In the

    United States Court of Appeals
                 For the Seventh Circuit
                     ____________________
No. 18-2604
UNITED STATES OF AMERICA,
                                                   Plaintiff-Appellee,
                                 v.

MICHAEL CLARK,
                                               Defendant-Appellant.
                     ____________________

         Appeal from the United States District Court for the
                   Western District of Wisconsin.
          No. 3:17-cr-53-JDP-1 — James D. Peterson, Judge.
                     ____________________

     ARGUED MAY 22, 2019 — DECIDED AUGUST 15, 2019
                ____________________

   Before BAUER, HAMILTON, and SCUDDER, Circuit Judges.
   HAMILTON, Circuit Judge. Defendant-appellant Michael
Clark was convicted of possessing a mixture containing fen-
tanyl in violation of 21 U.S.C. § 841(a)(1). Clark had been
found in a hotel room with more than 80 grams of a mixture
of heroin and fentanyl, a digital scale, and cellophane bags.
He does not appeal any aspect of his jury trial, but he chal-
lenges the denial of his motion for a Franks hearing challeng-
ing the issuance of the search warrant for the hotel room. He
2                                                   No. 18-2604

also challenges the denial of his motion to suppress without
an evidentiary hearing. And he challenges two aspects of his
sentence: the guideline treatment of his conviction for drug
distribution that occurred in Illinois seven months after his
Wisconsin arrest and one condition of supervised release. We
vacate Clark’s conviction and remand for an evidentiary hear-
ing on his Franks challenge. We affirm on the denial of his mo-
tion to suppress without a hearing. We also affirm on the
guideline issue and determine that the supervised release
challenge was waived. We address in Part I the need for a
Franks hearing and in Part II the need for an evidentiary hear-
ing on the motion to suppress. We address the sentencing is-
sues in Part III.
I. The Need for a Franks Hearing
    The Fourth Amendment’s strong preference for the use of
search warrants calls for probable cause determinations by a
“neutral and detached magistrate” as opposed to “officer[s]
engaged in the often competitive enterprise of ferreting out
crime.” Johnson v. United States, 

, 14 (1948). The ap-
plication for a warrant “must provide the magistrate with a
substantial basis for determining the existence of probable
cause.” Illinois v. Gates, 

, 239 (1983).
    The ability of the neutral and detached magistrate to de-
termine probable cause depends on the accuracy of the infor-
mation the police submit. “[A] search warrant is not valid if
the police obtain it by deliberately or recklessly presenting
false, material information,” or by omitting material infor-
mation from the affidavit provided to the issuing judge.
United States v. McMurtrey, 

, 508 (7th Cir. 2013),
citing Franks v. Delaware, 

, 15556 (1978). To inval-
idate a warrant on this basis, a defendant at a so-called Franks
No. 18-2604                                                     3

hearing must prove by a preponderance of the evidence either
falsity or recklessness, as well as materiality. 

.
    Merely to obtain a Franks hearing, however, a defendant
does not need to prove the Franks violation. A defendant must
only make a substantial preliminary showing (1) that the war-
rant application contained a material falsity or omission that
would alter the issuing judge’s probable cause determination,
and (2) that the affiant included the material falsity or omitted
information intentionally or with a reckless disregard for the
truth. United States v. Glover, 

, 820 (7th Cir. 2014);
see also, e.g., United States v. Hancock, 

, 708 (7th
Cir. 2016); United States v. Mullins, 

, 861–62 (7th
Cir. 2015); United States v. Robinson, 

, 887–88 (7th
Cir. 2008). “Proof by a preponderance of the evidence is not
required until the Franks hearing itself.” 

.
    Clark asserted in the district court that the police investi-
gator who applied for the search warrant of the hotel room
deliberately or recklessly omitted critical information affect-
ing the credibility of the unidentified informant who told the
officer about drug distribution at the hotel where Clark was
arrested. The district court denied the motion for an eviden-
tiary hearing on the question. The court agreed that the police
had provided no information about the informant’s credibil-
ity. The court found, however, that the police had provided
sufficient corroboration for the informant’s tip so that the war-
rant did not depend on the informant’s credibility. That
meant the omitted credibility information was not material
for Franks purposes. We disagree and find that a hearing is
needed. “[W]e review the denial of a Franks hearing for clear
4                                                   No. 18-2604

error, but any legal determinations that factored into the rul-
ing are reviewed de novo.” 

