NOT RECOMMENDED FOR PUBLICATION File Name: 25a0425n.06
No. 23-5579
UNITED STATES COURT OF APPEALS FILED FOR THE SIXTH CIRCUIT Sep 16, 2025 KELLY L. STEPHENS, Clerk
)
UNITED STATES OF AMERICA, )
Plaintiff-Appellee, ) ON APPEAL FROM THE ) UNITED STATES DISTRICT v. ) COURT FOR THE EASTERN ) DISTRICT OF TENNESSEE CODY BALLINGER, )
Defendant-Appellant. ) OPINION )
Before: COLE, GIBBONS, and BUSH, Circuit Judges.
BUSH, Circuit Judge. Defendant-Appellant Cody Ballinger pleaded guilty to a single count of being a felon in possession of a firearm in violation of 18 U.S.C. § 922(g)(1). The district court sentenced Ballinger to 180 months imprisonment based on an enhancement under the mandatory minimum of the Armed Career Criminal Act (ACCA). Ballinger now appeals his sentence. We AFFIRM.
I.
Cody Ballinger cooperated when detectives arrived at his girlfriend’s home in July 2021.
After speaking with the detectives, Ballinger led them to the woods near the house. There they found a black duffel bag with a pistol inside. On the pistol were Ballinger’s fingerprints. Given his past felonies, the government planned to charge Ballinger as a felon in possession of a firearm in violation of 18 U.S.C. § 922(g)(1).
In February 2022, a federal grand jury indicted Ballinger of that single count. Close to a year later, Ballinger pleaded guilty to the charged offense. During the change-of-plea hearing, the Government informed Ballinger that he could face a higher sentence if he was found to be an Armed Career Criminal at sentencing. Ballinger did not object and acknowledged the possible consequences of his guilty plea.
After the guilty plea, the probation office filed a Presentence Report and a Revised Presentence Report (PSR). In the PSR, the probation office listed Ballinger’s twenty-two prior felony convictions, eleven of which were for various forms of burglary. Each of Ballinger’s burglaries took place in Bradley County, Tennessee, with nine separate victims noted. The dates of these burglaries bear on this appeal, so we recount them in full:
1. Burglary—Other than Habitation: November 22, 2010, convicted on February 28, 2011;
2. Burglary: January 4, 2016, convicted on November 14, 2018;
3. Aggravated Burglary: July 1, 2016, convicted on November 14, 2018;
4. Burglary: July 1, 2016, convicted on November 14, 2018;
5. Burglary: July 1, 2016, convicted on November 14, 2018;
6. Burglary: July 4, 2016, convicted on November 14, 2018;
7. Aggravated Burglary: July 6, 2016, convicted on November 14, 2018;
8. Aggravated Burglary: October 14, 2016, convicted on November 14, 2018;
9. Burglary: October 15, 2016, convicted on November 14, 2018;
10. Burglary: October 15, 2016, convicted on November 14, 2018;
11. Burglary: October 17, 2016, convicted on November 14, 2018.
Ballinger objected to the PSR in May 2023, but he did not dispute or object to the characterization of his criminal history. At sentencing, the district court found that the dates listed in the PSR accurately reflect the facts of Ballinger’s convictions. Once again, Ballinger did not object or point to any inaccuracy in the PSR’s recitation. Because this sentencing took place before
3
the Supreme Court decided Erlinger v. United States, 602 U.S. 821 (2024), the district court made the determination from the PSR that Ballinger qualified as an Armed Career Criminal under the ACCA and imposed a mandatory minimum sentence of 180-months’ imprisonment.
II.
The ACCA imposes an enhanced mandatory minimum sentence for felons found in possession of a firearm who have “three previous convictions” for a “violent felony or a serious drug offense, or both,” that were “committed on occasions different from one another[.]” 18 U.S.C. § 924(e)(1). The ACCA defines violent felonies as those crimes punishable by more than one-year imprisonment and involve the use, attempted use, or threat to use “physical force against the person of another; or . . . is burglary, arson, or extortion[.]” Id. § 924(e)(2)(B) (emphasis added). Ballinger does not dispute that his previous burglary convictions constitute violent felonies under the ACCA. He instead focuses on the different-occasions element of the ACCA enhancement.
In 2022, at the time of Ballinger’s sentencing, our precedent allowed district judges to determine whether predicate offenses were committed on different occasions. See United States v. Williams, 39 F.4th 342, 351 (6th Cir. 2022). Erlinger abrogated this rule. 602 U.S. at 833. There, the Supreme Court held that the question of whether predicate violent felonies were committed on different occasions “must be submitted to a jury and found unanimously and beyond a reasonable doubt.” Id. at 833–34 (quotation omitted).
