BARRON, Circuit Judge.

Nicholas Occhiuto brings two challenges to his convictions for conspiracy to distribute heroin and distribution of heroin in violation of 21 U.S.C. §§ 846 and 841. He contends that the government relied on testimony that violated the Confrontation Clause and that the District Court denied him constitutional due process in preventing him from calling a witness critical to his defense. Occhiuto also challenges his sentence on the grounds that the District Court clearly erred in its factual determinations under the Sentencing Guidelines and imposed a term of imprisonment that was substantively unreasonable. We find no merit to any of these challenges and therefore affirm.

I.

A.

We begin with the Confrontation Clause challenge. See U.S. Const. amend. VI ("In all criminal prosecutions, the accused shall enjoy the right ... to be confronted with the witnesses against him...."). Occhiuto's challenge takes aim at the testimony FBI Special Agent Jeffrey Wood provided at his trial.

Wood oversaw an investigation into Occhiuto in the fall of 2009. The investigation into Occhiuto was undertaken as part of a broader investigation by federal, state, and local law enforcement into drug trafficking in Lynn, Massachusetts, and neighboring areas.

Wood testified at trial about how the investigating officers arranged for a cooperating informant, whom we will call "A.J.",1 to make undercover purchases from Occhiuto that law enforcement agents would secretly record by video. The government also presented surveillance evidence and testimony to establish that Occhiuto made sales of heroin to A.J. on October 1, 2009, and October 5, 2009, and a sale of cocaine to her on September 29, 2009.

In his testimony, Wood described the measures that law enforcement agents undertook to record A.J.'s undercover encounters with Occhiuto. Wood testified that a team of officers was nearby during the encounters, placed audio-visual recording equipment in A.J.'s vehicle, and had A.J. wear a wire. Wood also testified that A.J. was a known drug user and that law enforcement agents undertook some additional steps to ensure the retrieval of physical evidence from A.J. after each of her recorded encounters with Occhiuto. To that end, Wood testified, law enforcement agents met with A.J. before and after each encounter and searched her person for contraband.

Occhiuto objects specifically to the portion of Wood's testimony in which Wood stated that the drugs in evidence were the same drugs that Occhiuto had sold to A.J. on the dates covered by the surveillance. Occhiuto argues that Wood's testimony about the controlled buys relied on, and thus necessarily relayed to the factfinder, what A.J. had said to Wood about what occurred during her encounters with Occhiuto. And thus, Occhiuto argues, this aspect of Wood's testimony violated the Confrontation Clause because it implicitly — but necessarily — related the out-of-court statements made by the informant, A.J., about her transactions with Occhiuto.

To support this contention, Occhiuto relies on United States v. Meises, 645 F.3d 5 (1st Cir.2011). There, we found a Confrontation Clause violation based on an agent's testimony that implicitly related the statements of an informant who did not testify at trial. See id. at 18-21. The agent in Meises was asked on the stand if anything the informant had said during an interview "changed the targets of the investigation and prompted the defendants' arrests," and the agent then answered in the affirmative. Id. at 21.

Meises held, on the basis of that record, that the government had sought to prove its case with "testimony that plainly told the jurors that [the informant] said [the defendants] were co-conspirators rather than with the available evidence circumstantially pointing to their culpability." Id. (emphasis omitted). Meises explained that it "ma[de] no difference that the government took care not to introduce [the informant's] `actual statements'" since a "reasonable jury could only have understood [the agent] to have communicated that [the informant] had identified appellants as participants in the drug deal." Id.

