PUBLISHED UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT No. 20-4105 UNITED STATES OF AMERICA, Plaintiff – Appellee, v. MAXIMO H. GONDRES-MEDRANO, Defendant – Appellant. Appeal from the United States District Court for the District of Maryland, at Baltimore. George L. Russell, III, U.S. District Court Judge. (1:17-cr-00507-GLR-1) Argued: May 7, 2021 Decided: July 8, 2021 Before FLOYD, RICHARDSON, and QUATTLEBAUM, Circuit Judges. Affirmed by published opinion. Judge Richardson wrote the opinion, in which Judge Floyd and Judge Quattlebaum joined. ARGUED: Alfred Guillaume, III, LAW OFFICES OF ALFRED GUILLAUME III, Greenbelt, Maryland, for Appellant. Jeffrey Morgan Hann, OFFICE OF THE UNITED STATES ATTORNEY, Baltimore, Maryland, for Appellee. ON BRIEF: Robert K. Hur, United States Attorney, OFFICE OF THE UNITED STATES ATTORNEY, Baltimore, Maryland, for Appellee. RICHARDSON, Circuit Judge: Confidential informants are often a fixture of investigations in popular media. And real-life informants provide vital help to law enforcement, especially in drug cases. This case is a prime example. In 2017, a known confidential informant with a track record of providing reliable information told law enforcement that Maximo Humberto Gondres- Medrano was trafficking drugs in Baltimore, Maryland. The informant’s insight led law enforcement to watch Gondres-Medrano put a shoebox in the informant’s car and drive off to allegedly sell heroin. When the car was stopped, officers found heroin in the shoebox. Gondres-Medrano was indicted, convicted, and sentenced to 121 months in prison. Gondres-Medrano appeals three issues. First, he argues that the police lacked probable cause to search the shoebox so the heroin should have been suppressed. Second, he argues that the court erred in admitting a video that showed him opening a heroin package because the unfair prejudice from the video was too high under Rule 403. Finally, he claims that the court erred in applying an obstruction-of-justice enhancement for perjury during sentencing. We disagree and affirm. I. Background A. The investigation and arrest In 2017, a long-standing confidential informant told FBI Task Force Officer Mark Williams that an individual later identified as the defendant, Gondres-Medrano, was involved in heroin and cocaine trafficking between New York and Baltimore. This informant had been used several times before, leading to several arrests. So Williams began to investigate. The investigation included Williams listening to a recorded 2 conversation between the informant and a person the informant identified as Gondres- Medrano. Gondres-Medrano talked about trafficking cocaine between his place in Maryland and New York. Based on this information, Williams obtained a warrant to track Gondres-Medrano’s cellphone. Williams also learned Gondres-Medrano’s full name and address in Baltimore. Williams then got in contact with Homeland Security Investigations Special Agent Edward Kelly who ran an immigration check and learned that Gondres- Medrano had overstayed his immigration visa, had a disqualifying drug conviction, and was subject to removal. Agent Kelly issued a Notice to Appear and obtained an arrest warrant for Gondres-Medrano. The informant then told Williams that Gondres-Medrano intended to transport heroin from his residence on September 8th and that the informant was to provide transportation. So the officers set up surveillance that day at Gondres-Medrano’s residence. Before arriving at Gondres-Medrano’s residence, the informant and his car were searched by the officers. The officers then watched the informant pull up to the residence, get out and enter the house, and return with Gondres-Medrano two or three minutes later. Agent Kelly positively identified Gondres-Medrano, connecting him to the residence associated with the tracked phone number. The officers then witnessed Gondres-Medrano carry a shoebox that he put in the backseat before getting in the passenger side of the informant’s car and driving off. The officers stopped the car and handcuffed Gondres- Medrano and the informant. The officers then found the shoebox on the backseat floorboard. They opened it and found a white powdery substance wrapped in gray tape. Later testing established that the substance was heroin and fentanyl. 