United States v. Tyrique Davis: Post-2024 §2B1.1 “Loss” Is Guideline Text, and Large Upward Variances May Rest on Similarity/Recency of Prior Crimes Beyond Criminal-History Scoring

I. Introduction

In United States v. Tyrique Davis (6th Cir. Feb. 10, 2026) (unpublished), the Sixth Circuit affirmed a 48-month sentence for bank larceny involving an attempted ATM theft (Guidelines range: 21–27 months). The defendant, Tyrique K. Davis, challenged both the procedural and substantive reasonableness of an upward variance.

The appeal presented recurring sentencing issues—loss calculation under U.S.S.G. § 2B1.1, “double-counting” concerns when a court relies on criminal history, mitigation based on age, and the weight given to disparity data under 18 U.S.C. § 3553(a)(6). It also implicated an important post-2024 Guidelines development: the “loss” definitions previously located in commentary were moved into the guideline text.

The parties were the United States (appellee) and Davis (appellant), who pleaded guilty without a plea agreement after being apprehended during an ATM break-in at a Giant Eagle supermarket in Summit County, Ohio.

II. Summary of the Opinion

The Sixth Circuit affirmed. It held:

• The district court properly included both ATM damage costs and the cash removed from the ATM in the loss amount, supporting a six-level increase under § 2B1.1(b)(1).
• Davis’s argument that “loss” is defined only in commentary failed because the 2024 Guidelines place the definition in the guideline itself, not merely in commentary.
• The upward variance did not rest on impermissible double-counting: the district court emphasized distinct aspects of Davis’s criminal history (violence, similarity and temporal proximity of a prior attempted ATM theft, and poor probation performance) that are not captured with that granularity by criminal-history scoring.
• The court adequately addressed mitigation (including age) and did not clearly err in finding Davis posed a danger to the community.
• The 48-month sentence—though a large variance—was substantively reasonable in light of the totality of the § 3553(a) factors, particularly deterrence and protection of the public.

III. Analysis

A. Precedents Cited

The panel situated its review within well-established sentencing-review doctrine:

• Appellate framework and deference. Gall v. United States supplied the “deferential abuse-of-discretion” framework for reviewing sentences and required deference to a district court’s balancing of factors, even when a sentence varies from the Guidelines. The panel also relied on United States v. Wilson (quoting United States v. Elias) for the Sixth Circuit’s articulation of “abuse of discretion.”
• Procedural error taxonomy. The opinion drew its catalog of procedural errors from United States v. Rayyan and United States v. Rosenbaum—miscalculation, mandatory treatment of Guidelines, ignoring factors, erroneous facts, impermissible factors, failure to address non-frivolous arguments, or inadequate explanation.
• Loss methodology vs. loss factfinding. On the standard of review for loss: de novo for methodology and clear error for factual calculation, the court cited United States v. Maddux (citing United States v. Warshak).
• Post-2024 “loss” definition—commentary dispute avoided. The opinion’s key doctrinal move relied on the Sentencing Commission’s relocation of the “loss” definition into the guideline text, referencing United States v. Prather and the Commission’s notice-and-comment materials. This allowed the court to treat Davis’s “commentary-only” argument as a non-starter under the 2024 Manual applied.
• Deference to loss estimates. For appellate deference to plausible loss computations, the court cited United States v. Mitan (quoting United States v. Estrada-Gonzalez), and the “definite and firm conviction” standard from United States v. Karasarides (quoting United States v. White). For the “universe of acceptable computations,” it invoked United States v. Mahbub (quoting United States v. Raithatha).
• Double-counting vs. distinct aspects of conduct. On criminal history and “double-counting,” the panel cited: United States v. Lanning (with United States v. Battaglia), then refined the principle through United States v. Boucher (citing United States v. Warren), and United States v. Morgan distinguishing punishing the “precisely same aspect” twice from relying on distinct aspects.
• Similarity/recency as a valid reason to rely on criminal history beyond scoring. The opinion leaned heavily on United States v. Lee and United States v. Axline for the idea that similarities between past and present offenses, and temporal proximity, can demonstrate a need for deterrence beyond what the Guidelines already capture.
• Duty to address mitigation arguments. The court applied United States v. Wallace (quoting United States v. Gapinski), and the “apparent from the transcript as a whole” approach from United States v. Haile (quoting United States v. Brinda). On plain-error review for arguments not squarely presented, it used United States v. Taylor.
• Clearly erroneous facts and community danger. For clear-error review of factual findings at sentencing, it cited United States v. Vowels.
• Substantive reasonableness and factor-weighting. The opinion relied on United States v. Adkins (quoting United States v. Brooks) for when a sentence is substantively unreasonable and for the notion that appellate courts do not re-weigh factors. It also cited United States v. Sexton and United States v. Ely on deference to factor balancing. For district courts’ superior position, it invoked Kimbrough v. United States, Rita v. United States, and Gall v. United States. The “greater than necessary” framing came from United States v. Axline (quoting United States v. Vowell).
• Variance magnitude and justification. The requirement that justification be sufficiently compelling for the degree of variance was drawn from United States v. Aleo (quoting Gall v. United States) and referenced United States v. Harrison (order). The “larger variance needs more compelling justification” line came from United States v. Aleo (quoting United States v. Poynter).
• “Heartland” and closer review of mine-run cases. The “heartland” concept and “closer review” for mine-run deviations came from United States v. Herrera-Zuniga (quoting Kimbrough v. United States), and the requirement to explain why a case is outside the heartland was attributed to United States v. Axline (quoting United States v. Boucher).
• Comparative variance examples. The court contextualized the variance by citing affirmances in United States v. Noble, United States v. Harrison, United States v. Smith, United States v. Fievet, and United States v. Johnson.
• Disparity data as a “starting point.” On how to use sentencing data under § 3553(a)(6), it cited United States v. Perez-Rodriguez (quoting United States v. Stock), and reiterated that divergence from norms is not necessarily an abuse of discretion (citing Axline).
• Recent Sixth Circuit deference in outside-heartland cases. In concluding the case was outside the heartland and deferring accordingly, it cited United States v. Penny and United States v. Solano-Rosales.

