support for my conclusion. Brooks considered whether an affidavit was sufficient to establish probable cause to search the defendant's residence for evidence of drug crimes. Crucially, the affiant-officer arrested the defendant for aggravated drug trafficking at the defendant's residence and, in the process, smelled marijuana and observed marijuana seeds in plain view. The officer also found $ 1,000 in cash on the defendant after conducting a pat-down search. Later that day, the affiant-officer executed the affidavit in support of the search warrant. This court held that the officer's observations alone were sufficient to support probable cause. Brooks , 594 F.3d at 495.
But the affidavit in Brooks also contained several other pieces of information that, by themselves, were held to be insufficient to establish probable cause. These were: (1) four tips from confidential informants, stating that the defendant was trafficking *327in cocaine, with the tips ranging from one to five years old at the time that the affidavit was executed; (2) a 20-month-old tip from a confidential informant, stating that the defendant was selling cocaine from his residence; and (3) two controlled buys made by a confidential informant almost eight months before the affidavit was executed. The court noted:
[T]here is no question but that this information is stale for purposes of establishing probable cause in its own right. All of the information is regarding drug transactions that took place, at the most recent, approximately six months prior to the date of the affidavit. Given the mobile and quickly consumable nature of narcotics, evidence of drug sales or purchases loses its freshness extremely quickly.
Id. at 493 n.4. Similarly, the single controlled buy conducted in the present case was stale when the warrant was executed eight months later. Cf. United States v. Yates , 501 F. App'x 505, 511 (6th Cir. 2012) (concluding that evidence of a single drug transaction occurring at a residence was not stale when the transaction occurred within ten days of the affidavit's execution); United States v. Pinson , 321 F.3d 558, 565 (6th Cir. 2003) (concluding that evidence of a single controlled purchase was not stale when the warrant was issued three days later).
D. Criminal history
The next matter to be considered is Christian's criminal record. Although "a person's criminal record [demonstrating multiple drug offenses] alone does not justify a search of his or her home[,]" United States v. Payne , 181 F.3d 781, 790-91 (6th Cir. 1999), it is relevant to the probable-cause inquiry, United States v. Berry , 565 F.3d 332, 339 (6th Cir. 2009). The affidavit in question here asserts that Christian, at the time that the warrant was issued, had a 19-year-old conviction for possession of less than 25 grams of cocaine and a 13-year-old conviction for an unspecified second controlled-substance offense. Christian has an additional six-year-old conviction for the delivery/manufacture of marijuana and a four-year-old conviction for the delivery/manufacture of cocaine.
The majority's contention that these convictions support a conclusion that Christian was engaged, at some point, in protracted drug activity is problematic. See Maj. Op. 309. Precedent instructs us to consider "[t]he relative recency of a set of actions and their relative closeness in time to each other." United States v. Perry , 864 F.3d 412, 415 (6th Cir. 2017). These convictions are each several years apart, and even the most recent conviction predates the January 2015 controlled buy by four years. Nothing about these old convictions and the controlled buy is inconsistent with a conclusion that Christian was simply "effectuat[ing] the occasional sale from his or her personal holdings of drugs to known acquaintances." See United States v. Hython , 443 F.3d 480, 485 (6th Cir. 2006).
But even assuming that these convictions, combined with the fact that two search warrants were executed at the Residence in 2009 and 2011, support a conclusion that Christian was engaged in protracted drug sales at the Residence at some point, there is no evidence to suggest that these sales were continuous at the time the warrant was sought and executed in September 2015. See United States v. Helton , 314 F.3d 812, 822 (6th Cir. 2003) (noting that even where "the likely duration of th[e] evidence is relatively long," the evidence may still be stale if enough time has passed between the tip and the execution of the warrant); United States v. Brown , 828 F.3d 375, 384 n.3 (6th Cir. 2016) (concluding that a 12-year-old conviction *328for conspiracy to distribute marijuana was insufficient to establish that an individual was a known drug dealer at the time the warrant was executed).