3 Gondres-Medrano waived his Miranda rights and consented to a search of his phone and apartment. The officers then interviewed him and asked, “How did that box get to you?” and “How did you get possession of that box?” J.A. 385. Gondres-Medrano responded, “It’s that . . . they were sending it.” Id. Gondres-Medrano then asked if he could show them a video on his cellphone, explaining: “I will show them on there, how it arrived.” J.A. 386 (emphasis added). The video on his phone from weeks before the arrest showed him opening an envelope, peeling the liner apart, and removing a hidden plastic pouch with a white substance while giving a thumbs up to the camera, all at his residence. The video also showed a digital scale. Phone records showed that he sent the video to at least one person. When asked about the substance, Gondres-Medrano said it was 333 grams of heroin that he received at his house. Later, Gondres-Medrano discussed various cocaine and heroin sources, noting that someone named “Freddie” sent the seized drugs and that he was taking the drugs to a friend, who would pay him. The government also retrieved an image from Gondres-Medrano’s phone that showed a shoebox matching the one recovered from the car. B. Motions and trial Gondres-Medrano was charged with possession with intent to distribute a mixture containing fentanyl and heroin. See 21 U.S.C. § 841(a)(1). Gondres-Medrano sought to suppress and exclude certain evidence. First, he moved to suppress the drugs found in the car as a violation of his Fourth Amendment rights, arguing there was no warrant, probable cause, or warrant exception for the search and seizure. The court denied the motion, ruling that the search was lawful. The court did so 4 because it found that the informant was reliable and there was “no question that law enforcement at the very least had an articulable suspicion, if not probable cause that the defendant was, in fact, engaged in a narcotics transaction.” J.A. 113. The court then said the search met the automobile exception. The court also said there was probable cause to search the box based on the corroborated narcotics activity associated with Gondres- Medrano. Gondres-Medrano also tried to exclude the video. He argued that the evidence would confuse the jury and be prejudicial under Federal Rule of Evidence 403. The court rejected his argument, allowing the video to show intent, knowledge, and that the drugs in the video were the same as the drugs in the shoebox. Gondres-Medrano asked about a potential limiting instruction as to what the video could be used for, and the court discussed options. The court did not make a decision about an instruction and noted that Gondres- Medrano might change his mind about whether he wanted such a limiting instruction because an instruction might highlight the video in a way that would ultimately hurt his case. The trial proceeded with the agents discussing the investigation, drugs, the video, and the interview. Gondres-Medrano decided to testify about where he got the drugs and about the video. He said he gave the video to the police because he recorded the video to “file a complaint” against the person who had sent him the package. He testified that he expected the package to contain documents when he received it. However, once Gondres- Medrano followed up with the person who sent him the package, that person told Gondres- Medrano that the package contained drugs and there would be problems with Gondres- 5 Medrano’s family if he did not deliver the drugs to New York. He later admitted to bringing those drugs to New York but said he did so under duress. He claimed that he had never handled drugs or seen them in person before that package arrived. Gondres-Medrano then claimed that on September 8th, his friend had come to Gondres-Medrano’s house to pick up drugs that Freddie had dropped off there a week earlier. Although Gondres- Medrano had spoken to Freddie on his friend’s behalf, Gondres-Medrano denied any further participation in the deal. He claimed that after Freddie brought him the drugs, he agreed to keep the drugs at his house but never looked inside the package or used the drugs. Finally, Gondres-Medrano claimed that the friend came to his home on September 8th and that it was him, not Gondres-Medrano, who carried the shoebox to the car that day. During cross, Gondres-Medrano said the video was taken two weeks before his arrest and acknowledged the digital scale in the video. But he maintained that he only recorded the video to provide evidence to the police. Even so, he testified that he did not, in fact, notify the police at any point before his arrest and that he delivered the drugs to a person in New York and agreed to get paid for future deliveries. Gondres-Medrano then explained that the drugs in the shoebox were different from the drugs seen on the video. He reiterated that he had been threatened when he received the packaged drugs and that he was “worried” and under duress when he opened the drugs on the video. In response, the government replayed the scene in the video where Gondres-Medrano provides a thumbs up sign to the camera. The defense rested, and the jury found Gondres-Medrano guilty. 6 C. Sentencing and appeal Based on the drug weight and his criminal history, Gondres-Medrano faced a guidelines range of 97 to 121 months. But the court also applied a two-level enhancement for obstruction of justice. U.S.S.G. § 3C1.1. So his guidelines range increased to 121 to 151 months. Either way, Gondres-Medrano faced a 10-year (120-month) statutory minimum. 