But this case is not at all like Meises. Nothing in Wood's testimony suggested that a reasonable factfinder "could only have understood [Wood] to have communicated that [A.J.] had" told Wood that Occhiuto had sold her the drugs. Wood did not purport to recount anything during his testimony that A.J. had said to him about the buys. Wood did recount that he "debriefed" A.J. after the controlled buys in his description of what control measures that law enforcement undertook. But, despite Occhiuto's assertion that A.J. must have told Wood about the controlled buys during the debriefings, nothing in Wood's testimony referenced the content of these debriefings. Wood testified, throughout, without reference — direct or indirect — to any statement A.J. may have made to him about what had been exchanged, and Occhiuto identifies nothing in the record to suggest otherwise. See, e.g., United States v. Foster, 701 F.3d 1142, 1154 (7th Cir. 2012) ("The challenged testimony ... exclusively concerned the agents' personal observations and actions: the agents personally witnessed the controlled buys, searched the CI before and after each transaction, and followed the CI to the debriefing location after each transaction to collect the drugs, money, and recording equipment. Accordingly, their own actions formed the basis for their testimony, and their testimony did not relay `nonverbal conduct' statements of the CI.").

Instead, Wood, by virtue of the surveillance, could testify about what had been exchanged during the encounters from his personal observations. Wood was on the scene on each of the relevant dates. And he testified that he had searched A.J. before and after each buy and monitored her interactions with Occhiuto. Thus, his testimony that the transactions involved the same drugs put in evidence did not necessarily rely on any statements from A.J.

Occhiuto nonetheless argues that Wood must have been relying on statements that A.J. made to him at the time of the transactions because the evidence showed gaps in Wood's knowledge of what happened during the controlled buys that could only have been filled by such statements. And thus, Occhiuto contends, Wood's testimony must be understood to have relayed those same statements.

Occhiuto notes in this regard that there is no direct evidence to confirm that A.J. did not tamper with the evidence before handing it to the investigating agents, who (led by Wood) were conducting the contemporaneous surveillance. And Occhiuto contends that the agents, according to Wood's own testimony, did not do the most thorough search possible of A.J.'s body before and after each encounter she had with Occhiuto. As a result, Occhiuto says, there is a chance that A.J. could have concealed drugs on her person and thus a chance that Wood would not actually have seen the drugs A.J. purchased.

To bolster this argument, Occhiuto notes that Wood's testimony showed that A.J. had even, in other instances, manipulated controlled buys and stolen the government's buy money. Occhiuto thus concludes that "[t]he evidence created a concrete possibility that, for whatever reason, [A.J.] had delivered to Wood something other than the items Occhiuto had delivered to her."

But the fact that there may have been holes in Wood's testimony that drugs were exchanged and that those drugs were the drugs in evidence does not show that Wood was therefore relaying out-of-court statements from A.J. in offering that testimony. Wood based this testimony on surveillance that might have been imperfect. But, at most, the imperfections could support a challenge to the weight to be accorded Wood's testimony. They cannot show that Wood's testimony, because less than airtight, was other than it was: testimony that relayed Wood's own observations and not statements made out of court by someone Occhiuto could not confront at trial.2 Thus, there was no Confrontation Clause violation.

B.

Occhiuto next argues that the District Court deprived him of his constitutional right to present a defense by denying his request to call a particular witness, Victor Bizzell. Occhiuto contends that Bizzell's testimony would have provided "evidence in the form of an admission by [A.J.] that she ripped off the buy money in one of the controlled buys with Occhiuto." According to Occhiuto, such evidence would serve two purposes: to undermine the government's theory that the controlled drug buys had been legitimate and to show A.J. to be an unreliable cooperating informant. And each point, presumably, would have given reason to doubt Agent Wood's testimony concerning what had been exchanged during the encounters between A.J. and Occhiuto that Wood observed.

"[D]istrict courts must carefully balance an accused's right to present evidence with [other] considerations," United States v. Brown, 500 F.3d 48, 57 (1st Cir. 2007), like the "integrity of the adversary process, which depends both upon the presentation of reliable evidence and the rejection of unreliable evidence, [and] the interest in the fair and efficient administration of justice," United States v. Levy-Cordero, 67 F.3d 1002, 1013 (1st Cir.1995) (quoting Taylor v. Illinois, 484 U.S. 400, 414-15, 108 S.Ct. 646, 98 L.Ed.2d 798 (1988)). For that reason, a district court's determination as to the admissibility of witness testimony is reviewed for abuse of discretion. Brown, 500 F.3d at 57-58 & n. 4.

