In this sentencing appeal, defendant-appellant Héctor Cortés-Medina insists that his 168-month sentence is both procedurally flawed and substantively unreasonable. After careful consideration, we affirm the sentence.

This appeal has its roots in an indictment returned by a federal grand jury sitting in the District of Puerto Rico. The indictment alleged that the defendant served as an “enforcer” for a drug-trafficking ring and charged him as a participant in a conspiracy to possess with intent to distribute controlled substances within 1,000 feet of a protected location. See 21 U.S.C. §§ 841(a)(1), 846, 860.

In due course, the defendant entered into a non-binding plea agreement (the Agreement) with the government. The Agreement provided that, in exchange for his guilty plea to the conspiracy charge and to a related forfeiture allegation, the government would recommend a 121-month prison term; provided, however, that the defendant’s criminal history category (CHC) was IV or lower. The district court accepted the plea, and the probation office prepared a presentence investigation report (PSI Report). Neither side objected to anything contained in the PSI Report, which (among other things) recommended a series of guideline calculations culminating in a total offense level of 30, a CHC of IV, and a guideline sentencing range (GSR) of 135 to 168 months.

At the disposition hearing, the government recommended the agreed 121-month sentence, even though that sentence was below the nadir of the GSR. The district court heard statements from defense counsel and the defendant himself, and the *65court acknowledged the parties’ joint sentencing recommendation. The court then engaged in a dissection of the defendant’s criminal history.

To begin, the court examined the four convictions on which the defendant’s CHC was premised. It then catalogued several arrests that had terminated either in acquittals or in dismissals. These included two charges for first-degree murder, two charges relating to destruction of evidence, and an assortment of charges for drug and firearm violations.1 Noting that none of these charges had resulted in any punishment, the district court expressed frustration. The court said: “This is what I just don’t understand, how these things are happening.” It then added, cryptically, that “lightning doesn’t strike twice in the same place.”

The district court proceeded, without objection, to ratify and adopt the guideline calculations limned in the PSI Report. It stressed that the defendant was an enforcer in the drug-trafficking organization, adding “[w]e know what that means.” In the end, the court sentenced the defendant to a term of immurement at the apex of, but within, the GSR: 168 months.

This timely appeal ensued. Although the Agreement contains a waiver-of-appeal provision, that provision, by its terms, is operative only if the court sentences the defendant in accordance with the Agreement’s “terms, conditions and recommendations.” Because the sentence imposed by the district court exceeded'the sentence recommended in the Agreement, the waiver-of-appeal provision is a dead, letter. See, e.g., United States v. Fernández-Cabrera, 625 F.3d 48, 51 (1st Cir. 2010).

Overall, “[a]ppellate review of federal criminal sentences is characterized by a frank recognition of the substantial discretion vested in a sentencing court.” United States v. Flores-Machicote, 706 F.3d 16, 20 (1st Cir. 2013). The “process is bifurcated: we first determine whether the sentence imposed is procedurally reasonable and then determine whether it is substantively reasonable.” United States v. Clogston, 662 F.3d 588, 590 (1st Cir. 2011). Generally, both aspects of this review are for abuse of discretion. See Gall v. United States, 552 U.S. 38, 46, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007); United States v. Martin, 520 F.3d 87, 92 (1st Cir. 2008). When-assessing the procedural reasonableness of a sentence, however, appellate review is more nuanced: we afford de novo consideration to the sentencing court’s interpretation and application of the sentencing guidelines and assay the court’s factfinding for clear error. See Flores-Machicote, 706 F.3d at 20.

These standards of review are altered when an objection is not preserved in the court below. In that event, review is for plain error. See United States v. Duarte, 246 F.3d 56, 60 (1st Cir. 2001). Plain error review is not appellant-friendly. It “entails four showings: (1) that an error occurred (2) which was clear or obvious and which not only (3) affected the defendant’s substantial rights, but also (4) seriously impaired the fairness, integrity, or public reputation of judicial proceedings.” Id.

Against this backdrop, we turn to the defendant’s claims of error: three procedural objections and a plaint of substantive unreasonableness. We address these matters sequentially.