21 U.S.C. § 841(b)(1)(A)(vi). Gondres-Medrano argued that the obstruction enhancement should not be applied. The government countered that Gondres-Medrano had committed perjury by purposefully lying about the package and drugs in the video, which were material facts. Gondres- Medrano asserted that his testimony did not amount to perjury, either because it was not material or because the evidence failed to show it was false. The court found that Gondres- Medrano presented perjured testimony and that the two-level enhancement would apply. The court sentenced Gondres-Medrano to 121 months in prison and noted that it would have imposed the same sentence even without the obstruction enhancement. Gondres-Medrano timely appeals the district court’s decision to admit the drugs from the shoebox and the video and to apply the obstruction enhancement. The district court had jurisdiction over this federal criminal case. See 18 U.S.C. § 3231. And we have jurisdiction on appeal. See 18 U.S.C. § 3742(a); 28 U.S.C. § 1291. II. Analysis A. Probable cause existed to search the shoebox Police, without a warrant, may “search an automobile and the containers within it where they have probable cause to believe contraband or evidence is contained.” 7 California v. Acevedo,
, 580 (1991). At base, Gondres-Medrano claims the government lacked probable cause. We disagree. 1 The information from the known reliable informant established probable cause that the shoebox contained heroin. Probable cause has long been understood to encompass circumstances that, while less than a preponderance, “warrant suspicion.” Locke v. United States, 11 U.S. (7 Cranch) 339, 348 (1813). And it is the “combined circumstances,” and not each circumstance “separately,” by which we must evaluate probable cause.
–48. As the Supreme Court explained, probable cause turns on a “totality-of-the-circumstances analysis” and requires “a practical, common-sense decision whether . . . there is a fair probability that contraband or evidence of a crime will be found in a particular place.” Illinois v. Gates,
, 238 (1983). One circumstance law enforcement relies on is information provided by citizens, whether victims, bystanders, or informants. Citizen informants in particular often contribute to probable cause. But the use of informants has long bedeviled courts considering searches and seizures under the Fourth Amendment. For example, in 1932 the Supreme Court relied on information provided by an informant to find probable cause to stop and search a car. Husty v. United States,
We review probable-cause determinations de novo but must “take care both to review findings of historical fact only for clear error and to give due weight to inferences drawn from those facts by resident judges and local law enforcement officers.” Ornelas v. United States,
, 699 (1996). As a result, “we must construe the evidence in the light most favorable to the prevailing party.” United States v. Lewis,
, 197 (4th Cir. 2010). In doing so, we are not limited by the evidence at the suppression hearing and can consider the evidence produced at trial. United States v. Han,
, 539 (4th Cir. 1996); Carroll v. United States,
, 162 (1925). 8 694, 701 (1931). The officer was “well acquainted with his informant,” who had provided “similar information to the [officer] before, which had always been found to be reliable.”
. By telephone, the informant explained that “Husty had two loads of liquor in automobiles of a particular make and description, parked in particular places on named streets.”
The Court found that information, along with the officer’s own knowledge of Husty’s prior illegal activities, their discovery of the automobiles on the identified street along with Husty and two companions, and the companions’ prompt escape attempt when hailed by officers, provided the officers with probable cause.
; see also Draper v. United States,
, 309, 312–13 (1959) (information from an informant who had proved reliable in the past established probable cause when, as the informant said would happen, an unknown man matching the informant’s description exited from a certain train on one of the two days the informant had predicted, wearing particular clothing, carrying a tan zipper bag, and walking fast). In the 1960s, the Supreme Court adopted a two-part test for evaluating an informant’s information. See Aguilar v. Texas,
, 114 (1964); Spinelli v. United States,
, 413, 415–17 (1969). Aguilar and Spinelli required police to separately establish both (1) the informant’s basis of knowledge and (2) the tip’s veracity, that is, the informant’s credibility. Aguilar,
. In 1983, however, the Supreme Court rejected this rigid two-part test, returning to “the totality-of-the-circumstances analysis that traditionally has informed probable cause determinations.” Gates,
. In Gates, the Court found probable cause based on an anonymous handwritten letter stating that the Gates were selling drugs at their home 9 in Bloomingdale, Illinois.