The key issue is whether a search-warrant affidavit establishes a fair probability that the evidence sought will be found at the place identified at the time the warrant is executed . Hython , 443 F.3d at 485. Emphasizing the temporal requirement of this test, this court found in Hython that "[e]ven had the affidavit stated that from time out of mind, [the residence to be searched] had been a notorious drug den, some recent information would be necessary to eliminate the possibility that a transfer in ownership or a cessation of illegal activity had not taken place." Id. at 486 ; see also United States v. McPhearson , 469 F.3d 518, 524 (6th Cir. 2006) (concluding that the magistrate may "draw the inference that evidence of wrongdoing would be found in the defendants' homes" when the affidavit reflects "the independently corroborated fact that the defendants were known drug dealers at the time the police sought to search their homes").
Neither the majority nor the government has identified any case in which a record of past drug convictions, without recent credible evidence of drug activity, was sufficient to establish that a defendant was engaged in protracted and continuous drug dealing. This court has generally relied on past drug convictions in combination with a defendant's recent drug activity in applying the principle that, "[i]n the case of drug dealers, evidence is likely to be found where the dealers live." United States v. White , 874 F.3d 490, 501 (6th Cir. 2017) (alteration in original) (quoting United States v. Jones , 159 F.3d 969, 975 (6th Cir. 1998) ); see also UnitedStates v. Miggins , 302 F.3d 384, 393 (6th Cir. 2002) (finding probable cause to search the defendant's residence where his criminal record indicated that he had been convicted of cocaine charges and officers observed him signing for a package of cocaine delivered at a second location immediately before the issuance of the warrant).
Absent additional recent reliable evidence, then, old criminal convictions cannot support a finding that drug activity is continuous at the time of the search. Our legal system has long held a strong policy against using propensity evidence to suggest an inference that an individual who has once committed a crime continues to engage in criminal activity. See Fed. R. Evid. 404(b)(1) ("Evidence of a crime, wrong, or other act is not admissible to prove a person's character in order to show that on a particular occasion the person acted in accordance with the character."). Although the Federal Rules of Evidence do not come into play when deciding the validity of a search warrant, the aim of Rule 404 is similar to the purpose of the staleness rule: to ensure that decisionmakers-whether jurors or magistrates-do not improperly assume based on past wrongs that an individual is currently engaging in the specific criminal conduct at issue. See Old Chief v. United States , 519 U.S. 172, 179-82, 117 S.Ct. 644, 136 L.Ed.2d 574 (1997) (discussing the prejudicial nature of propensity evidence). By allowing the government to rely in part on Christian's stale prior convictions, the majority is lowering the probable-cause threshold for former convicts and stripping away their rights guaranteed by the Fourth Amendment.
I therefore conclude that Christian's prior criminal convictions, even when considered with the eight-month-old controlled buy, do not establish that he was engaged in protracted and continuous drug activity. And absent some "independently corroborated fact that the defendant[ ] w[as a] known drug dealer[ ] at the time the police *329sought to search [his] home," probable cause did not exist to search the Residence based on Christian's criminal record. See McPhearson , 469 F.3d at 524.
E. Totality of the circumstances
As discussed above, the caselaw makes clear that the probable-cause determination must be based on the "totality of the circumstances." United States v. Williams , 544 F.3d 683, 686 (6th Cir. 2008). The court should therefore evaluate the weight of the evidence when considered as a whole. At this juncture, I note that the equivalent of only one page of the affidavit is dedicated to facts specifically related to Christian. The majority repeatedly points out that the affidavit is five pages long, see Maj. Op. 307, 312, but the affidavit mostly concerns generic information, including Officer Bush's qualifications and the general nature of drug investigations.
Even though the relevant portion of the affidavit is short and the information contained therein problematic, I recognize that this is not necessarily fatal. For instance, "[w]here recent information corroborates otherwise stale information, probable cause may be found." United States v. Spikes , 158 F.3d 913, 924 (6th Cir. 1998) (alteration in original) (quoting United States v. Henson , 848 F.2d 1374, 1381-82 (6th Cir. 1988) ) (concluding that evidence of drug residue in a residence's trash cans and an officer's recent observations of individuals leaving the residence to sell drugs nearby "refreshed ... otherwise stale information" contained in the affidavit). But no reliable evidence corroborates the stale evidence in the affidavit under review.