*66The defendant first argues that the district court erred by taking into account several dismissed or acquitted charges because the facts underlying those charges were not proven by a preponderance of the evidence. He says that he preserved this argument by means of a statement contained in the “Background of the Defendant” section of his sentencing memorandum:

As evidence showed in the court files, that were examined, many of the indictments got dismissed because of lack of proof related to the supposed direct participation of the defendant and in others there was no proof at all. During the investigations as is shown as well in the Pre-Sentence Report the defendant has been accused of many different illegal acts as, for which many of these accusation [sic] were dismissed because of insufficient evidence.

However, no mention of the dismissed or acquitted charges was made in the “Application of Law and Arguments” section of the defendant’s sentencing memorandum.

At the outset of the disposition hearing, the district court confirmed with defense counsel that the defendant had no objections to the PSI Report. During that hearing, defense counsel did not mention the dismissed or acquitted charges at all.

Generally, a party has 14 days after receipt of a presentence report within which to object in writing to, inter alia, “material information” contained in that report. Fed.R.Crim.P. 32(f)(1). A failure to object constitutes a waiver of any objection to such information. See United States v. Serrano-Mercado, 784 F.3d 838, 846, 847 (1st Cir. 2015); United States v. Turbides-Leonardo, 468 F.3d 34, 37 (1st Cir. 2006). Such a waiver occurred here.

Nor did the passing reference to the charges in the background section of the sentencing memorandum cure this omission. That reference, particularly when not followed up by some corresponding reference in the argument section of the memorandum, did nothing to call to the sentencing court’s attention that the defendant objected to any consideration of those parts of his arrest record that had not ripened into convictions. While our dissenting brother cavalierly proclaims that the argument made on appeal was “implicit in [the defendant’s] contentions” at sentencing, post at 71, a finding to that effect would render normal principles of waiver meaningless. We conclude, therefore, that the defendant’s argument is un-preserved and engenders plain error review.

We turn to that review. The defendant bases his claim of error on the Supreme Court’s opinion in United States v. Watts, 519 U.S. 148, 157, 117 S.Ct. 633, 136 L.Ed.2d 554 (1997) (per curiam). In that case, the Court concluded that, when imposing an offense-level enhancement, a sentencing court may consider acquitted conduct only if that conduct is proven by a preponderance of the evidence.2 See id. at 153, 157, 117 S.Ct. 633. Here, however, Watts is inapposite: the sentencing court did not use dismissed or acquitted conduct in its sentencing calculus. Rather, the court used the defendant’s arrest record, which was laid out in the PSI Report and not contested by the defendant. The arrest record was, therefore, a proven fact, *67see United States v. Jiménez, 512 F.3d 1, 7 (1st Cir. 2007), and thus properly before the district court.

Nor did the district court plainly err in taking into account the prolific arrest record of the defendant — a drug-gang enforcer — solely for the purpose of determining at what point within the GSR the defendant’s sentence should be set. As we previously have explained, “a criminal defendant’s ‘history and characteristics’ are among the considerations that a court ought to take into account at sentencing.” Flores-Machicote, 706 F.3d at 21 (quoting 18 U.S.C. § 3553(a)(1)). This includes the defendant’s record of past arrests or dismissed charges, as such a record “may indicate a pattern of unlawful behavior even in the absence of any convictions.” Id. (internal quotation marks omitted). That the defendant had several such arrests is an important datum, for we have distinguished a series of arrests “which might legitimately suggest a pattern of unlawful behavior even in the absence of any convictions” from, say, a single arrest. United States v. Lozada-Aponte, 689 F.3d 791, 792 (1st Cir. 2012) (quoting United States v. Zapete-Garcia, 447 F.3d 57, 61 (1st Cir. 2006) (internal quotation marks omitted)); accord United States v. Ocasio-Cancel, 727 F.3d 85, 91-92 (1st Cir. 2013). Based on these authorities, it defies reason to assert (as does our dissenting brother) that there was a clear or obvious error in the district court’s consideration of the defendant’s full arrest record at the final step of the sentencing proceeding. Consequently, we find no plain error.3 Cf. United States v. Vega-Santiago, 519 F.3d 1, 5 (1st Cir. 2008) (en banc) (“Garden variety considerations of culpability, criminal history, likelihood of re-offense, seriousness of the crime, nature of the conduct and so forth should not generally come as a surprise to trial lawyers who have prepared for sentencing.”).

The defendant’s next claim of error posits that the district court abused its discretion by not adequately considering the factors set forth in 18 U.S.C. § 3553(a). Our review of this claim is for abuse of discretion. See Gall, 552 U.S. at 51, 128 S.Ct. 586.