–46. The letter described how the Gates obtained their drugs: the wife drove to Florida and left their car to be loaded with drugs while the husband flew to Florida and drove their car back.
. The letter explained that the wife would be driving down again on May 3, with the husband flying down a few days later.
After confirming the Gates lived in Bloomingdale, officers surveilled the husband on a flight to Florida on May 5, after which he checked into a hotel room in his wife’s name.
. The next morning the Gates were seen leaving in a car with plates registered to a different car owned by Gates.
–27. They left on an interstate highway used by travelers to drive from Florida to Chicago.
Based on these facts, a state-court judge found probable cause to issue a search warrant for the car and residence for when they returned.
. The Supreme Court acknowledged that the anonymous letter, standing alone, would not provide probable cause.
. But the Court rejected the “elaborate set of legal rules that have developed among various lower courts to enforce the ‘two-pronged test.’”
. Instead, the Court returned to the “practical, nontechnical conception” of probable cause.
(quoting Brinegar v. United States,
, 176 (1949)). An informant’s “veracity” or “reliability” and his “basis of knowledge” are relevant considerations in the totality-of-the-circumstances analysis, but neither consideration is necessary.
; see also Massachusetts v. Upton,
, 732 (1984). For in “determining the overall reliability of a tip,” a “deficiency” in one consideration may be “compensated for” by some other indicia of reliability or “a strong showing as to” another consideration. Gates,
. The Court then found that the police corroboration 10 of some details in the informant’s letter, including the Gates’ predicted actions, “indicated . . . that the informant’s other assertions also were true” and made up for the letter’s anonymity.
–46. The corroboration also reflected the informant had firsthand knowledge of the Gates’ activities.
–46. In this way, the Court emphasized that the corroboration verified the informant’s reliability. And this “suffices for the practical, common-sense judgment called for in making a probable cause determination.”
. After Gates, the Court has emphasized the corroborating aspects of an anonymous tip as establishing the anonymous informant’s reliability. See Alabama v. White,
, 329 (1990). Standing alone, an anonymous tipster’s veracity or reliability is “by hypothesis largely unknown, and unknowable.”
). But where a tipster is shown to be right about some details, he is probably right about other alleged facts, including the alleged criminal activity, justifying reliance on the anonymous tip.
–32. For example, in White, an anonymous call reported that Vanessa White would be leaving a particular apartment at a particular time in a brown Plymouth station wagon with a broken right taillight to go to a particular motel.
. The tip also recounted that she would have a brown case with an ounce of cocaine.
At the indicated time, police saw a woman with nothing in her hands leave the apartment and get in the described station wagon.
After she began on the path to the hotel, police stopped the car and found marijuana in a case inside.
Relying on the corroboration of the tip’s predictive aspects, the Court found there was good reason to believe the caller was honest, 11 well informed, and reliable.
; cf. United States v. Perkins,
, 325 (4th Cir. 2004). Similarly, corroboration can confirm the reliability of an informant who is known (rather than anonymous) but whose credibility is unknown to the officer. See United States v. Harris,
, 579–84 (1971); United States v. White,
, 952 (4th Cir. 2008). 2 Some corroboration is still required with no track record of reliability. 3 But since a known informant exposes himself to possible criminal prosecution or other consequences for giving false information, his reliability is enhanced and less corroboration is required as compared to an anonymous informant. 4 2 This often arises when a recently arrested person provides the police information about other potential targets. See White,
; United States v. Miller,
, 699–700 (4th Cir. 1991). 3 In Adams v. Williams,
, 146 (1972), the Court addressed a known informant with no indication that prior information had proven reliable. In that context, the Court found reasonable suspicion but questioned whether such an informant’s unverified tip could establish probable cause.
. Similarly, in considering a recently arrested informant with no known history of reliability, we generally stated that “[a]n informant’s tip is rarely adequate on its own to support a finding of probable cause.” Miller,
. Without a record of reliability, even a known informant must be verified by the nature of the statement or other information showing the informant is honest and well informed.