; see also Han-

08.
    A. The Warrant Application
    Investigator Todd Maas is a police officer in Superior, Wis-
consin. He prepared the warrant application and signed the
supporting affidavit. Maas said that a confidential informant
contacted him on October 14, 2015 and told him that earlier
that day, he had driven someone to a parking lot adjacent to
the Baywalk Inn in Superior to buy heroin from a black male
called “Big Mike,” the brother of “Toonchie.” Maas said he
and another officer then performed their own investigations,
including surveillance of the parking lot. Maas observed a
black male leave the hotel and enter and then exit at least five
cars in the hotel parking lot. He also learned that the guest
staying in Room 203 was the only hotel guest who both had
paid in cash and was staying only one night, all behavior that
Maas said was typical of drug trafficking, based on his train-
ing and experience. Maas also said he had spoken to a woman
(referred to in this case as the “mom on a mission”) who said
that her daughter was a heroin addict and that she (the
mother) had followed a man she suspected of drug-dealing to
Room 203.
    Maas included all of this information in his affidavit,
which convinced a state trial judge to issue a search warrant
for Room 203. Maas did not include any damaging infor-
mation about the credibility of his confidential informant,
who was the only source of information specifically about
drug trafficking. The informant was being paid for his ser-
vices. He also had two pending criminal charges against him,
fifteen prior convictions, and a history of opiate and cocaine
No. 18-2604                                                   5

abuse, and he was hoping to receive a reduced sentence in ex-
change for his cooperation.
   B. Materiality
    Where an affidavit is based primarily on tips from an in-
formant, probable cause can be shown based on the totality of
the circumstances. 

; 

. In cases based on informants’ tips, we have identified five
factors of particular relevance: (1) the level of detail the in-
formant provided; (2) the extent to which the informant’s in-
formation is based on his or her own first-hand observations;
(3) the degree to which police have corroborated the inform-
ant’s information; (4) the time elapsed between the events re-
ported and the warrant application; (5) and “whether the in-
formant appeared or testified before the magistrate.” 

, citing United States v. Johnson, 

,
600 (7th Cir. 2011). Information omitted from a warrant appli-
cation is material when its omission affects the probable cause
determination. 

.
    In performing this probable cause inquiry, courts must
keep in mind that a search warrant for a home, business, or
even hotel room can authorize highly intrusive and even de-
structive actions by the police, and that informants, especially
in drug cases, can be unreliable and motivated by rivalries or
revenge. United States v. Lopez, 

, 478 (7th Cir.
2018), quoting Terry v. Ohio, 

, 1617 (1968) (“What
we blandly call ‘Terry stops’ can be highly intrusive.” They are
“not just ‘a “petty indignity,”’ wrote the Supreme Court, but
‘a serious intrusion upon the sanctity of a person, which may
inflict great indignity and arouse strong resentment, and it is
not to be undertaken lightly’”); United States v. Bell, 

, 1050 (7th Cir. 2009) (“For all we know, [the informant]
6                                                   No. 18-2604

could have been a rival drug dealer, an angry customer, or
had some other beef” with the defendant).
    Our cases do not hold that a Franks hearing is required
every time some substantial adverse information about an in-
formant’s credibility is omitted from a probable cause affida-
vit. 

. We have said generally that
where a warrant is obtained based on an informant’s tip, “in-
formation about the informant’s credibility or potential bias
is crucial.” 

; see, e.g., United States v.
Bradford, 

, 503–04 (7th Cir. 2018). That being said,
we also have upheld warrants tainted by police omission of
adverse informant credibility information. Our Franks hearing
cases show that when police have sufficiently corroborated an
informant’s tip, the omission of facts pertaining to the inform-
ant’s credibility may not be material.
   In United States v. Musgraves, for example, the police omit-
ted from the affidavit important and damaging credibility in-
formation about one informant. 

, 45960 (7th Cir.
2016). We affirmed the denial of a Franks hearing because a
second affidavit included information from a second inform-
ant without credibility problems who had provided the police
with specific and timely information about the suspect’s drug
trafficking to support the warrant. 

   In United States v. Bradford, we also held that the affidavit’s
omission of facts damaging to an informant’s credibility did
not require suppression of evidence seized with a search war-
rant for drug and weapons 

05. In
Bradford, the application failed to disclose that the informant
had three felony convictions, was on probation, and was be-
ing paid for his help. 