The defendant’s claim centers on his assertion that the district court failed to consider not only that he had already served time for a matter incident to the offense of conviction but also that he had been rehabilitated. But these potentially mitigating factors were before the district court at sentencing; indeed, they were vigorously pressed by defense counsel. There is not the slightest reason to think that the district court overlooked them.4

*68No more is needed to defeat this claim of error. Even though a sentencing court is charged with a duty to “consider all relevant section 3553(a) factors, it need not do so mechanically.” Clogston, 662 F.3d at 592 (internal quotation marks omitted). While the court below did not squarely address the two factors cited by the defendant, we have warned against “read[ing] too much into a district court’s failure to respond explicitly to particular sentencing arguments.” Id. This court has not required sentencing courts to walk, line by line, through the section 3553(a) factors. See United States v. Dixon, 449 F.3d 194, 205 (1st Cir. 2006) (explaining that a sentencing court need not “address [the section 3553(a) ] factors, one by one, in some sort of rote incantation when explicating its sentencing decision”). We have no occasion to impose such a requirement today. Thus, we discern no abuse of discretion in the sentencing court’s failure to acknowledge explicitly that it had mulled the defendant’s arguments.

The defendant’s last procedural claim implicates 18 U.S.C. § 3553(c). This statute provides in pertinent part that the court “at the time of sentencing, shall state in open court the reasons for its imposition of the particular sentence” and, if the GSR spans more than 24 months, shall also state “the reason for imposing a sentence at a particular point within the range.” 18 U.S.C. § 3553(c). The defendant says that the sentencing court did not adequately comply with these strictures and that, therefore, his sentence must be vacated.

The defendant’s premise is sound: the sentencing court’s explanation of its reason for choosing a top-of-the-range sentence of 168 months is recondite at best. But the conclusion that the defendant seeks to draw from this premise is unfounded. The defendant did not raise this objection below, and we have held that a district court’s failure to provide an adequate explanation of a sentence, without more, is not sufficient to constitute plain error. See United States v. Medina-Villegas, 700 F.3d 580, 583 (1st Cir. 2012).

Here, there is no “more”: the district court’s rationale is readily apparent from the sentencing transcript. The court made no bones about its belief that the defendant’s criminal history score underrepresented his culpability because of his pattern of arrests and the persistent lack of follow-up with respect to the charges that were initially preferred against him. Such items speak directly to the character of the individual, the risk of recidivism, and the need to protect the public from future crimes. See United States v. Rivera Calderón, 578 F.3d 78, 104-05 (1st Cir. 2009).

Transparency at sentencing is important, and we do not readily condone a district court’s failure to comply with the obligations imposed by section 3553(c). But neither do we condone a defendant’s failure to object in a seasonable manner and call such an oversight to the sentencing court’s attention in time to correct it at the disposition hearing. The failure to voice a contemporaneous objection constrains our review to plain error, and we find no plain error here. There is simply no reason to believe that if the district court had effected a more literal compliance with section 3553(c), it would have handed down a milder sentence. See Medina-Villegas, 700 F.3d at 584; United States v. Mangual-Garcia, 505 F.3d 1, 16 (1st Cir. 2007); see also Turbides-Leonardo, 468 F.3d at 39 (explaining that an appellant hoping to prevail on plain error review must show “a reasonable probability that, but for the error claimed, the result of the proceeding would have been different” (internal quotation marks and alterations omitted)).

*69The defendant’s final claim of error em-. bodies a challenge, raised for the first time on appeal, to the substantive reasonableness of his sentence. We recently have explained that, in such circumstances, the appropriate standard of appellate review is uncertain. See United States v. Vargas-García, 794 F.3d 162, 167 (1st Cir. 2015); United States v. Ruiz-Huertas, 792 F.3d 223, 228 (1st Cir. 2015). We need not resolve that uncertainty today: even assuming, favorably to the defendant, that his claim of substantive unreasonableness is reviewable for abuse of discretion, it nonetheless fails.

We start with first principles. When evaluating the substantive reasonableness of a sentence under the abuse of discretion rubric, an inquiring court must take into account the totality of the circumstances. See Martin, 520 F.3d at 92. A principal goal of sentencing is to fashion a sentence that is “sufficient, but not greater than necessary.” United States v. Carrasco-de-Jesús, 589 F.3d 22, 29 (1st Cir. 2009) (quoting 18 U.S.C. § 3553(a)). In determining whether a sentencing court has achieved this goal, we assess the plausibility of the sentencing court’s rationale and the appropriateness of the sentence itself. See id. at 30.