U.S. at 146–47 (1972); White,
; see also Navarette v. California,
, 400–01 (2014) (finding a tipster reliable when the tipster’s statement about being run off the road by a specific vehicle was “made under the stress of excitement caused by a startling event” using the 911 system, which recorded information about the caller and could subject the caller to prosecution); Harris,
–84 (finding known informant with no track record making admissions of having committed crimes credible enough to support finding probable cause because statements against penal interest carry their own indicia of credibility); United States v. Gagnon,
, 236 (2d Cir. 2004) (A “face-to-face informant must . . . be 12 But corroborating details of an informant’s tip is not the only way to establish the informant’s reliability. See Miller,
–99 (explaining that corroboration need not come from independent investigation). Just as the veracity of the anonymous tip made up for the tipster’s total lack of reliability or accountability in Gates, so too can other factors compensate for deficiencies. In considering the reliability of a known informant, the Supreme Court has repeatedly emphasized the import of an informant’s track record of providing reliable information. See, e.g., McCray v. Illinois,
, 303–04 (1967) (finding “no doubt” that probable cause existed based on information provided by an informant with a record of providing “information concerning narcotics ‘20 or 25 times’” which had led to arrests and convictions). This is because the informant’s track record provides verification and corroboration that his future tips will likewise be reliable. See Draper,
(finding probable cause based on an informant when the officer “had always found the information given [by the informant] to be accurate and reliable”); Husty,
(relying on an informant who was “well acquainted” with the officer and who had given “similar information” to the officer before, “which had always been found to be reliable”); see also Florida v. J.L.,
, 270 (2000) (finding no reasonable suspicion based on a vague anonymous tip and contrasting that with “a known informant whose reputation can be assessed and who can be held responsible if her allegations turn out to be fabricated”). 5 Indeed, corroboration can come from “showing thought more reliable than an anonymous telephone tipster, for the former runs the greater risk that he may be held accountable if his information proves false.”). See also United States v. Ross,
, 800 (1982) (describing “an informant 5 who had previously proved to be reliable”); Arkansas v. Sanders,
, 761 (1979) 13 the trustworthiness of the informant, such as a history of providing accurate (corroborated) information.” United States v. Long,
, 659 (10th Cir. 2014). So “probable cause can be based solely on information from a reliable, credible informant.”
; see also United States v. Searcy,
, 1123–24 (7th Cir. 2011) (noting that an informant with a track record is less likely to be biased against a target since they have provided information against multiple targets). But we still look to the totality of the circumstances. Gates,
–31. And so outlandish, vague, or contradictory information provided by a known reliable informant, even an informant who had given verified tips before, may defeat probable cause. So too may corroboration and independent police work strengthen the probable cause deriving from information provided by a known and reliable informant. While the “practical, nontechnical conception” of probable cause fails to provide rigid rules, these considerations provide a framework for evaluating an informant’s information in establishing probable cause. In this case, we have little doubt that the officers had probable cause to believe the shoebox contained drugs. The informant providing that information was known and reliable, having given the police accurate information in the past that led to several (finding “ample probable cause to believe that respondent’s green suitcase contained marihuana” based on information from a “previously reliable informant”), abrogated on other grounds by Acevedo,
; cf. United States v. Shepherd,
, 317 (4th Cir. 1983) (crediting the tip because the source was a “highly reliable informant” whose tips “netted over twenty-five convictions”). 14 successful arrests. The officers had also met with this informant and could evaluate his demeanor and hold him responsible for lying. Moreover, the officers corroborated significant parts of the informant’s information. Williams overheard a phone call between the informant and someone the informant identified as Gondres-Medrano discussing trafficking cocaine between Maryland and New York. This provided probable cause to obtain a warrant to track Gondres-Medrano’s cellphone. The officers also confirmed Gondres-Medrano’s identity and residence and learned of his prior drug conviction and immigration history. See United States v. Bynum,
, 197–98 (4th Cir. 2002) (relying on a target’s prior criminal activity); Miller,
–700 (same). Then on September 8th, officers met with the informant, searched him and the car, then surveyed Gondres-Medrano’s house and told the informant when to arrive. Consistent with the informant’s assertion that Gondres-Medrano would be transporting drugs from his residence on that day, when the informant pulled up to the residence, Gondres-Medrano let him in, came out with a shoebox, and got in the car to drive away. That Gondres-Medrano climbed into the informant’s car with a shoebox at the time and place described shows the informant’s firsthand knowledge and corroborates his story. These circumstances established probable cause to believe that Gondres-Medrano was trafficking drugs and had drugs in the shoebox. Gondres-Medrano and the government spar over how to apply a few cases, but none offer much guidance. First, in Miller,
, we found probable cause based on a tip from a recently arrested individual with no record of providing reliable information.