There was no abuse of discretion here. Occhiuto, by his own description, sought to admit Bizzell's testimony in order to prove the truth of what A.J. had said to Bizzell — namely, that she was actually involved in a scam at the time of her "controlled buys" with Occhiuto. Thus, Bizzell's testimony as to A.J.'s out-of-court statements to Bizzell — because such testimony would have been introduced to prove the truth of the matter asserted within — would have constituted hearsay. See Fed. R. Evid. 801(c). And Occhiuto has not demonstrated how Bizzell's testimony would have been admissible under any exception to the hearsay rule. In consequence, the District Court acted well within its discretion in declining to permit Bizzell's testimony.3

Occhiuto responds that Bizzell's testimony had to be admitted "irrespective of whether it[] [was] hearsay or not" given that the testimony "went to the heart of Occhiuto's defense." Occhiuto argues in this regard that evidentiary rules designed to ensure reliability can be trumped by the defendant's right to present a defense. See Chambers v. Mississippi, 410 U.S. 284, 299, 302-03, 93 S.Ct. 1038, 35 L.Ed.2d 297 (1973).

But Occhiuto's appeal to that general proposition does not explain either why the hearsay component of Bizzell's testimony would not have been available from some other source or how that testimony had other indicia of reliability and persuasiveness. See id. at 302, 93 S.Ct. 1038. And thus Occhiuto does not explain how the District Court erred in exercising its discretion to deny hearsay testimony from this particular witness, in light of the specific grounds that the District Court gave for deciding to do so.

Finally, Occhiuto suggests that Bizzell's testimony could also have aided his defense even if that testimony had been admitted only to impeach the reliability of the government's cooperating informant, A.J. Testimony about a statement, if not offered to prove the truth of the matter asserted in the statement, falls outside of the definition of hearsay. See Fed. R. Evid. 801(c). But, Bizzell's testimony — if used only for that purpose — still would raise the issue of whether it was cumulative evidence that could be excluded on that basis. See Fed. R. Evid. 403. The District Court explained that if Occhiuto sought to introduce Bizzell's testimony "simply [for] the impeachment value with respect to the information that comes from [A.J.] ... I think the record is uncontested on that." And the record supports that conclusion. Wood testified that A.J. had stolen the government's buy money on several instances and that she was a known drug user. Wood also explained that A.J. was recruited to assist the investigation because she had access to suspected drug dealers. The District Court therefore acted well within its discretion in excluding what it reasonably determined to be cumulative evidence. See United States v. Marino, 277 F.3d 11, 24 (1st Cir.2002).

II.

Occhiuto also challenges his sentence. He contends that the District Court erred in finding certain facts that it used to determine the appropriate sentence under the Sentencing Guidelines and that the imposed sentence was substantively unreasonable.

A.

A district court's factual findings at sentencing are reviewed for clear error. United States v. Ramos-Paulino, 488 F.3d 459, 463 (1st Cir.2007). Clear error will be found only when, upon whole-record review, a reviewing court "form[s] a strong, unyielding belief that a mistake has been made." United States v. Cintron-Echautegui, 604 F.3d 1, 6 (1st Cir.2010) (alteration in original) (quoting Cumpiano v. Banco Santander P.R., 902 F.2d 148, 152 (1st Cir.1990)).

The District Court had to calculate the recommended sentencing range under the Sentencing Guidelines. That calculation depended on a factual finding about the weight of drugs that Occhiuto sold. Occhiuto contends that the District Court's finding was clearly erroneous because it relied on a determination contained in the presentence report, the accuracy of which Occhiuto disputes now and challenged at the time. See United States v. Sepulveda, 15 F.3d 1161, 1198 (1st Cir. 1993).