Challenging a sentence as substantively unreasonable is a heavy lift. That lift grows even heavier where, as here, the sentence falls within a properly calculated GSR. See Clogston, 662 F.3d at 592-93; see also United States v. Jiménez-Beltre, 440 F.3d 514, 518 (1st Cir. 2006) (en banc) (explaining importance of advisory guidelines in the sentencing calculus). Indeed, a reviewing court may apply “a presumption of reasonableness” to a within-the-range sentence. Rita v. United States, 551 U.S. 338, 347, 351, 127 S.Ct. 2456, 168 L.Ed.2d 203 (2007). At a bare minimum, a defendant “must adduce fairly powerful mitigating reasons and persuade us that the district court was unreasonable in balancing pros and cons.” Clogston, 662 F.3d at 593 (internal quotation mark omitted).

In the case at hand, the defendant asseverates that his sentence is substantively unreasonable on two fronts. He first submits that he deserves a more lenient sentence because of his rehabilitation.5 Second, he complains that he already has served a sentence in a Puerto Rico prison for a 2004 drug crime—a crime that he says is incident to the charged conspiracy.

The charge of substantive unreasonableness is futile. The offense of conviction is serious: the defendant served as an enforcer for a thriving conspiracy that sold drugs in a protected area. The defendant’s criminal history is bleak. And though his efforts at rehabilitation are laudable, the district court is in the best position to weigh the credibility of a claim of rehabilitation and to balance the sentencing scales in light of such a claim. See Gall, 552 U.S. at 51-52, 128 S.Ct. 586.

So, too, the defendant’s suggestion that the sentence imposed punishes him twice for the same criminal conduct is unavailing. In support, the defendant relies on a guideline provision, USSG § 5K2.23. That provision, however, states that a downward departure may be warranted if the defendant has completed a term of imprisonment for a crime incident to the offense of conviction and that crime “was the basis for an increase in the offense level for the instant offense.” United States v. Korne-*70gay, 410 F.3d 89, 99 (1st Cir. 2005) (emphasis omitted). Here, however, the Puérto Rico drug-trafficking conviction identified by the defendant was not assigned any criminal history points in the calculation of his CHC. Thus, that. conviction did not serve to increase his offense level, and section 5K2.23 does not apply.

To say more would be to paint the lily. Here, the sentencing court offered a plausible rationale for the sentence imposed, and that within-the-range sentence represents a defensible outcome. Having in mind the totality of the circumstances, we conclude that the district court did not abuse its considerable discretion in sentencing the defendant at the top of — but within — the GSR. In other words, the sentence was sufficient but not greater than necessary to achieve the legitimate goals of sentencing.

The fact that the parties jointly agreed to recommend a lower (downwardly variant) sentence does not alter this conclusion. In the absence of exceptional circumstances (such as the applicability of a statutory mandatory minimum sentence), the starting point for a court’s sentencing determination is the guideline range, not the parties’ recommendations. Thus, we have consistently refused to accord any decretory significance to such non-binding recommendations — or even to require a sentencing court to explain why it decided to eschew those recommendations. See Vargas-García, 794 F.3d at 167; United States v. Vega-Salgado, 769 F.3d 100, 104 (1st Cir. 2014). See Carrasco-de-Jesús, 589 F.3d at 29.

We need go no further.6 For the reasons elucidated above, the sentence is

Affirmed.

dissenting.

My colleagues treat defendant Cortés-Medina’s dismissed and acquitted charges as if they have significance separate and apart from the conduct that they may, or may not, reflect. Hence, they find no error in the district court’s reliance on the defendant’s arrest record to justify a sentence at the top of the guidelines range — a term of imprisonment nearly four years longer than the sentence recommended by the government — despite the absence of any evidence about the conduct underlying those ultimately unproven charges. Neither Supreme Court precedent nor our own cases support treating as fact mere allegations of criminal behavior that are not substantiated by at least a preponderance of the evidence. Indeed, the cases make clear that, where the evidence of culpability does not meet that level of reliability, the district court errs by factoring unproven charges into the sentence. Accordingly, Cortés-Medina is entitled to a resentencing in which the unsubstantiated charges play no role.