In addition, evidence from an informant whose reliability is not known can be corroborated by independent information from police officers. McPhearson , 469 F.3d at 524 n.3 ; United States v. Hammond , 351 F.3d 765, 772 (6th Cir. 2003) (noting the "minimal probative value" of a tip from an informant of unknown reliability, but concluding that "the tip can take on an increased level of significance for probable cause purposes, if corroborated by the police through subsequent investigation"). But there is no evidence that the officers here attempted to corroborate the information provided by the unidentified subjects. And even if the previous controlled buy could be considered to corroborate subsequent information from unidentified sources, the single buy did not corroborate allegations that protracted and continuous drug activity was occurring at the Residence.
In sum, the affidavit shows that (1) two search warrants were executed for drugs at the Residence years ago, (2) Christian has a history of years-old drug convictions, (3) he engaged in one sale of drugs at the Residence eight months prior to the execution of the search warrant, (4) unidentified subjects of unknown reliability reported that Christian was selling drugs at unspecified times in more recent months, and (5) a man with no known connection to Christian was found to be in possession of drugs after leaving "the area" of the Residence on the date of the search-warrant affidavit. This evidence, even under the totality-of-the-circumstances approach, fails to establish a "fair probability" that drug activity was occurring at the Residence at the time the search warrant was executed. See United States v. Brooks , 594 F.3d 488, 492 (6th Cir. 2010).
The majority relies on United States v. Hines , 885 F.3d 919 (6th Cir. 2018), in arguing to the contrary. Maj. Op. 312 (quoting Hines , 885 F.3d at 921-22 ) (" 'Not all search warrant affidavits include the same ingredients,' we said before recognizing that " '[i]t is the mix that courts review to decide whether evidence generated from *330the search may be used or must be suppressed.' " (alteration in original)). But the facts of Hines actually lend further support to my position that the warrant here did not establish probable cause to search the Residence.
The affidavit at issue in Hines contained the following evidence in support of a warrant to search the house in question, owned by Hines's mother: (1) a reliable confidential informant told officers five months prior to the warrant's execution that the defendant was selling large amounts of heroin from the house; (2) a statement from the same informant that he had seen heroin at the house the day before the search; (3) several months of surveillance of the house by law-enforcement officers documented the defendant's comings and goings; (4) a tip from a second reliable confidential informant the day before the warrant's execution stated that he was meeting the defendant at a nearby club to discuss an incoming heroin shipment; (5) officers' observations of the defendant driving "in a manner consistent with narcotics traffickers" to the club at the designated time; (6) statements from the second informant that he had received heroin from the defendant numerous times and was always instructed to meet him at the house to receive that heroin; (7) a tip from the second informant that he was instructed to collect heroin from the defendant at the house on the day that the warrant was executed; (8) three-year-old wiretaps identifying the defendant as a significant heroin trafficker; (9) the two-year-old seizure of $ 33,500 from a third individual outside the house (believed to be payment from Hines for a kilogram of cocaine); and (10) a subsequent statement from this individual that he had previously provided the defendant with heroin and cocaine. Hines , 885 F.3d at 922.
All of the evidence detailed above in Hines directly linked the residence to heroin trafficking at the time of the search through information from reliable informants and specific observations by officers that corroborated the information provided by those informants. And after comparing this evidence with that used to support affidavits in other cases, the court in Hines ultimately concluded that the affidavit at least satisfied the Leon good-faith standard, if not probable cause. Id. at 924-28.
The affidavit at issue here, in contrast, fails to establish anything more than a speculative connection between drug activity and the Residence at the time of the search. This is the "hunch upon hunch" approach found unacceptable in United States v. Valenzuela , 365 F.3d 892, 897 (10th Cir. 2004). Unlike the affidavit in Hines , the affidavit here contains no recent evidence of drug activity at the Residence. The search warrant was therefore not supported by probable cause. Under these circumstances, the deference that would otherwise be due to the issuing magistrate is unjustified. See United States v. Leon , 468 U.S. 897, 914, 104 S.Ct. 3405, 82 L.Ed.2d 677 (1984) ("Deference to the magistrate ... is not boundless."); Massachusetts v. Upton , 466 U.S. 727, 733, 104 S.Ct. 2085, 80 L.Ed.2d 721 (1984) (noting that we apply a "deferential standard of review" to an issuing magistrate's probable-cause determination, but that the determination will not be upheld if the evidence, viewed as a whole, does not provide a "substantial basis" for that determination).