–700. The officers corroborated the informant’s predictive details: The target would 15 arrive from New York by bus on a certain date, described her clothes, identified her in a picture, and said she would be carrying a brown bag.
. And the officer knew the target based on previous drug charges.
So even though the source had no history of reliability, we found probable cause because the informant had an interest in providing accurate information, would face consequences for lying, much of the tip was corroborated, and the officer knew the target from the previous drug charge.
–700. 6 Gondres-Medrano also relies on United States v. Wilhelm,
, 117, 120 (4th Cir. 1996), where we found no probable cause based on an unnamed, unknown informant’s claims that he believed he saw the defendant selling marijuana at his house and gave the police directions to the house. The government relied on “conclusory descriptions” of the informant’s demeanor and maturity to prove that the informant was reliable.
. With no other information about the informant’s reliability, the lack of corroboration and meaningful indications of the informant’s reliability doomed the warrant.
. 7 Unlike Miller and Wilhelm, the officers here relied on a known informant with a track record of providing reliable information. 6 The parties disagree over whose side United States v. Blackwood,
(4th Cir. 1990), supports. There, the police were given information that in the previous seventy-two hours crack cocaine was sold from the defendant’s residence, prompting the police to enlist a reliable informant who explained how he would conduct a controlled buy from the defendant.
. The police watched the purchase go just as the informant said and then field tested the drugs and identified them as crack.
We found probable cause based on the totality of the circumstances, as the police had received information from a reliable informant and had corroborated the information received.
. Although the police’s independent investigation helped establish probable cause, such evidence is not required. Just because something is sufficient does not render it necessary. 16 With probable cause, the police could search the shoebox under the automobile exception. Maryland v. Dyson,
, 467 (1999); Acevedo,
; see also Shepherd,
–17 (finding probable cause to search vehicle based on a reliable informant telling the police about a moonshining operation at a residence and law enforcement surveilling the residence and seeing the defendant put translucent jugs in the trunk). 8 B. The court did not abuse its discretion in admitting the video under Rule 403 Federal Rule of Evidence 403 says that the “court may exclude relevant evidence if its probative value is substantially outweighed by a danger of one or more of the following: unfair prejudice, confusing the issues, misleading the jury, undue delay, wasting time, or needlessly presenting cumulative evidence.” (emphasis added). “This Court reviews evidentiary rulings for abuse of discretion.” United States v. Hill,
, 304 (4th 7 We distinguished Wilhelm in Bynum because the police in Bynum used a reliable informant, knew that the defendant was a felon, knew that he resided at the location, and had seized drugs at that location less than four months earlier. Bynum,
– 98; see also United States v. Lalor,
, 1581 (4th Cir. 1993) (finding sufficient evidence to conclude the informants were reliable where the police corroborated the informants’ information about the defendant’s address, vehicle, and alias and determined that he had been arrested for drug possession just a few days before). 8 Gondres-Medrano also argues that the automobile exception is inapplicable here because the vehicle was not readily mobile during the search as he was detained first. This argument has been rejected by the Supreme Court many times, making it clear that the car need only be capable of movement. See Florida v. Meyers,
, 382 (1984). (upholding a search after the car had been impounded); Michigan v. Thomas,
, 261 (1982) (same); Texas v. White,
, 68 (1975) (same); see also Pennsylvania v. Labron,
, 940 (1996) (explaining that past cases require only probable cause and that the vehicle be “readily mobile” to permit a search). 17 Cir. 2003). But when reviewing a district court’s balancing under Rule 403, we apply a “highly deferential” standard of review, and do not overturn the district court’s decision “‘except under the most extraordinary circumstances, where that discretion has been plainly abused.’” United States v. Hassan,
, 132 (4th Cir. 2014) (quoting United States v. Udeozor,
, 265 (4th Cir. 2008)). And since the probative value must be “substantially outweighed” by one or more of the listed dangers, relevant evidence should be excluded only “when there is a genuine risk that the emotions of a jury will be excited to irrational behavior, and this risk is disproportionate to the probative value of the offered evidence.”