I. Standard of Review

My colleagues apply the plain error standard of review because Cortés-Medina did not object to the portion of his presentence investigation report (“PSR”) listing charges against him that either were dismissed or resulted in acquittal.7 I disagree that Cortés-Medina’s challenge to the court’s use of these unsubstantiated charges was unpreserved. Certainly, *71Cortés-Medina was not required to object to the inclusion of these charges in his PSR, as he has not argued that the arrests and subsequent proceedings did not occur. He did, however, point out in his sentencing memorandum the flimsy foundation for many of the charges.8 Referring to his multiple indictments in state court, he noted that “many of the indictments got dismissed because of lack of proof related to the supposed direct participation of the defendant and in others there [was] no proof at all.” He further stated that “many of these accusationfs] were dismissed because of insufficient evidence.” Although Cortés-Medina did not repeat his objection in the argument section of his memorandum, or explicitly assert that the court should not take his dismissed and acquitted charges into account, that objection and assertion are implicit in his contentions that the charges lack support.9

Moreover, even if plain error review applies, Cortés-Medina would satisfy its requirements.*7210 As I explain below, a sentencing judge may not properly rely on dismissed and acquitted charges to increase a defendant’s sentence without determining, by a preponderance of the evidence, that the charges reflect culpable conduct. Absent record support for treating the unproven charges as sufficiently well founded to meet that modest standard, a court’s use of them to justify a higher sentence is error that must be characterized as plain.11 Here, the district court invoked such charges when selecting the high end of the applicable Guidelines range, despite the Probation Office’s inability to explain the underlying conduct or give reasons for the dismissals. The error was thus manifestly prejudicial. As for the miscarriage-of-justice prong, we previously have recognized that “the difference in potential jail time would be a concern in any balance.” United States v. Ramos-Gonzalez, 775 F.3d 483, 507 (1st Cir. 2015) (quoting United States v. Torres-Rosario, 658 F.3d 110, 117 (1st Cir. 2011)).

Accordingly, under either standard of review, Cortés-Medina prevails on his claim of procedural error.

II. The Standard of Reliability

This should be an easy case for concluding that a remand is necessary because, as the majority acknowledges, the Probation Office was unable to obtain information about the conduct underlying the unproven or acquitted charges reported in Cortés-Medina’s PSR. The district court thus had no evidence that those charges in fact re-fleeted criminal behavior. As the majority reports, the court nonetheless chose a sentence based, in part, on Cortés-Medina’s “pattern of arrests and persistent lack of follow-up with respect to the charges that were initially preferred against him.” My colleagues have no problem with that rationale, quoting language from one of our precedents stating that a record of arrests or dismissed charges “may indicate 'a pattern of unlawful behavior even in the absence of convictions.’” United States v. Flores-Machicote, 706 F.3d 16, 21 (1st Cir. 2013) (quoting United States v. Lozada-Aponte, 689 F.3d 791, 792 (1st Cir. 2012)). They maintain that, because Cortés-Medina has not contested his arrest record, “[t]he arrest record was ... a proven fact, and thus properly before the district court.” (Citation omitted.)

I agree with the general proposition that past criminal acts that did not result in conviction may be given weight in sentencing determinations if the available proof of those criminal acts meets some standard of reliability. I do not understand my colleagues to disagree with that proposition. Nevertheless, when the defendant’s criminal history contains a series of dismissed or acquitted charges, my colleagues hold that a district court may accord that pattern of alleged criminal activity probative weight based solely on the arrests, even if the court has no facts substantiating the conduct underlying the alleged charges.

The majority is wrong. Sentencing based on unexamined allegations of criminal behavior is not permitted by our own *73or Supreme Court precedent. In articulating their view, my colleagues summarily dismiss as inapposite the Supreme Court’s decision in United States v. Watts, 519 U.S. 148, 117 S.Ct. 633, 136 L.Ed.2d 554 (1997) (per curiam). There, the Court held that a sentencing court may consider acquitted conduct to impose an offense-level enhancement as long as that conduct is proven by a preponderance of the evidence. Id. at 157, 117 S.Ct. 633. The majority insists that Watts is inapplicable because the district court in this case relied on Cortés-Medina’s pattern of dismissed or acquitted charges, not his conduct. And, they say, because CortésMedina’s record of past charges was “a proven fact,” no error occurred.12