B. Legal Reasoning

1. Loss calculation after the 2024 Guidelines revision. Davis attempted a familiar line of attack: that “loss” is defined only through commentary and thus should not drive an enhancement. The panel treated that argument as obsolete under the 2024 Manual because the Sentencing Commission relocated the definitions (actual loss, intended loss, reasonably foreseeable pecuniary harm) into the guideline itself. The court therefore reviewed the district court’s application as an application of guideline text, not commentary.
2. Including the ATM cash as “intended loss.” On the facts, the district court found—and the Sixth Circuit accepted—that Davis intended to steal the cash inside the ATM, whatever its precise amount. The cash boxes in fact contained $45,346, and the participants took them into their possession, losing the proceeds only because police intervened. With no alternative “intended loss” figure proposed beyond machine damage, the district court’s estimate was “plausible” and comfortably within the range of permissible computations.
3. Upward variance and “double-counting.” The core procedural challenge to the variance was framed as double-counting criminal history. The panel acknowledged the general caution that criminal history is “typically” not a proper reason to vary when the Guidelines already account for it, but it applied Sixth Circuit doctrine allowing reliance on criminal history when the court is reacting to aspects not meaningfully captured by the criminal-history score.
   • Granularity matters. Criminal-history points under U.S.S.G. § 4A1.1 do not differentiate in a fine-grained way among (i) violent vs. non-violent priors, (ii) priors that closely mirror the instant offense vs. unrelated priors, (iii) priors that are temporally proximate vs. remote, or (iv) patterns of probation noncompliance.
   • Similarity + recency as a deterrence signal. Following Lee and Axline, the district court could permissibly treat the near-identical, recent attempted ATM theft—followed by quick recidivism—as evidence that ordinary guideline deterrence had failed.
   This reasoning let the variance rest on “distinct aspects” rather than punishing the “precisely same aspect” twice.
4. Mitigation (age) and explanation sufficiency. Although the district court did not provide an extended discussion of youth as mitigation, the panel found that it said enough—acknowledging juvenile offending at “younger ages of 14 and 15”—to show the argument was considered, with the reasons for rejecting a lower sentence apparent from the transcript as a whole. To the extent Davis argued that age 26 at the time of the instant offense demanded mitigation, the panel applied plain-error constraints because the argument was not presented “in earnest” in that form at sentencing.
5. “Danger to the community” is not limited to formally violent offenses. The panel rejected the idea that only “offenses of violence” can support a finding of community danger. It emphasized that an ATM theft scheme can become dangerous due to police involvement and flight, and it noted other risk-relevant facts (including stolen vehicles and a violent background) supporting the district court’s assessment.
6. Substantive reasonableness despite a large variance. The panel recognized that a 48-month sentence was a substantial (about 78%) upward variance from the top of the range, requiring a compelling justification under Aleo/Poynter/Gall. It nonetheless deferred because the district court articulated why this defendant fell outside the §2B1.1 heartland: repeated similar conduct, short time since release from a lengthy prison term for an almost identical scheme, a violent criminal history, and failures under supervision—collectively supporting deterrence and public protection as overriding concerns.