But his information was fresh,
specific, and corroborated by his having carried out
No. 18-2604                                                   7

controlled drug buys from the target and by specific infor-
mation from another informant without his credibility prob-
lems. 

    In United States v. Hancock, we affirmed the denial of a
Franks hearing in similar 

. An
informant provided fresh, detailed information about a sus-
pect’s drug sales, but the affidavit for the warrant left out in-
formation about the informant’s own criminal history and the
fact that he was in custody when he provided the information.

We found that the omissions were not material
because the warrant application included extensive corrobo-
ration of the suspect’s drug trafficking and intimidation of
witnesses. That corroboration came from other informants, a
search of a cellular telephone showing texts referring to drug
deals, and from an earlier interview with the suspect himself.
The quantity and quality of corroborating information meant
that the omitted adverse credibility information about one in-
formant did not require a Franks hearing. 

    This case is readily distinguishable. This warrant applica-
tion did not include any of the substantial adverse information
Maas had about the informant’s credibility. The government
does not try to justify Maas’s omissions, but it argues that the
informant’s credibility was not material to the warrant appli-
cation because Maas and his colleague investigated and cor-
roborated the informant’s tip sufficiently to provide probable
cause independent of that tip. We disagree.
    The corroboration in this case was much weaker than in
the cases where we upheld warrants’ validity in the face of
credibility omissions. This case is more akin to Glover, where
the warrant application depended so heavily on the credibil-
ity of the informant. Here, the police had no controlled buys.
8                                                   No. 18-2604

They never saw money or drugs change hands. Even the in-
formant did not claim to have seen drugs or money change
hands. If his tip did not pan out, he had plenty of deniability.
He claimed only that his passenger had met with “Big Mike”
outside his presence to buy heroin. And the “mom on a mis-
sion” had no track record of credibility and provided no spe-
cifics to support her suspicions of the man she had followed
to Room 203. In sum, the foundation for probable cause inde-
pendent of the credibility of the informant was so meager in
this case that the credibility of the informant was material for
Franks purposes.
    C. Deliberate or Reckless Omissions?
    To obtain a Franks hearing, Clark needed to “offer direct
evidence of [Maas’s] state of mind or circumstantial evidence
that [Maas] had a subjective intent to deceive based on the na-
ture of [his] omissions.” 

. In his report
and recommendation to the district court, we must note, the
magistrate judge in Clark’s case incorrectly read our decision
in Glover to “mandate that … court[s] infer recklessness from
the fact that an officer omitted known and substantial adverse
information from a search warrant affidavit….” That is not
correct. In Glover, we held that “credibility omissions them-
selves, even in the absence of more direct evidence of the of-
ficer’s state of mind, provide sufficient circumstantial evi-
dence to support a reasonable and thus permissible inference
of reckless disregard for the truth.” 

In
other words, credibility omissions do not require courts to in-
fer recklessness. Such omissions are matters for a trier of fact
to weigh, just as when a witness testifies incorrectly or falsely
about one or a few matters, and the trier of fact must weigh
the rest of the witness’s testimony.
No. 18-2604                                                    9

    In this case, the questions of materiality and the police af-
fiant’s state of mind are intertwined. To the extent that cor-
roboration might have avoided the need for a Franks hearing
in this case, the reliability of the corroboration all depended,
from the issuing judge’s point of view, on the same officer
whose credibility is at issue: Maas omitted all adverse infor-
mation he had about the credibility of the informant who pro-
vided the most specific (but still second-hand) information
about drug trafficking. As noted, the omission of so much im-
portant information permits (but does not require) an infer-
ence that the omissions were deliberate or reckless. If the
showing of probable cause in the warrant application de-
pended on the credibility of the informant, that permissible
inference should be enough to obtain a Franks hearing.
    The district court’s and government’s rationale for deny-
ing a Franks hearing is that Maas’s own observations supply
probable cause, independent of the informant. We assume
that if Maas’s own observations are credited, they could lead
a reasonable judge to find probable cause, though the appli-
cation would have been much weaker without the inform-
ant’s tip to Maas. The problem is that Maas’s credibility can
legitimately be questioned here based on what could have
been deliberate or reckless omissions of information about the
credibility of the informant.
    This is where materiality and intent become intertwined.
If the omissions were deliberate or reckless, it would not be
unreasonable to take seriously the prospect that other por-
tions of the same warrant application may have been deliber-
ately or recklessly false. Judges often instruct jurors that if
they believe a witness has lied intentionally about a material
matter, they may (but are not required to) discount the
10                                                  No. 18-2604

witness’s testimony on other matters. See United States v.
Weinstein, 

, 713–14 (2d Cir. 1971) (no positive rule
of law excludes witness’s testimony entirely based on willful
falsehood, but such falsification presents issue for jury to con-
sider in weighing rest of testimony), quoting Knowles v. People,