This attempt to distinguish Watts is misguided. A series of arrests or past charges is inseparable from the underlying conduct. Watts refers to the defendant’s conduct because that is the pertinent information in assessing whether the defendant’s past criminal activity is understated by reference to his convictions alone. The relevant question cannot be whether the defendant’s non-conviction arrests and prosecutions are reported accurately — in all likelihood,' the PSR will contain an accurate record of detentions and charges — but whether the defendant in fact committed the listed offenses notwithstanding the absence of convictions for those crimes. Indeed, the defendant’s conduct must be the focus because, where the charges initially filed against him did not bear fruit, they do not demonstrate culpability. A sentencing court cannot simply presume that the lack of convictions is attributable to flawed or lax prosecutorial or judicial systems rather than the defendant’s innocence. Sometimes, systemic flaws lead to arrests without justification. See United States v. Za-pete-Garcia, 447 F.3d 57, 61 (1st Cir. 2006) (noting that “arrest ‘happens to the innocent as well as the guilty’ ” (quoting Michelson v. United States, 335 U.S. 469, 482, 69 S.Ct. 213, 93 L.Ed. 168 (1948))).

The next question, then, is what standard of reliability applies to the inquiry into the conduct underlying unproven charges. As I explain below, the answer is unequivocally provided by both Supreme Court and First Circuit caselaw..

A. The Teaching of Watts

The Supreme Court’s decision in Watts confirms that a sentencing court may not give weight to unproven crimes — whether uncharged, dismissed, or acquitted — unless the court finds by at least a preponderance of the evidence that the conduct underlying those charges occurred. In Watts, the Court rejected an argument that principles of due process foreclose reliance on acquitted conduct to calculate the Guidelines range, stating that “a jury’s verdict of acquittal does not prevent the sentencing court from considering conduct underlying the acquitted charge, so long as that conduct has been proved by a preponderance of the evidence.” 519 U.S. at 157, 117 S.Ct. 633 (emphasis added). By including this qualification, the Court reaffirmed its prior holding that “application of the preponderance standard at sentencing generally satisfies due process.” Id. at 156, 117 S.Ct. 633 (citing McMillan v. Pennsylvania, 477 U.S. 79, 91-92, 106 S.Ct. 2411, 91 L.Ed.2d 67 (1986)).

Although the focus in Watts was on the use of acquitted conduct to set the Guidelines range, the Court did not suggest that a standard less demanding than prepon*74derance-of-the-evidence applies to the use of acquitted conduct — or any other unproven criminal activity — in choosing a sentence within the range.13 To the contrary, multiple statements in the Watts opinions reflect an assumption that any facts used in sentencing — pertaining to allegations of past criminal conduct, or otherwise — must be proven by a preponderance of the evidence or an even higher standard of reliability. First, the Court quotes commentary from Guidelines § 6A1.3 stating that “it is ‘appropriate’ that facts relevant to sentencing be proved by a preponderance of the evidence,” 519 U.S. at 156, 117 S.Ct. 633, and the majority goes on to make the observation quoted above linking the preponderance standard with the requirements of due process. Id.14 In addition, as described above, the Court framed its holding in Watts broadly, without any suggestion that the preponderance standard applies only for the purpose of selecting the Guidelines range: a sentencing court is permitted, in general, to consider “conduct underlying the acquitted charge, so long as that conduct has been proved by a preponderance of the evidence.” Id. at 157, 117 S.Ct. 633.

Second, Justice Scalia points out that the preponderance of the evidence standard — the measure of reliability the Court has endorsed for other sentencing facts— is also consistent with due process for conduct underlying an acquittal. He asserts that neither the Sentencing Commission nor the courts may entirely exclude from the sentencing calculus “information which would otherwise justify enhancement of sentence or upward departure,” or impose “some higher standard of probative worth than the Constitution and laws require,” simply because that information “pertains to acquitted conduct.” See id. at 158, 117 S.Ct. 633 (Scalia, J., concurring).15