(quoting United States v. Siegel,
, 319 (4th Cir. 2008)). The court here acted well within its wide discretion. Gondres-Medrano argues that the drugs in the video are different from the drugs in the shoebox, so admitting the video might confuse the jury and unfairly prejudice him with no relevance or real probative value. Not so. The video helped establish the elements of the possession-with-intent-to-distribute charge: possession, knowledge, and intent. United States v. Randall,
, 209 (4th Cir. 1999). The video shows Gondres-Medrano’s handling and comfort with drugs. He chose to film himself taking the drugs out and showed no surprise when drugs were inside. Upon carefully removing the drugs from a hidden liner, he gave a thumbs up. And a digital scale could be seen behind him. It remains unclear whether the heroin in the video was the same as the heroin seized in the shoebox. Gondres-Medrano argues the government failed to prove that they were. But there was some evidence that the drugs were the same, as 18 Gondres-Medrano said he would show the police the video to explain “how it arrived” in response to questions about the shoebox drugs. The video also undermined Gondres-Medrano’s primary defense at trial: that he held the drugs under duress and did not intend to distribute them. Nothing about Gondres- Medrano’s actions or reactions on the video suggests any duress or distress in receiving the heroin. Quite the contrary. That Gondres-Medrano began filming before opening the package, quickly removed the drugs from a hidden liner, was unsurprised by the drugs, gave a thumbs up, and had a scale nearby all convey knowledge and intent rather than duress. Gondres-Medrano also argued that he made the video to give to police, so it was evidence that he never intended to distribute the heroin. But Gondres-Medrano did not give the video to police until he was arrested. And by that time, he had circulated it to others. The video contradicts Gondres-Medrano’s testimony, which undermines his credibility, an important part of any case in which the defense turns on the defendant’s testimony and state of mind. And while he essentially admitted that he knew that the substance in the shoebox was drugs and that he possessed them, Gondres-Medrano did not stipulate to those elements and did not admit this at the suppression hearing. It was only after the prosecution’s case that Gondres-Medrano admitted these facts. So without a stipulation, the government had to prove each element. The video, as the court correctly found, was therefore highly probative to the government’s case and to undermining Gondres- Medrano’s defense. See J.A. 435 (explaining the closeness in time and nature of the video 19 made it relevant and probative to the drugs seized, knowledge, and intent). So the probative value was not “substantially outweighed” by anything else, and the video’s probative value was certainly not so plainly outweighed that the court abused its discretion in admitting it. Hassan, 742 F.3d at 132. 9 The court then brought up the possibility of crafting a curative instruction but stopped short of finding one necessary, instead suggesting that decision be made by Gondres-Medrano at a later point in the trial. J.A. 435–36. The court never gave this instruction. But the court made it clear that it was trying to decide when and whether to provide any limiting instruction, and it put the onus on defense counsel to ask for the instruction if counsel wanted it, since counsel may have changed his mind during trial. The court never promised to give an instruction or make the video’s admission conditional on giving that instruction. Instead, the court made it clear it would wait on defense counsel to decide whether he wanted the limiting instruction, and counsel crossed on the issue but never asked for an instruction. Although Gondres-Medrano never asked for the instruction, he argues not giving one was a plain error. United States v. Marcus,
, 262 9 Gondres-Medrano relies on Hill,
, to argue that the district court abused its discretion. There, a defendant accused of defrauding a retiree sought to introduce a gift letter showing that the victim had previously promised to gift an associate some money just to turn around and sue him for failing to pay a loan.
. We found that this evidence was properly excluded because what happened with the third-party associate was disputed and figuring that out would require a “case within a case that would have confused the jury as to the issues to be decided.”