, 412 (1867); Kevin F. O’Malley, Jay E, Grenig, and
William C. Lee, 1A Fed. Jury Prac. & Instr. §§ 15:0106 (6th ed.
2019).
    The same point is at the heart of the Supreme Court’s ju-
risprudence on impeachment evidence in Brady cases. Brady
v. Maryland, 

(1963), and its progeny establish that
government-suppressed evidence is material “if there is a rea-
sonable probability that, had the evidence been disclosed, the
result of the proceeding would have been different.’” Kyles v.
Whitely, 

, 43334 (1995), quoting United States v.
Bagley, 

, 682 (1985). Strong, non-cumulative evi-
dence that undermines the credibility of an important govern-
ment witness satisfies that definition. Sims v. Hyatte, 

, 1087 (7th Cir. 2019). The reason is simple. Evidence in-
dicating that a witness whose testimony “may well be deter-
minative of guilt or innocence” is unreliable might prompt a
jury to distrust everything the witness says. Giglio v. United
States, 

, 154 (1972). The same logic applies in this
context. Maas’s potentially intentional or reckless omission
compromises his credibility, which determines whether prob-
able cause for the warrant existed or not.
   To be clear, we are not saying or finding here that Maas
was in fact deliberately or recklessly deceptive. The issue on
appeal is only whether Clark was entitled to a hearing, one
where Maas’s credibility could be addressed with evidence
from both sides. The informant was the only source of
No. 18-2604                                                                11

information (and again, that was only second-hand) who said
the person in Room 203 was actually dealing drugs. All of the
other information in the warrant application indicated
grounds for suspicion, but we can assume that Maas had a
good reason for including the informant’s more specific infor-
mation. The complete omission of the available damaging in-
formation about the informant’s credibility permits an infer-
ence that Maas was not being honest and careful with the is-
suing court. Whether he in fact was honest and careful about
the facts is an appropriate subject for the Franks hearing on
remand.1
II. The Timing of the Warrant and Search
   Defendant Clark was in Room 203 when police executed
the warrant. Also in the room were cellophane bags, a digital
scale, and approximately 82 grams of a mixture of heroin and
fentanyl. The police photographed the hotel room during
their search. In a photograph of suspected drugs on a
nightstand, a digital clock is visible, reading 8:15 p.m. If the
clock was correct, the search was executed at least 25 minutes
before the judge issued the warrant.


    1 Maas’s application also did not explain why the informant would
know heroin if he saw it. See United States v. Peck, 

, 75657 (7th
Cir. 2003) (no probable cause for search warrant where affidavit was based
on informant’s report of seeing drugs in defendant’s house, but according
to affidavit, informant failed to give details about specific location and
amount of drugs, or any explanation as to how informant knew that what
she saw were controlled substances). Nor did Maas provide the kind of
detail about his own training and experience that is needed to justify reli-
ance upon them. See generally, e.g., United States v. Scott, 

, 845
(7th Cir. 2018) (“Details … are vital when an officer proposes his own
training and experience as the basis of a warrant”).
12                                                            No. 18-2604

    The police of course are not permitted to rely on a search
warrant that has not yet been approved by a judge. The dis-
trict court denied Clark’s motion to suppress without a hear-
ing. The issue here is whether the photograph of the clock was
sufficient evidence for Clark to have an evidentiary hearing
on his motion to suppress.
    A defendant who requests a suppression hearing must
present “definite, specific, detailed, and nonconjectural facts
… demonstrat[ing] that there is a disputed material issue of
fact.” United States v. Rodriguez, 

, 141 (7th Cir. 1995);
see United States v. Curlin, 

, 564 (7th Cir. 2011).
“Reliance on vague, conclusory allegations is insufficient.”
United States v. Randle, 