But the majority contends that the Supreme Court's decision in District of Columbia v. Wesby , --- U.S. ----, 138 S. Ct. 577, 199 L.Ed.2d 453 (2018), prevents us from "discounting each item [in the affidavit] one by one." Maj. Op. 312. Similarly, according to the majority, " Hines requires *331us to look holistically at what the affidavit does show, instead of focusing on what the affidavit does not contain, or the flaws of each individual component of the affidavit." Maj. Op. 313. The majority seems to believe that those cases prevent us from assessing the probative value of each bit of material information contained in an affidavit. That is not what the caselaw forbids, and nor should it.
To the contrary, under the totality-of-the-circumstances approach, we assess the probative value of each piece of evidence in the affidavit and then determine whether those pieces of evidence are, as a whole, sufficient to establish probable cause-in other words, we review the "mix" of unique "ingredients" in the affidavit. See Hines , 885 F.3d at 921-22 ; see also Gardenhire v. Schubert , 205 F.3d 303, 315 (6th Cir. 2000) (explaining that, in the context of an arrest, "[p]robable cause determinations involve an examination of all facts and circumstances within an officer's knowledge at the time of an arrest" (quoting Estate of Dietrich v. Burrows , 167 F.3d 1007, 1012 (6th Cir. 1999) )); Valenzuela , 365 F.3d at 897 ("[I]n assessing the totality of the circumstances, a reviewing court must examine the facts individually in their context to determine whether rational inferences can be drawn from them that support a probable cause determination." (citations and internal quotation marks omitted)).
The D.C. Circuit's error in Wesby was that it "viewed each fact 'in isolation, rather than as a factor in the totality of the circumstances.' " Wesby , 138 S. Ct. at 588 (quoting Maryland v. Pringle , 540 U.S. 366, 372 n.2, 124 S.Ct. 795, 157 L.Ed.2d 769 (2003) ). Specifically, the Court of Appeals erred by engaging in a "divide-and-conquer analysis," id. (quoting United States v. Arvizu , 534 U.S. 266, 274, 122 S.Ct. 744, 151 L.Ed.2d 740 (2002) ), whereby it dismissed facts that were "not sufficient standing alone to create probable cause," id. (emphasis in original). I have not done that here. Instead, I have considered each fact "as a factor in the totality of the circumstances," but nonetheless conclude that the affidavit was insufficient to establish probable cause. See id. (quoting Pringle , 540 U.S. at 372 n.2, 124 S.Ct. 795 ).
The majority's approach, on the other hand, is problematic because it contains "inferences drawn upon inferences." See United States v. Laughton , 409 F.3d 744, 750 (6th Cir. 2005). It infers that the officers saw Thomas leave the Residence, that the heroin found in Thomas's vehicle was connected to the Residence, that tips from unidentified informants are reliable without any indicia of credibility and without corroboration, and that a stale controlled buy and old criminal convictions establish that Christian was engaged in continuous and protracted drug activity at the time of the search. The totality-of-the-circumstances test is not a license for the majority to list problematic evidence, stacking inference upon inference, and contend in a conclusory manner that "taken together [the affidavit] point[s] clearly to one conclusion: that Christian was dealing drugs from 618 Grandville." Maj. Op. 311. Rather, the Supreme Court in Wesby instructed lower courts to view each fact "as a factor in the totality of the circumstances" and to "consider 'the whole picture.' " See Wesby , 138 S. Ct. at 588 (citations omitted). This requires us to explain how individual pieces of evidence corroborate one another, which the majority has failed to do.
II. LEON GOOD-FAITH EXCEPTION
This brings me to the Leon good-faith exception. Under the Leon good-faith standard, suppression should be limited to "circumstances in which the benefits of police deterrence outweigh the heavy costs of *332excluding 'inherently trustworthy tangible evidence' from the jury's consideration." United States v. White , 874 F.3d 490, 496 (6th Cir. 2017) (quoting United States v. Leon , 468 U.S. 897, 907, 104 S.Ct. 3405, 82 L.Ed.2d 677 (1984) ). The test is "whether a reasonably well trained officer would have known that the search was illegal despite the magistrate's decision." Id. (quoting United States v. Hodson , 543 F.3d 286, 293 (6th Cir. 2008) ).