C. Impact

• Post-2024 loss litigation narrows (at least for §2B1.1). By emphasizing that “loss” is now defined in the guideline text, the decision reduces the force of attacks premised on “commentary-only” definitions in cases sentenced under the 2024 Manual. Practically, defendants must contest the application of the definition (foreseeability, intent, proof of amounts), not the definitional source.
• ATM/theft cases: contents can be “intended loss” even when the amount is unknown ex ante. The court’s approach supports including the full amount of cash actually present and targeted—especially where the defendant offers no principled limiting figure and the plan was to take “whatever was inside.”
• Upward variances: courts may rely on criminal history when the Guidelines fail to capture “why this history matters now.” The decision reinforces a Sixth Circuit throughline (notably Lee and Axline): similarity and temporal proximity between prior and current offenses can justify treating deterrence as inadequately served by the advisory range. Expect future sentencing arguments to focus on whether the district court identified truly “distinct aspects” versus merely repeating what the criminal-history category already reflects.
• Disparity statistics remain a “starting point,” not a ceiling. Even where PSR data show that similarly scored defendants typically receive much less time, district courts may depart or vary if they explain why the comparator set is not meaningfully “similar” in the ways that matter under § 3553(a).

IV. Complex Concepts Simplified

• Procedural vs. substantive reasonableness. Procedural review asks whether the court followed the right steps (correct Guidelines calculation, considered the right factors, relied on accurate facts, explained itself). Substantive review asks whether the final sentence is too long (or too short) given the totality of circumstances.
• Actual loss vs. intended loss (under §2B1.1). “Actual loss” is the reasonably foreseeable monetary harm that actually resulted. “Intended loss” is the monetary harm the defendant purposely sought to cause—even if completing it was unlikely or impossible. Under the 2024 Manual, these definitions are placed in the guideline text.
• Variance. A variance is a sentence outside the Guidelines range based on the statutory factors in 18 U.S.C. § 3553(a), rather than a Guidelines “departure” based on a Guidelines-specific adjustment.
• Double-counting. Impermissible double-counting occurs when the court punishes the same aspect of conduct twice in two different ways. It is generally permissible, however, to use criminal history to justify a variance when the court is focused on distinct, case-specific features that the criminal-history score does not meaningfully reflect (e.g., close similarity and recency showing deterrence failure).
• “Heartland” / “mine-run.” The “heartland” is the set of typical cases contemplated by a Guideline. If a case is outside that typical set, a district court may have stronger grounds to deviate—so long as it explains why the case is atypical.
• Plain error. If a defendant did not adequately present an argument below, appellate review is limited; the defendant must show an obvious error that likely affected the outcome and undermined the fairness of the proceeding.

V. Conclusion

United States v. Tyrique Davis affirms two practical sentencing propositions in the Sixth Circuit. First, under the 2024 Guidelines Manual, “loss” for § 2B1.1 is defined in the guideline itself, substantially weakening challenges premised on commentary-only definitions and reinforcing inclusion of targeted ATM cash as intended loss. Second, a sizable upward variance may be sustained where the district court ties the variance to features of criminal history that demonstrate deterrence failure beyond what criminal-history points capture—especially similarity and temporal proximity to prior, near-identical conduct—while adequately addressing mitigation and disparity concerns.

Note: The opinion is labeled “NOT RECOMMENDED FOR PUBLICATION,” so it is nonprecedential, but it is still instructive for how the Sixth Circuit applies its existing sentencing framework to post-2024 Guidelines text and to large variances in repeat-pattern property crimes with aggravating history.