, 1212 (7th Cir. 1992).
    In response to Clark’s motion to suppress, the government
submitted the “metadata” from the digital camera. The
metadata showed a host of data about each photograph, in-
cluding the time each was taken. The metadata showed that
all of the search photographs were taken between 10:36 and
10:52 p.m., well after the judge issued the search warrant.
Clark responded in turn not with an affidavit from any wit-
ness to the search but with results of Google searches about
how to edit metadata for photographs. In light of the
metadata, and in the absence of any specific indications that
data had been altered, the district court decided that there was
no need for a hearing and denied Clark’s motion to suppress.
   The situation here is similar to United States v. Woods, 

(7th Cir. 1993).2 In Woods, the police conducted a

     2 Another portion of the Woods opinion was abrogated by the Supreme

Court’s decision in Bailey v. United States, 

(1995), as we noted
in United States v. Monroe, 

, 133 (7th Cir. 1995).
No. 18-2604                                                 13

search on February 12th. The problem was that both the war-
rant and the supporting affidavit bore handwritten dates of
February 14th, though both were also date-stamped February
11th. The defendant moved to suppress, arguing that the
search had been executed two days before the warrant was
issued. The government responded with an affidavit from a
prosecutor testifying that the application and warrant were
both prepared and signed on February 11th and that the Feb-
ruary 14th dates were mistakes. Using language associated
with civil motions for summary judgment—“disputed issues
of material fact” and “genuine factual dispute”—we affirmed
the district court’s decision to deny the motion to suppress
without a hearing. Because the defendant had failed to re-
spond to the government’s explanation, we agreed that he
had failed to show a genuine factual dispute and a need for
an evidentiary hearing. 

    One might argue that in Woods the government submitted
an affidavit to explain the discrepancy, while the government
here submitted letters and copies of documents, without any
affidavits. Both parties here submitted various documents to
the district court with this less formal approach. But the bur-
den is on the defendant to support his motion to suppress.
Missing from this record is an assertion by any witness—de-
fendant Clark or anyone else—to the effect that the search ac-
tually occurred at 8:15 p.m. and that the clock was correct.
Without such an assertion, whether in an affidavit or other-
wise (given the less formal procedures being used in the dis-
trict court), the Google search results about how to alter
metadata merely invite speculation. The district court did not
err by denying the motion to suppress, to the extent it was
based on the photograph of the clock, without a hearing.
14                                                 No. 18-2604

III. Sentencing Issues
     A. Relevant Conduct Under the Guidelines
    Clark also challenges two aspects of his sentence. The first
is a Sentencing Guidelines issue concerning the scope of his
relevant conduct and criminal history, which are two sides of
the same coin here. The U.S. Probation Office prepared a Pre-
Sentence Report for Clark. Pursuant to U.S.S.G. § 2D1.1(a)(5)
and (c)(8) of the 2016 Sentencing Guidelines Manual, the PSR
calculated Clark’s base offense level as 24 because he pos-
sessed between 40 and 160 grams of a substance containing
fentanyl. No other adjustments applied, so his total offense
level remained 24. The PSR also determined that Clark had 11
criminal history points and was in criminal history category
V. Three of those 11 points were from a 2017 federal convic-
tion for drug distribution stemming from events that occurred
in Illinois after Clark was arrested in this Wisconsin federal
case. Clark also received three points for a prior felony drug
conviction in Minnesota. Because of the Minnesota convic-
tion, Clark faced a 120-month mandatory minimum pursuant
to 21 U.S.C. § 851. (If not for the enhancement, Clark would
have had an advisory guideline range of 92 to 115 months.)
    Both the government and Clark argued for the 120-month
mandatory minimum sentence. The district court adopted the
facts of the PSR and determined that the PSR had calculated
correctly Clark’s offense level, criminal history, and guideline
range. The court sentenced Clark to 120 months in prison to
be served consecutively to the 71-month sentence he had re-
ceived in the Illinois case.
    On appeal, Clark argues that his later drug offense in the
Illinois conviction should be treated under the Sentencing
No. 18-2604                                                    15

Guidelines as “relevant conduct” for this case rather than as
part of his criminal history. That means, he contends, his 120-
month sentence in this case should run concurrently with the
71-month sentence he received in Illinois. Clark did not raise
this argument in the district court. At sentencing, the govern-
ment asked that Clark’s sentence run consecutively to his Illi-
nois sentence. Clark did not object. He requested only that
this sentence run concurrently with any state sentence. The
government argues that Clark therefore waived this issue.
    Waiver is the intentional relinquishment of a known right,
whereas forfeiture occurs when a defendant “negligently fails
to assert a right in a timely fashion.” United States v. Brodie,