Third, and consistently, the Watts Court acknowledged the possibility that, in some circumstances, the more demanding clear- and-convineing evidence standard might be appropriate. Id. at 156-57, 117 S.Ct. 633. In a lengthy footnote citing cases reflecting “a divergence of opinion among the Circuits,” id. at 156, 117 S.Ct. 633, the Court quotes an Eighth Circuit case characterizing the Supreme Court’s McMillan decision as approving the preponderance standard only “ ‘for garden variety sentencing determinations,’ ” id. at 156 n. 2, 117 S.Ct. 633 (quoting United States v. Townley, 929 F.2d 365, 369 (8th Cir. 1991)). *75In other words, the Court in Watts considered the possibility that, at times, an assessment more reliable than the preponderance standard might be applicable to sentencing facts. Neither the Court nor the circuits it quoted in Watts contemplated the possibility of proof less reliable than preponderance of the evidence. This view that Waits reaffirms preponderance of the evidence as the minimum standard of reliability is also reflected in academic literature. See, e.g., Claire McKusker Murray, Hard Cases Make Good Law: The Intellectual History of Prior Acquittal Sentencing, 84 St. John’s L.Rev. 1415, 1468 (2010) (“Under Watts, prior acquittal sentencing is permitted but not mandated, and a hard floor of reliability is established in the form of the requirement that prior acquitted conduct be proved to a preponderance of the evidence.”).

B. First Circuit Law

The preponderance-of-the-evidence baseline for considering sentencing facts is also well established in our circuit. Indeed, we have applied the standard in this very context, i.e., to the choice of sentence within the Guidelines range where the court sought to rely on unproven criminal conduct. See United States v. Lombard, 102 F.3d 1, 4 (1st Cir. 1996) (“[T]he district court may ... choose to give weight to the uncharged offenses in fixing the sentence within the statutory range if it finds by a preponderance of evidence that they occurred .... ”); see also United States v. Munyenyezi, 781 F.3d 532, 544 (1st Cir. 2015) (“[A] judge can find facts for sentencing purposes by a preponderance of the evidence, so long as those facts do not affect either the statutory minimum or the statutory maximum.... ” (citations omitted)); United States v. Fermin, 771 F.3d 71, 82 (1st Cir. 2014) (“While the jury must, of course, find facts beyond a reasonable doubt, a preponderance-of-the-evidence standard applies to the sentencing court’s factual findings.”); United States v. Gobbi 471 F.3d 302, 314 (1st Cir. 2006) (stating that “acquitted conduct, if proved by a preponderance of the evidence, still may form the basis for a sentencing enhancement”).

I recognize that, although this standard of reliability is well established, we have not always used the words “preponderance of the evidence” when considering a district court’s reliance on charges that did not lead to conviction. See, e.g., Flores-Machicote, 706 F.3d 16; Lozada-Aponte, 689 F.3d at 792; Zapete-Garcia, 447 F.3d at 61. Nonetheless, we have applied that standard even when we have not referred to it by “name,” routinely scrutinizing the facts underlying the unproven criminal charges to ensure the necessary degree of reliability. See, e.g., United States v. Hinkley, 803 F.3d 85, 93 (1st Cir. 2015) (upholding court’s reliance on reports of inappropriate sexual contact with minors where district court “found that it was reasonable to rely on the experience of the detective who prepared the police reports” and where “certain details reported by [a victim] made the reports ‘almost self-authenticating’ ”); United States v. Díaz-Arroyo, 797 F.3d 125, 127, 130 n. 3 (1st Cir. 2015) (noting prosecutor’s explanation that charges for murder and attempted murder were dropped “only after the sole surviving witness to the incident (a minor who was able positively to identify the defendant as the shooter) was threatened and fled the jurisdiction,” and that defense counsel “did not directly challenge the prosecutor’s account of the circumstances surrounding the dismissal of the charges”); Flores-Machicote, 706 F.3d at 21 (noting that the district court “went to considerable lengths to walk through the defendant’s prior interactions with the law ... [and] explained, in some detail, why [it] *76believed the outcome of these interactions underrepresented the seriousness of the defendant’s past criminal conduct”); United States v. Gallardo-Ortiz, 666 F.3d 808, 814-15 (1st Cir. 2012) (noting that the district court took into account, inter alia, that numerous charges were dismissed on speedy trial grounds (i.e., not the merits), and rejecting defendant’s contention that the court relied on “the dismissed charges when concluding that he displayed a violent character”); United States v. Tabares, 951 F.2d 405, 411 (1st Cir. 1991) (noting that some charges were dismissed “not because of any finding on the merits of the case,” but because the defendant was deported, and that defendant did not “deny the facts, as set forth in the presentence report, upon which these charges rested”).