Four situations have been identified by the Supreme Court in which an officer could not reasonably believe that a search was valid, despite the issuance of a warrant. See Laughton , 409 F.3d at 748 (citing Leon , 468 U.S. at 914-23, 104 S.Ct. 3405 ). One of those is where the affidavit is "so lacking in indicia of probable cause as to render official belief in its existence entirely unreasonable." Leon , 468 U.S. at 923, 104 S.Ct. 3405 (quoting Brown v. Illinois , 422 U.S. 590, 610-11, 95 S.Ct. 2254, 45 L.Ed.2d 416 (1975) (Powell, J., concurring in part)). Such an affidavit has been characterized as "bare bones." Id. at 915, 926, 104 S.Ct. 3405. A "bare bones affidavit is one that merely 'states suspicions, beliefs, or conclusions, without providing some underlying factual circumstances regarding veracity, reliability, and basis of knowledge.' " United States v. McPhearson , 469 F.3d 518, 526 (6th Cir. 2006) (quoting United States v. Weaver , 99 F.3d 1372, 1378 (6th Cir. 1996) ). In contrast, a sufficient affidavit must contain some "particularized facts that indicate veracity, reliability, and basis of knowledge and go beyond bare conclusions and suppositions." Id.
This court held in Laughton "that a determination of good-faith reliance, like a determination of probable cause, must be bound by the four corners of the affidavit." Laughton , 409 F.3d at 751. Judge Thapar, concurring in the present case, now recommends that we overrule Laughton . But I believe that Laughton correctly decided that "the good faith exception to the exclusionary rule does not permit consideration of information known to a police officer, but not included in the affidavit, in determining whether an objectively reasonable officer would have relied on the warrant." Id. at 752.
The test for good-faith reliance is an objective one. Leon , 468 U.S. at 919 n.20, 104 S.Ct. 3405. In Leon , the Supreme Court reasoned that "sending state and federal courts on an expedition into the minds of police officers would produce a grave and fruitless misallocation of judicial resources." Id. at 922 n.23, 104 S.Ct. 3405 (quoting Massachusetts v. Painten , 389 U.S. 560, 565, 88 S.Ct. 660, 19 L.Ed.2d 770 (1968) (White, J., dissenting)). This court held in Laughton that the same reasoning applies to the issue of considering information outside of the affidavit "in determining whether an objectively reasonable officer would have relied on the warrant." Laughton , 409 F.3d at 752. Allowing courts to consider such extrinsic information would "lead to the very kind of subjectivity that the Supreme Court has repeatedly and explicitly rejected." Id. Future cases would require courts to engage in the subjective and time-consuming inquiry of "determin[ing] not only how much affiants knew, but also when and from whom they learned it." See id.
Judge Thapar disagrees, citing Herring v. United States , 555 U.S. 135, 129 S.Ct. 695, 172 L.Ed.2d 496 (2009), for the proposition that courts already consider "a particular officer's knowledge and experience, but that does not make the test any more subjective than the one for probable cause, which looks to an officer's knowledge and experience, but not his subjective intent." Id. at 145-46, 129 S.Ct. 695 (internal citations omitted). But the Supreme Court in Herring was noting only that we should *333consider a police officer's general background knowledge and experience. In support of this proposition, the Court cited Ornelas v. United States , 517 U.S. 690, 116 S.Ct. 1657, 134 L.Ed.2d 911 (1996), in which Chief Justice Rehnquist observed that "a police officer views the facts through the lens of his police experience and expertise" and that "a police officer may draw inferences based on his own experience in deciding whether probable cause exists." Id. at 699-700, 116 S.Ct. 1657. Considering an officer's general background knowledge and experience is an entirely different inquiry from considering what relevant facts were known to the officer at the time of the search. The latter requires us to "inquir[e] into the subjective awareness of arresting officers," see Herring , 555 U.S. at 145, 129 S.Ct. 695, whereas the former does not.