, 530 (7th Cir. 2007); see United States v. Olano, 

, 733 (1993). Concurrent sentences would reduce sig-
nificantly the time Clark spends in prison. There is no reason
for us to draw the inference here that Clark intentionally
waived this argument. “Waiver principles should be con-
strued liberally in favor of the defendant,” and we are hesi-
tant, without more, to construe a defendant’s failure to object
to a PSR as an intentional relinquishment of a known right.
United States v. Jaimes-Jaimes, 

, 84849 (7th Cir.
2005). This is especially the case where, as here, the defendant
had no strategic reason to forgo this sentencing argument. See
United States v. Jenkins, 

, 1096 (7th Cir. 2014).
    Plain-error review, however, does not win relief for Clark
on this issue. Plain-error review allows reversal where: (1)
there was an error; (2) the error was plain; (3) the error af-
fected the substantial rights of the defendant; and (4) the error
seriously impacted the fairness, integrity, or public reputation
of the proceedings. 

38; United States v.
Duran, 

, 834 (7th Cir. 2005). Here, we need not
16                                                 No. 18-2604

address the third and fourth elements. If any error occurred
at all, it was not “plain.”
   Under § 5G1.3(b) of the Guidelines, where a defendant has
an undischarged term of imprisonment, and where the con-
duct that led to the undischarged sentence
      is relevant conduct to the instant offense of con-
      viction under the provisions of subsections
      (a)(1), (a)(2), or (a)(3) of § 1B1.3 (Relevant Con-
      duct), the sentence for the instant offense shall
      be imposed as follows:
          (1) The court shall adjust the sentence for
          any period of imprisonment already
          served on the undischarged term of im-
          prisonment if the court determines that
          such period of imprisonment will not be
          credited to the federal sentence by the
          Bureau of Prisons; and
          (2) the sentence for the instant offense
          shall be imposed to run concurrently to
          the remainder of the undischarged term
          of imprisonment.
Section 5G1.3(d) further advises that “in any other case in-
volving an undischarged term of imprisonment, the sentence
for the instant offense may be imposed to run concurrently,
partially concurrently, or consecutively to the prior undis-
charged term of imprisonment to achieve a reasonable pun-
ishment for the instant offense.” So if Clark’s Illinois offense
was relevant conduct here, the Guidelines would recommend
concurrent sentences. If his Illinois offense was not relevant
conduct here, the choice of consecutive or concurrent
No. 18-2604                                                    17

sentences is left to the discretion of the sentencing judge, and
there is no reason to find an abuse of discretion here.
    As for the scope of relevant conduct, § 1B1.3(a)(1)(A) pro-
vides that “relevant conduct” is based on “all acts and omis-
sions committed, aided, abetted, counseled, commanded, in-
duced, procured, or willfully caused by the defendant.” “Sec-
tion 1B1.3(a)(2) applies to grouped offenses and includes ‘all
acts and omissions described in subdivisions 1(A) and 1(B)
that were part of the same course of conduct or common
scheme or plan as the offense of conviction.’” United States v.
Schrode, 

, 551 (7th Cir. 2016), quoting U.S.S.G.
§ 1B1.3(a)(2).
    “In assessing whether offenses are part of the same course
of conduct, we focus on whether the government has demon-
strated a significant similarity, regularity, and temporal prox-
imity.” United States v. Baines, 

, 963 (7th Cir. 2015)
(internal citations omitted). In considering whether an offense
is relevant within the meaning of § 1B1.3, our court has em-
phasized several factors that should guide courts’ analysis.
“Sufficient factual overlap exists where the past and present
offenses involve the same victims, stem from the same under-
lying conduct, or are in fact similar offenses.” United States v.
Orozco-Sanchez, 

, 850 (7th Cir. 2016); see also
United States v. Sumner, 

, 889 (7th Cir. 2003) (in
deciding relevant conduct in drug cases, courts look for “a
strong relationship between the uncharged conduct and the
convicted offense, focusing on whether the government has
demonstrated a significant similarity, regularity, and tem-
poral proximity”), quoting United States v. Acosta, 