It is true that a presentence report is not beyond scrutiny, as a "district court is obliged to resolve any genuine and material dispute on the merits." United States v. Cyr, 337 F.3d 96, 100 (1st Cir. 2003). But the dispute must be genuine, and thus "if the defendant's objections to the [presentence investigation report] are merely rhetorical and unsupported by countervailing proof, the district court is entitled to rely on the facts in the [presentence investigation report]." Id.

Here, the District Court made factual determinations based upon information in the presentence investigation report,4 which in turn reflected information from earlier forensic drug testing by the crime lab. And Occhiuto did not offer any countervailing proof to rebut the drug weights reported from that testing.5 Occhiuto does note that the drug weights testified to at trial differed from the drug weights reported in the earlier tests. But the government's expert witness explained that, because the first chemist who tested the drugs was unavailable at trial, the government undertook re-testing of the drugs in evidence. This second chemist then testified and explained that the testing procedures used up some of the drugs, so that there was a lower reserve weight at each subsequent test. At sentencing, the government reminded the District Court of the chemist's trial testimony, and the surveillance evidence corroborating the controlled buys. In crediting this explanation for the variance in reported weights, the District Court did not clearly err in making its factual finding. See Cintron-Echautegui, 604 F.3d at 6-7.

B.

Finally, Occhiuto challenges the substantive reasonableness of his sentence. The District Court imposed a sentence of 105 months, a term at the higher end of the applicable Guidelines range. The substantive reasonableness of a sentence is reviewed for abuse of discretion, considering the totality of the circumstances. United States v. Gibbons, 553 F.3d 40, 47 (1st Cir.2009).

Occhiuto argues that the District Court inadequately considered his history of mental illness and previous brain injury because the District Court did not mention these mitigating factors at sentencing. But the District Court did state at the sentencing hearing that it had read the materials submitted by the parties, and Occhiuto's materials identified these same factors. The record thus supports the conclusion that the District Court simply focused on other considerations that it implicitly deemed more important, including the defendant's history of violent behavior. See United States v. Suarez-Gonzalez, 760 F.3d 96, 102 (1st Cir.2014) ("[T]he appellant's real complaint is not that the district court overlooked or misapprehended relevant sentencing factors but, rather, that the court gave more weight to factors that the appellant regarded as unimportant and less weight to factors that the appellant regarded as salient.... [S]uch selective triage is precisely the function that a sentencing court is expected to perform.").

Occhiuto also argues that the District Court lengthened his sentence in order to promote his "rehabilitation" in violation of Tapia v. United States, ___ U.S. ___, 131 S.Ct. 2382, 180 L.Ed.2d 357 (2011). Tapia held that the Sentencing Reform Act "prevents a sentencing court from imposing or lengthening a prison term because the court thinks an offender will benefit from a prison treatment program." Tapia, 131 S.Ct. at 2392. But the record does not show that the District Court imposed the sentence — or extended its length — in order to effect the defendant's rehabilitation. The District Court instead repeatedly referred to the need to protect the public. See Sentencing Tr. at 34-35 ("There are a couple of ... statutory factors that I think are particularly important here and I agree with the government that the most important one is the need to protect the public from future crimes of the defendant.... The [defendant's] institutional history is remarkable.... So I think incapacitation is apparently the predominant nonguideline factor.").

And while the District Court did mention the "need for the defendant to get the... services that can help him eventually become a productive law-abiding citizen, including educational, vocational, and perhaps mental health treatment options," the context shows that the District Court did not make the reference in order to justify the sentence itself. In like circumstances, we have held that "no Tapia error occurs unless rehabilitative concerns are being relied upon either in deciding whether to incarcerate or in deciding the length of the incarcerative sentence to be imposed. Thus, the mere mention of rehabilitative needs, without any indication that those needs influenced the length of the sentence imposed, is not Tapia error." United States v. Del Valle-Rodríguez, 761 F.3d 171, 175 (1st Cir.2014). And so we find no such error here as well.

III.

For these reasons, we affirm Occhiuto's conviction and sentence.