In this appeal, Ballinger asserts two errors based on Erlinger. First, Ballinger claims that applying the ACCA enhancement to his sentence amounts to structural error because the government failed to charge the ACCA enhancement in the indictment and failed to obtain a guilty plea where Ballinger knowingly “admits all the elements of the enhanced offense[.]”
4
Appellant Br. 11. Second, Ballinger claims that even if harmless error review applies, the error here was not harmless.
We find neither argument persuasive.
A.
We have already decided that Erlinger errors are not structural. United States v. Campbell, 122 F.4th 624, 630–31 (6th Cir. 2024); see also United States v. Thomas, 142 F.4th 412, 417 (6th Cir. 2025). Campbell considered whether the failure to submit the different-occasions question to the jury amounted to a structural error—and concluded that it did not. 122 F.4th at 630–31. Following Campbell, the Erlinger error here was not structural.
B.
Because Ballinger’s first claim of error fails, we turn to the second question: whether the error was harmless. Errors are harmless only “if the court can conclude beyond a reasonable doubt that the jury would have found the defendant’s offenses occurred on different occasions.” United States v. Durham, —F.4th—, No. 23-5162/5173, 2025 WL 2355998, at *2 (6th Cir. Aug. 14, 2025) (per curiam). Given Ballinger’s criminal history, we have no difficulty reaching that conclusion here.
Ballinger makes two arguments to support his cause: (1) the prior convictions could not have been presented through admissible evidence because Erlinger prohibited the use of Shepard documents to determine facts about prior offenses, and (2) some juries have acquitted defendants of ACCA offenses by finding that a long lapse in time was a single occasion. We reject both arguments.
Ballinger claims that Erlinger “expressly recogniz[ed] that Shepard documents are inherently unreliable” in this context. Appellant Br. 20. Once again, our precedent forecloses this
5
argument. Shepard documents may still be relied on to conduct a harmless error review in the different-occasions context. Thomas, 142 F.4th at 418–19. This includes information contained in a PSR. Id. at 419. Erlinger only clarifies that courts should take caution with Shepard documents if a court has reason to “question the reliability of the PSR’s findings.” Id. But absent the defendant providing reason to doubt the information provided in the PSR, courts can still rely on the PSR when conducting a harmless-error inquiry. Id.; see United States v. Geerken, 506 F.3d 461, 467 (6th Cir. 2007) (“When a defendant fails to produce any evidence to contradict the facts set forth in the PSR, a district court is entitled to rely on those facts when sentencing the defendant.”).
Now we apply this to Ballinger. Ballinger has provided no reason to doubt the dates and locations of his previous offenses laid out in the PSR. He makes the same critique of the PSR that Erlinger discussed for all Shepard documents. He does not provide a specific critique of the reliability of this PSR. Because Ballinger did not provide a basis for doubt, the district court was entitled to rely on the PSR to inform itself of the basic facts relating to his predicate crimes. Thomas, 142 F.4th at 419.
We also reject Ballinger’s second argument that this could not be harmless error because some juries have held that large gaps in time still constitute a single occasion. That argument misapprehends the nature of the analysis. When courts ask what a reasonable jury would have done, they are not asking empirically what past juries have done, but whether the evidence “could rationally lead to a contrary finding with respect to the omitted element.” Neder v. United States, 527 U.S. 1, 19 (1999). The question is an objective one, and we have already rejected the use of actual jury decisions to inform the analysis. See Campbell, 122 F.4th at 633 (“After all, if only a
6
theoretical possibility of acquittal were sufficient to defeat demonstrated harmless error, harmless error could not exist.”); Thomas, 142 F.4th at 419.
Consider this issue in a different context: a manifest-weight-of-the-evidence challenge to a jury verdict. When a party challenges a jury verdict as being against the manifest weight of the evidence, we ask whether a “reasonable juror could reach the challenged verdict[.]” Debity v. Monroe Cnty. Bd. Of Educ., 134 F.4th 389, 404 (6th Cir. 2025) (quoting Barnes v. Owens-Corning Fiberglas Corp., 201 F.3d 815, 821 (6th Cir. 2000)). We conduct this review after the jury already reached the disputed verdict. But when undertaking this review, we have never concluded that because a jury, in fact, reached the challenged verdict, a reasonable jury must be able to reach the same verdict. Looking to real world juries would defeat the entire inquiry. Ballinger’s argument to the contrary results in a circularity that no case requires. Instead, as with manifest-weight-of- the-evidence challenges, we examine the harmless error question through the objective lens of a reasonable jury. See United States v. Cogdill, 130 F.4th 523, 528–29 (6th Cir. 2025). This inquiry does not turn on the empirics of what a specific jury did in another case, but on what a reasonable jury would have done here. See United States v. Kerley, 838 F.2d 932, 938 (7th Cir. 1988).