,
279 (7th Cir. 1996). The relevant conduct issue is similar to the
issue in calculating criminal history under U.S.S.G. § 4A1.2
18                                                    No. 18-2604

(defining “prior sentence”), where two similar prior convic-
tions separated by an intervening arrest are counted sepa-
rately. E.g., United States v. Morgan, 

, 623 (7th Cir.
2003); United States v. Bradley, 

, 673 (7th Cir. 2000).
    The district court did not plainly err in not treating Clark’s
later offense in Illinois as relevant conduct for his Wisconsin
conviction. To support his argument, Clark points to facts set
forth in his Wisconsin PSR. The PSR explained that Clark trav-
eled to Chicago to obtain the drugs that he would then sell in
Wisconsin. This fact, Clark asserts, establishes that his drug
activities in both states were part of the same course of con-
duct, his fentanyl-trafficking enterprise.
    Under our precedents, this argument faces obvious “tem-
poral proximity” and “similarity” problems. First, Clark was
arrested in Wisconsin seven months before he was arrested
for selling a fentanyl mixture in Illinois. Further, the offenses
were dissimilar. In Wisconsin, Clark possessed more than 80
grams of fentanyl, most of which was packaged in retail pack-
ages of 0.5 grams. In Illinois, by contrast, he was arrested for
selling a confidential informant more than 65.2 grams of fen-
tanyl. And even if the offenses had been more similar, “the
fact that a defendant commits a certain type of offense on mul-
tiple occasions does not inevitably mean that each separate
offense is part of the same course of conduct.” 

. There is not enough here to have required the dis-
trict court to have found such strong connections between
these two drug offenses, especially without an objection by
the defense to that effect. The offenses involved different buy-
ers, in different cities, in different types of quantities, months
apart, and with an intervening arrest for the Wisconsin of-
fense.
No. 18-2604                                                    19

    “An error is ‘plain’ if the law at the time of appellate re-
view shows clearly that it was an error.” United States v.
Pierson, 

, 919 (7th Cir. 2019), citing Henderson v.
United States, 

, 279 (2013). Clark cannot establish
with precedent or otherwise that the district court should
have acted sua sponte to treat his Illinois offense as conduct
relevant to his Wisconsin charge. Even if there might have
been an error, it was not plain error to run Clark’s two federal
sentences consecutively.
   B. Supervised Release Condition
   Finally, the district court imposed a supervised release
condition that Clark “Not meet, communicate or spend time
with any persons known by defendant to be a member of or
affiliate of any known street gang.” On appeal, Clark argues
that the term “affiliate” is unconstitutionally vague. He did
not object to this condition at sentencing.
    In at least scores of appeals over the past five years, in the
wake of United States v. Siegel, 

(7th Cir. 2014),
United States v. Thompson, 

(7th Cir. 2015), and
United States v. Kappes, 

(7th Cir. 2015), this court
has wrestled with how best to handle appellate challenges to
supervised release conditions that were not raised in the dis-
trict court. Our recent opinion in United States v. Flores, 

(7th Cir. 2019), should bring some order to our
caselaw in this niche of waiver law, particularly since it was
circulated to the entire court under Circuit Rule 40(e), and no
judge voted to hear the case en banc. Special considerations
apply to our handling of supervised release conditions both
because they are usually the least salient concern at the time
of sentencing and, more important, because a defendant may
challenge them in the district court at any time. See Flores, 929
20                                                   No. 18-2604

F.3d at 451, citing 18 U.S.C. § 3583(e)(2); see also United States
v. Lewis, 

, 1083 (7th Cir. 2016) (defendant waived
challenge to supervised release condition: “sentencing in the
district court is the main event”).
    This case meets the standards of waiver set forth in Flores:
Clark had prior notice of the proposed conditions, including
the “affiliate” condition. He had a meaningful opportunity to
object, and his lawyer told the judge that the proposed condi-
tions were acceptable. Clark even raised other objections to
the PSR prior to sentencing but did not challenge this condi-
tion, which signals an intentional decision not to object. See

.
    We REVERSE the district court’s denial of Clark’s motion
for a Franks hearing without an evidentiary hearing. We
VACATE the district court’s judgment and REMAND for a
Franks hearing and appropriate action based on the outcome
of that hearing, either granting Clark’s motion to suppress or,
if his motion is denied, reinstating the convictions and sen-
tence. If the judgment is reinstated, the denial of Clark’s mo-
tion to suppress and the district court’s sentence are
AFFIRMED.
No. 18-2604                                                    21