Furthermore, " Leon ... make[s] clear that the relevant question is whether the officer reasonably believed that the warrant was properly issued, not whether probable cause existed in fact." United States v. Carpenter , 360 F.3d 591, 598 (6th Cir. 2004) (en banc) (Gilman, J., concurring) (emphasis in original). Information extrinsic to the affidavit and not presented to the magistrate is not relevant to the inquiry of whether the officer reasonably believed that the warrant was properly issued. Accordingly, I find no basis to conclude that this court's decision in Laughton is inconsistent with Supreme Court precedent.
Applying the above principles, I acknowledge that whether the good-faith standard is met in this case is a close call. But I ultimately conclude that the affidavit falls short because it does not provide any particularized facts connecting the Residence to drug activity at the time that the search warrant was executed.
The majority, in concluding otherwise, argues that our decision in United States v. Hython , 443 F.3d 480 (6th Cir. 2006), "is almost completely inapposite here," Maj. Op. 313. I completely disagree. In Hython , the affidavit contained information that the officers had, at some unidentified point, conducted a controlled buy of crack cocaine at the residence to be searched. But because "the affidavit include[d] no observation of deliveries to the address, no monitoring of the frequency or volume of visitors to the house, no second controlled buy, [and] no further surveillance whatsoever," "the affidavit [wa]s patently insufficient" to allow a reasonable officer to believe that the affidavit established probable cause to search the residence. Hython , 443 F.3d at 486, 488-89.
Similarly, the affidavit in the present case primarily relies on a single stale controlled buy to link the Residence to drug activity at the time of the search. It does not provide any credible evidence that drug activity continued at the Residence in the eight-month interim, and the single instance of contemporary surveillance did not link the Residence to drug activity by anything more than speculation that Thomas purchased drugs at the Residence. This court's decision in Hython is thus very much on point with regard to the Leon good-faith issue.
Moreover, this court has held that the Leon good-faith standard was not satisfied where "the 'evidence in the affidavit connecting the crime to the residence [wa]s so vague as to be conclusory or meaningless.' " McPhearson , 469 F.3d at 527 (quoting United States v. Frazier , 423 F.3d 526, 537 (6th Cir. 2005) ) (finding that the Leon standard was not satisfied where the affidavit reflected that officers had arrested the defendant at his residence on an assault charge and found him in possession of cocaine, but where there was no evidence *334connecting the defendant or his residence to drug trafficking); see also United States v. Brown , 828 F.3d 375, 384-86 (6th Cir. 2016) (holding that, in the search of a residence, the Leon good-faith standard was not satisfied despite the affidavit's allegations that the defendant was arrested for attempting to deliver heroin 21 days prior to the search, a drug dog had alerted to the odor of narcotics in the defendant's car, the defendant exchanged text messages discussing drug prices, and the defendant had a 12-year-old conviction for conspiracy to distribute marijuana).
In contrast, this court has held that an affidavit was sufficient to satisfy the Leon good-faith standard where the affidavit provided a material link between the criminal activity alleged and the residence in question at the time of the search. See United States v. White , 874 F.3d 490, 494, 498 (6th Cir. 2017) (finding the Leon good-faith standard satisfied where the affiant stated that officers received a tip that the defendant, who had an extensive criminal history involving drugs, was selling drugs from his residence, and the officers initiated, observed, and recorded a controlled buy from the defendant in the driveway of the residence less than 72 hours before the affidavit was executed); United States v. Higgins , 557 F.3d 381, 391 (6th Cir. 2009) (concluding that the affidavit met the Leon good-faith standard where it stated that a named informant told officers that he had purchased drugs from the defendant's residence earlier that day); Frazier , 423 F.3d at 536 (finding the Leon good-faith standard satisfied when the affidavit reflected that two recorded controlled buys were conducted by an informant at the defendant's previous residence seven months before the search, that drugs were found at the defendant's previous residence two months before the search, that a named informant reported buying two pounds of marijuana from the defendant weekly, and that phone records showed that the defendant was in constant contact with known drug dealers); Carpenter , 360 F.3d at 593 (finding the Leon standard satisfied where the affidavit supporting a warrant to search the residence alleged that a police officer conducting an aerial search spotted numerous marijuana plants directly connected by a road to the residence).
Unlike the affidavit evidence considered in White , Higgins , Frazier , and Carpenter , Christian's criminal history and the January 2015 controlled buy do not establish a nexus between the Residence and drug activity at the time of the search. Such a nexus is required for the Leon good-faith exception to apply. See Hython , 443 F.3d at 488-89.