III.
Ballinger does not raise the argument that his burglaries were committed on a single occasion. And it easy to understand why, applying the factors laid out by the Supreme Court for a different-occasions inquiry. See United States v. Kimbrough, 138 F.4th 473, 476 (6th Cir. 2025) (citing Wooden v. United States, 595 U.S. 360, 369 (2022)). Wooden made clear that what constitutes an “occasion” is a multi-factored analysis that requires courts to look at timing, proximity of location, and the character and relationship of the offenses. 595 U.S. at 369. The
7
first two factors—timing and location—will often prove dispositive.1 Durham, 2025 WL 2355998, at *2 (citing Wooden, 595 U.S. at 369–70). And “[i]n many cases, a single factor— especially of time or place—can decisively differentiate occasions.” Wooden, 595 U.S. at 369– 70.
Even with the Supreme Court’s statement that courts “have nearly always treated offenses as occurring on separate occasions if a person committed them a day or more apart,” the different- occasions question has proven harder to apply in practice. Wooden, 595 U.S. at 370. This Court has published several recent decisions considering harmless error in the Erlinger context. See Durham, 2025 WL 2355998, at *3 (collecting cases). For those instances where we held the Erlinger error to be harmful, the decisions turned, at least in part, on the uncertainty in the record as to the timing of the predicate offenses. Id. (citing Cogdill, 130 F.4th at 529; Kimbrough, 138 F.4th at 478).
Ballinger points to no uncertainty in the record. Based on his own description of the PSR, Ballinger had prior convictions for one count of burglary in Bradley County in 2010, for which he was sentenced in 2011; a burglary in Bradley County on January 4, 2016; a series of five burglaries in July 2016; and a series of five burglaries in October 2016, with the last one taking place on October 17. We now must answer whether a reasonable jury could have reasonable doubt that the three burglaries furthest from each other (2010, January 4, 2016, and October 17, 2016) took place
1 Ballinger analogized the different occasions question to cases analyzing whether someone had a common scheme or plan. While a common scheme can be relevant to the character and relationship of offenses, this only accounts for a single factor in the analysis, and not one that carries the same weight as time and distance. Campbell, 122 F.4th at 629.
8
on “occasions different from one another.”2 18 U.S.C. § 924(e)(1). We hold here that no reasonable jury could conclude that any two of these burglaries took place on the same occasion.
In Durham, the “time and space” separating two burglaries “clearly establishe[d] the offenses occurred on different occasions.” 2025 WL 2355998, at *4. The burglaries there were separated by a little over three months and more than 15 miles. Id. Unlike Kimbrough, where at most nine days separated the two predicate offenses, the months-long gap in Durham was close to dispositive on its own. Id.; Kimbrough, 138 F.4th at 478.
Ballinger’s case is easier. His first predicate crime took place in 2010, followed by a conviction in Bradley County Criminal Court. His next predicate crime took place a little over five years later in January 2016. These events were followed by additional burglaries of various victims, and then a final burglary in October 2016—more than nine months after his January 2016 burglary. Ballinger cites no court decisions finding a harmful error when the timing of the predicate crimes showed a gap of more than nine months.3 And we have not found any such precedent.
Ballinger instead argues that the series of burglaries from January 2016 to October 2016, could be thought of as a single occasion. But even Ballinger’s recounting of the facts shows why this fails. He describes the 2016 crimes in groups, with one committed in January, five committed in July, and five in October, “with the five burglaries within each group spanning just a few days apart.” Reply Br. 8. This explanation provides the only reasonable way of viewing these crimes— as three (or more) separate occasions. Durham also explains why the January and October
2 Ballinger does not appear to dispute that the 2010 burglary was a separate occasion from the later burglaries, and we agree. 3 Ballinger instead points to past jury verdicts, but these are unpersuasive for the reasons already stated.
9
burglaries are separate occasions despite the other crimes in between. That case offers the example of an opening day baseball game as a separate occasion from “a September series finale[.]” 2025 WL 2355998, at *4 n.5. No reasonable jury would conclude that opening day was the same “occasion” as the September game, just because similar events took place regularly during the season. Id.