C. Applying the Standard

Despite the precedent described above, my colleagues accept a bare list of past arrests and charges as sufficiently reliable evidence that the defendant did, in fact, commit the crimes for which he was charged but not convicted. They claim that ample precedent in our circuit recognizes that a series of arrests — as distinguished from a single arrest — “might legitimately suggest a pattern of unlawful behavior even in the absence of any convictions.” Lozada-Aponte, 689 F.3d at 792 (quoting Zapete-Garcia, 447 F.3d at 61). They assert that Cortés-Medina’s “pattern of arrests” and the “lack of follow-up” with respect to the charges filed against him “speak directly to the character of the individual, the risk of recidivism, and the need to protect the public from future crimes.”

Yet, even where a defendant’s record contains a multitude of allegations of criminal conduct, the district court — and we in turn — must be certain that a preponderance of the evidence supports a conclusion that the allegations have merit. See Gal-lardo-Ortiz, 666 F.3d at 815 (“We have cautioned against district courts relying on mere arrests as indicative of a defendant’s character to justify an upward departure from the GSR since a criminal charge alone does not equate with criminal guilt of the charged conduct.”); Zapete-Garcia, 447 F.3d at 61 (noting a guideline policy statement that “highlights] the important distinction between direct evidence of past criminal behavior and mere arrests that may or may not have been the result of wrongdoing”). Here, the district court had no evidence concerning the conduct underlying the various charges against Cortés-Medina that ultimately were dismissed or resulted in acquittal. At the sentencing hearing, after listing those charges and noting the absence of explanation for the dismissals, the district court merely voiced its “firm belie[f] that lightning doesn’t strike twice in the same place.”

Presumably, the court meant to offer a different metaphor — “where there’s smoke, there’s fire” — to say that the unproven charges had substance because Cortés-Medina had other,' similar criminal convictions and also admitted participating in the drug conspiracy charged in this case. When additional years of incarceration are in the balance, however, due process requires more than metaphors. The fire that the district court infers — past criminal conduct — must be based on a preponderance of the evidence.

The cases cited by the majority to support the use of a series of arrests as a proxy for culpability do not hold otherwise. As described above, the panel in Flores-Machicote noted the district court’s careful examination of the defendant’s prior interactions with the law. 706 F.3d at 21. In Zapete-Garcia, the panel rejected reliance on a single arrest that occurred more than *77a decade earlier. 447 F.3d at 60-61. In the remaining two cases, where the court gave cursory attention to the district court’s reliance on a series of past arrests, it is nonetheless evident that the district courts had been presented with facts about the underlying conduct. See United States v. Ocasio-Cancel, 727 F.3d 85, 91-92 (1st Cir. 2013) (indicating that the defendant’s PSR contained detail on the events giving rise to the dismissed charges and noting that the defendant did not object to “any aspect” of the discussion); Lozada-Aponte, 689 F.3d at 792 (referring to “Lozada’s frequent run-ins with law enforcement in Florida, Illinois, and Puerto Rico, some of which apparently involved firearms”). I do not dispute that the district court may rely on a PSR’s depiction of the circumstances that led to the dismissed charges as sufficiently reliable evidence of the conduct to meet the preponderance of the evidence standard, particularly in the absence of any objection.

No such evidentiary support exists here. Cortés-Medina’s PSR contains an unelabo-rated list of his dismissed and acquitted charges, with notations stating that “Court documents were requested but have not been received.” The PSR states that some of the charges were dismissed for lack of probable cause, while others are simply described as “dismissed.” The district court thus relied improperly on those charges to sentence Cortés-Medina to a longer term of imprisonment than it otherwise would have imposed.

III. Conclusion

A district court may not rely on a defendant’s unproven past criminal activity to increase his sentence for a later crime unless the court determines, by a preponderance of the evidence, the prior criminal conduct in fact occurred. This requirement applies equally to a single instance of prior criminal activity and to a series of alleged crimes. Invocation of a pattern does not eliminate the need to examine each unproven criminal charge under the preponderance of the evidence standard.

In this case, the majority concedes that no factual support was offered to substantiate the charges on which the district court relied. The Probation Office has also acknowledged that it tried, but failed, to obtain the supporting information. Hence, on this record, defendant Cortés-Medina is entitled to resentencing without reliance on the dismissed and acquitted charges. I therefore respectfully dissent from my colleagues’ conclusion to the contrary.