And although closer in time to the execution of the search, the information received from the unidentified subjects indicating that Christian was engaged in large-scale drug trafficking from the Residence was "so vague as to be conclusory or meaningless." See Frazier , 423 F.3d at 536 (quoting Carpenter , 360 F.3d at 596 ). Where statements "are heavily discounted due to their minimal trustworthiness and reliability, they add little to the probable cause determination" and, accordingly, "a reasonable officer would recognize that without more corroboration, the ... affidavit came well short of establishing probable cause." United States v. Helton , 314 F.3d 812, 825 (6th Cir. 2003).
An investigation by law-enforcement officers can corroborate tips of unknown reliability. But the observation of Thomas "walk[ing] away from the area" of the Residence before he was later found with heroin in his vehicle does not provide this additional corroboration. At best, the observation allows for only speculation that Thomas purchased the drugs from the Residence. Such speculation cannot reasonably *335be thought to support a finding of probable cause. See White , 874 F.3d at 498 (noting that a bare-bones affidavit is one that contains "a mere affirmation of suspicion and belief without any statement of adequate supporting facts" (quoting Nathanson v. United States , 290 U.S. 41, 46, 54 S.Ct. 11, 78 L.Ed. 159 (1933) )). As a result, I conclude that no reasonable officer would have believed that the affidavit established probable cause to search the Residence at the time the affidavit was executed.
I also believe that my conclusion is in line with the policy behind the Leon good-faith exception to the exclusionary rule. The majority argues that "this is a case in the very heartland of the Leon exception," and that "[t]his is a particularly egregious case to misapply the good-faith exception given the utter lack of police wrongdoing." Maj. Op. 312, 313. I respectfully disagree. This court in United States v. McClain , 444 F.3d 556 (6th Cir. 2005), held that the Leon exception applied because "[t]here was indeed nothing more that [the officer] 'could have or should have done under these circumstances to be sure his search would be legal.' " Id. at 566 (quoting United States v. Thomas , 757 F.2d 1359, 1368 (2d Cir. 1985) ).
In the present case, however, the officers could have and should have done a lot more. There is no evidence in the affidavit that they engaged in ongoing or repeated surveillance, arranged subsequent controlled buys, or otherwise monitored for "hallmarks of drug dealing" at the Residence. See United States v. McPhearson , 469 F.3d 518, 527 (6th Cir. 2006) ; Hython , 443 F.3d at 486, 488-89. Suppressing the evidence in this case would incentivize police officers to take such actions to corroborate unreliable information and describe such actions in their search-warrant applications. Instead, the majority allows the Leon good-faith exception, which was intended to be applied only in "unique cases," see McClain , 444 F.3d at 565, to excuse sloppy police work.
III. TELEPHONE-CALL EVIDENCE
I will now briefly discuss Christian's challenge to the district court's decision to admit the recorded telephone call between Thomas and Thomas's girlfriend, Tanisha Edwards. The majority concludes that, if the district erred in admitting this evidence, the error was harmless. Maj. Op. 313-14. But the majority's determination relies on its conclusion affirming the district court's decision to admit the evidence obtained in accordance with the search warrant. Because I disagree and conclude that the court erred in admitting the evidence obtained pursuant to the warrant, I do not consider the court's error regarding the phone-call evidence harmless.
IV. CONCULSION
In sum, considering the totality of the circumstances, I believe that the information in the affidavit fails to establish probable cause and that the Leon good-faith exception does not apply. I would therefore hold that the district court erred in denying Christian's motion to suppress the evidence obtained pursuant to the search warrant. Accordingly, I respectfully dissent.
In a perfect world, Laughton would not be as problematic because the good-faith exception would apply unless the affidavit was skeletal. Yet courts have extended the "bare bones" exception to good faith well beyond Supreme Court precedent. This case is a perfect example. Leon itself said that an affidavit supports good faith when it "provide[s] evidence sufficient to create disagreement among thoughtful and competent judges as to the existence of probable cause." 468 U.S. at 926, 104 S.Ct. 3405. That is what we have here, as the dueling majority and dissent opinions show. But despite this thoughtful disagreement, the dissent continues further to say that the affidavit cannot even satisfy good faith.