. Our case is different. The disputed issue here is directly relevant in a way that the dispute in Hill, which happened years prior and involved a third party not privy to the case, was not. So any risk of confusion is substantially less. But even if this case were like Hill, Hill does not dictate that evidence must be excluded when a trial within a trial is possible, as Hill was affirming a district court’s discretionary choice to exclude the evidence. 20 (2010). We find no error, much less a plain one, in failing to provide a limiting instruction to the properly admitted video. See United States v. Johnson,
, 178 (4th Cir. 2019) (finding no plain error in failing to give a limiting instruction required by statute). 10 C. Applying the sentencing enhancement was harmless Gondres-Medrano finally argues that the court erred when it applied a two-level sentencing enhancement to his guidelines range based on perjury under Federal Sentencing Guideline § 3C1.1. It is not apparent that this was erroneous, but even if it was, we can find any error harmless if we have “(1) ‘knowledge that the district court would have reached the same result even if it had decided the guidelines issue the other way,’ and (2) ‘a determination that the sentence would be reasonable even if the guidelines issue had been decided in the defendant’s favor.’” United States v. Savillon-Matute,
, 123 (4th Cir. 2011) (quoting United States v. Keene,
, 1349 (11th Cir. 2006)); 10 Gondres-Medrano relies heavily on the framework laid out in United States v. Mitchell,
(4th Cir. 1993), to argue that there was a plain error here. But that case is inapposite. In Mitchell, we found plain error when the prosecution, during closing arguments, repeatedly emphasized the defendant’s brother’s conviction for participating in the same conspiracy the defendant was charged with and focused on the fact that the jury in his brother’s case did not believe his brother’s testimony.
–43. While evidence of the conviction was admissible to impeach the brother’s testimony, it was not admissible for the purpose the prosecutor used it for: to argue that because the brother had been convicted, the defendant should also be convicted.
To determine whether it was plain error not to stop the prosecutor from making the statements he made, we considered five factors, one of which was whether a limiting instruction was given.
But this framework is inapplicable here. In Mitchell, a limiting instruction was only one factor to consider in determining whether “the prosecution’s conduct prejudiced the appellant’s substantial rights,” one part of the plain-error inquiry.
. But here the plain-error review applies only to the failure to instruct the jury on how it could use the video, rather than the decision to admit the video. And as explained above, the court had broad discretion to admit the video for various purposes under Rule 403 and quite clearly did not commit a reversible error in doing so. 21 see also United States v. McDonald,
, 643 (4th Cir. 2017). Both requirements are met here. Gondres-Medrano admits that his 121-month sentence was substantively reasonable and does not contest the harmlessness analysis. The district court explicitly said that it would have given the same sentence without the enhancement, noting that it was only one month above the statutory minimum. J.A. 468; see also 21 U.S.C. § 841(b)(1)(A)(vi). This is enough to meet the first prong of the harmlessness analysis. See McDonald, 850 F.3d at 643–44. As for the second prong, Gondres-Medrano was given a sentence of 121 months, the bottom of his 121-to-151-month guidelines range that included the § 3C1.1 enhancement. J.A. 150. Even without the enhancement, however, 121 months would be within his guidelines range. See J.A. 445 (showing an offense level of 30 without the enhancement); U.S.S.G. § 5A (setting a guidelines range of 97 to 121 months for an offense level of 30 and criminal history category of I). And regardless, the statute under which he was sentenced required a sentence of at least 120 months. 21 U.S.C. § 841(b)(1)(A)(vi). So Gondres-Medrano’s sentence would have been substantively reasonable even if applying the § 3C1.1 enhancement was improper. Thus, any error was harmless, and the sentence must be affirmed. * * * Probable cause is an elusive and flexible concept. It may turn on a variety of circumstances. One common source police rely on, and that has been the crux of multiple Supreme Court cases, is the informant. Here, a known informant had a track record of providing reliable information to law enforcement. And in the circumstances here, the 22 information he provided readily established probable cause, justifying the search. Gondres-Medrano’s other claims also fail: The district court acted within its discretion in admitting the video, and the application of the perjury enhancement was, at most, harmless error. So we AFFIRM. 23