Ballinger’s reading would instead turn the occasions clause into a haven for repeat offenders. A burglar’s decades-long spree could be a single occasion if the burglar committed his crimes at regular intervals without getting caught. But a burglar who commits his fewer crimes more sporadically would face a stiffer punishment as a career criminal despite committing fewer violent crimes. It cannot be the case that more criminality becomes a shield against armed career criminal status. This reading undermines the ACCA’s purpose of targeting “those who have repeatedly committed violent crimes.” Wooden, 595 U.S. at 375.
Other circuits have been close to uniform in finding an Erlinger error harmless if months pass between predicate crimes. See United States v. Brown, 136 F.4th 87, 98–99 (4th Cir. 2025); United States v. Robinson, No. 23-14202, 2025 WL 1576122, at *2 (11th Cir. June 4, 2025); United States v. Xavior-Smith, 136 F.4th 1136, 1137 (8th Cir. 2025); United States v. Saunders, No. 23- 6735-cr, 2024 WL 4533359, at *3 (2d Cir. Oct. 21, 2024); United States v. Butler, 122 F.4th 584, 590 (5th Cir. 2024). The only exception to this uniformity comes from Cogdill, 130 F.4th at 529. But the gap in Cogdill was from June to September, only around three months, and this represents the largest gap where a court has found harmful error. Id. The gap between Ballinger’s predicate crimes spans over three times as long as that in Cogdill. Ballinger’s predicate crimes were separated by ten months, which strongly (almost dispositively) suggests harmless error.
10
Though the timing of the offenses would likely be sufficient to find harmless error here, nothing about the character of the offenses suggests a single occasion either. Take for example a case from the Fourth Circuit. In United States v. Browning, the defendant committed predicate crimes ranging “anywhere between four weeks and seven years apart, and had no apparent connection with one another.” No. 23-4038, 2025 WL 1823952, at *2 (4th Cir. July 2, 2025), pet. for cert. filed sub nom. Browning v. United States (No. 25-5344). To determine that the offenses were not connected, the court looked at the names of the victims listed in the PSR. Id. Combining the gap in time with the lack of repeat victims, the Fourth Circuit had no trouble finding harmless error. Id.; see Thomas, 142 F.4th at 418 (noting that different victims weighs toward a finding of different occasions).
The same result holds here. As shown in the PSR, Ballinger burglarized nine different people who have no apparent connection to one another. That fact combined with a large gap in time between offenses, provides a sufficient basis to conclude that the government has satisfied its burden to show that the Erlinger error was harmless.4 III.
Ballinger raises one last point on appeal. He claims (for the first time) that his ACCA enhanced sentence violates the Double Jeopardy clause. Because Ballinger did not make this argument before the district court, we apply plain error review.5 Campbell, 122 F.4th at 634.
4 Because we find that the Erlinger error was harmless, we do not address Ballinger’s arguments about the proper remedy for the error. 5 Ballinger contends that he did make the argument in general terms and so plain error review should not apply. But he did not make the argument with enough specificity “to give the district court an opportunity to correct the alleged error.” United States v. Day, 737 F. App’x 728, 730 (6th Cir. 2018) (quotation omitted). Unlike in United States v. Ehle, Ballinger did not make “arguments that support a double jeopardy claim[.]” 640 F.3d 689, 694 (6th Cir. 2011).
11
Under plain error review, Ballinger must show “(1) error, (2) that was obvious or clear, (3) that affected his substantial rights, and (4) affected the fairness, integrity, or public reputation of his judicial proceedings.” Id. (citing United States v. Johns, 65 F.4th 891, 893 (6th Cir. 2023)). Ballinger cannot show plain error.
Ballinger claims that United States v. Bell requires reversal. There, we decided that once a court accepts a guilty plea, the government cannot proceed to trial on any remaining uncharged counts because this would raise double jeopardy concerns. United States v. Bell, 37 F.4th 1190, 1194, 1998 (6th Cir. 2022). We need not engage in a lengthy analysis under Bell because, as we stated in Thomas, Bell concerned a “unique factual circumstance” that does not apply in the context of harmless error review for an Erlinger error. 142 F.4th at 422. The double jeopardy challenge in Thomas, almost identical to the one presented here, did not undermine the conclusion that the Erlinger error was harmless. Id. at 423.
While Thomas did not decide the double jeopardy issue for cases that present a harmful error, it went far enough to show that any error was not plain and obvious. Without a plain and obvious error, Ballinger’s argument fails.
IV.
Because the Erlinger error in Ballinger’s case was harmless and he did not show that his double jeopardy argument constituted plain error, we AFFIRM his enhanced sentence under the ACCA.
Ballinger’s arguments related only to the different occasions analysis and “the Court’s authority to apply ACCA absent a jury trial.” Sentencing Tr., R-59, PageID 609.