    SCUDDER, Circuit Judge, dissenting in part. The panel opin-
ion is well done, and I join all parts of it except the conclusion
that the district court should have granted Michael Clark’s
motion for a Franks hearing. In my view, Investigator Todd
Maas’s affidavit supporting the search warrant, while omit-
ting facts damaging to the confidential informant’s credibil-
ity—facts that unquestionably should have been included—
adequately corroborated the information supplied by the in-
formant. The district judge reached this precise conclusion,
and under the deferential standard that guides our review on
appeal, we should affirm. There was no clear error here.
                                I
    The majority’s opinion correctly articulates the standard
for obtaining a Franks hearing: a defendant must make a sub-
stantial preliminary showing that an officer intentionally or
recklessly omitted information from a search warrant affida-
vit and the omissions were material to the probable cause de-
termination. See United States v. Glover, 

, 820 (7th
Cir. 2014). The majority also rightly observes that the omis-
sion of adverse information about an informant’s credibility
does not automatically warrant a hearing. See United States v.
Hancock, 

, 709 (7th Cir. 2016). Instead the proper
question is whether the affidavit would have established
probable cause if the omitted information had been included.
See 

To make this determination, we consider the
totality of the circumstances, including “the extent to which
the police have corroborated the informant’s statements.”
United States v. Sutton, 

, 773 (7th Cir. 2014).
    All agree that Investigator Maas failed to disclose critical
facts about the informant’s background. In my judgment,
though, the warrant application had sufficient corroboration
22                                                  No. 18-2604

to overcome this omission and establish the requisite proba-
ble cause to search hotel room 203 at the Baywalk Inn. Maas
took care to include three key pieces of corroborating evi-
dence in his affidavit:
        In response to the informant’s tip about possible drug
         dealing at the Baywalk Inn, Investigator Maas sur-
         veilled the hotel and observed a black male exit the
         hotel and quickly enter and exit at least five cars.
         Drawing on his experience as a narcotics investigator,
         Maas characterized this conduct as “textbook behav-
         ior to dealing narcotics.”
        Maas discovered that only two of the hotel’s 25 rooms
         had been rented with cash, with only one of the two—
         room 203—being rented for one night, which Maas
         explained was consistent with drug trafficking be-
         cause “drug dealers usually pay in cash and only stay
         one or two nights so to throw off police.”
        Maas also reported that, upon arriving at the hotel, he
         had a conversation with a woman he identified by
         name in the warrant application, who said her 20-
         year-old daughter was addicted to heroin. The
         woman added that, before the officer arrived, she was
         “also watching the black male who [was] suspected to
         be dealing” and she followed him to room 203.
    The majority’s analysis discounts this information, reason-
ing that Investigator Maas’s omission of material facts about
the informant’s background compromised his credibility on
all other portions of the warrant application. Put another way,
if Maas omitted critical information about the informant’s re-
liability, nothing else he included in his affidavit deserved to
No. 18-2604                                                     23

be credited as part of the probable cause analysis. But our case
law does not require such broad discounting and distrust. See,
e.g., United States v. Bradford, 

, 503–05. The district
court was not required to view as worthless the entirety of the
corroborating information provided by Maas.
    To be sure, a court faced with a reckless or intentional
omission of material information may consider whether (and
to what extent) the omission bears on the officer’s credibility
as part of its probable cause analysis. But I cannot say that the
district court clearly erred, in light of other corroborating ev-
idence, by choosing to credit Maas’s affidavit despite his
omission. So I would affirm the court’s denial of Clark’s mo-
tion for a Franks hearing.
                                II
    A final observation seems worthwhile. When this case en-
tered federal court, a magistrate judge reviewed the state-
court warrant application and noted that it, like many others
prepared by law enforcement officers in Wisconsin, omitted
information about the confidential informant’s criminal his-
tory. The panel’s opinion, aligned with our precedent, rein-
forces that this information is essential to a proper probable
cause analysis under the Fourth Amendment. See, e.g., United
States v. Musgraves, 

, 460 (7th Cir. 2016). Given the
frequency with which search warrants sought and executed
at the state level result in federal prosecutions, Wisconsin law
enforcement would do well to revisit its warrant application
practices. Omitting information about an informant’s credi-
bility creates real yet avoidable peril. Today’s decision proves
the point.