Before JONES, SMITH, and ELROD, Circuit Judges. JERRY E. SMITH, Circuit Judge: Buck Brune is a methamphetamine (“meth”) dealer. In charging him, the government accidentally cited the wrong statutory subparagraph. After Brune had pleaded guilty, the court copied that error into its order accepting his plea but later corrected it. The court applied a sentencing enhancement on the ground that some of Brune’s meth was imported. Brune contends that the court’s correction of the erroneous citation amounted to double jeopardy and that the enhancement was erroneous. We find no error and affirm. I. Brune distributed at least 50–75 pounds of meth over nine months. For five months, he sold half a pound of meth to one coconspirator each day. His supplier was “a member of the Michoacán Cartel based in Dallas, Texas.”[1] Brune concedes that that cartel “borrow[s] its name from a state in Mexico.” The government filed a one-count information based on the conspiracy provision of 21 U.S.C. § 846, charging Brune with conspiracy to violate “21 U.S.C[.] §§ 841(a)(l) and (b)(l)(C), namely to possess with intent to distribute a mixture and substance containing more than 50 grams of methamphetamine.” But the information cited the wrong part of § 841(b)(1): Subparagraph B—not C—criminalizes possession of a substance containing more than 50 grams of meth. In contrast, subparagraph C provides “the baseline statutory penalty for any quantity of methamphetamine.” United States v. Doggett, 230 F.3d 160, 166 (5th Cir. 2000) (emphasis added). Subparagraph B’s penalty range is 5 to 40 years, § 841(b)(1)(B); subparagraph C’s is 20 years or less, § 841(b)(1)(C). The parties agree that subparagraph C is a lesser-included offense of subparagraph B. Despite that initial error and without any plea agreement, Brune pleaded guilty to subparagraph B,[2] referencing it nine times. For instance, Brune’s factual resume cited subparagraph B, twice indicated that Brune was subject to its penalty range, and twice parroted its 50-gram threshold. Brune’s waiver of indictment also cited subparagraph B. At arraignment, Brune admitted he understood its elements and penalty range. His lawyer admitted that “the intention of the parties was for Mr. Brune to enter a guilty plea to that offense, which was in the factual resume, and that would be a five to 40 count”—namely subparagraph B. In recommending that the district court accept Brune’s guilty plea, however, the magistrate judge copied the information’s erroneous citation. The district court adopted that recommendation, accepted the plea, and adjudged Brune guilty. Thus, the presentence investigation report came back with subparagraph C’s “maximum term of imprisonment,” namely “20 years,” even though it should have been 40 years under subparagraph B. The government raised two objections. First, the government noted that Brune pleaded guilty to subparagraph B—not C. Brune countered, contending, inter alia, that modification of the court’s order accepting his plea would violate the prohibition against double jeopardy. The district court rejected Brune’s contentions and amended its order to reflect that it was accepting Brune’s guilty plea to subparagraph B. Second, the government requested a sentencing enhancement for an offense involving “importation of . . . methamphetamine,” which would raise Brune’s offense level by two.[3] Brune countered that there was insufficient evidence for that enhancement, because Brune’s supplier was “based in Dallas.” The court found there was sufficient evidence that Brune conspired to possess meth that “originated in . . . Mexico.” II. The government contends that jeopardy never attached.[4] Our review is de novo. United States v. Dugue, 690 F.3d 636, 637–38 (5th Cir. 2012) (per curiam). We (A) determine that jeopardy does not always attach upon acceptance of a guilty plea, (B) explain the framework for analyzing attachment under Ohio v. Johnson, 467 U.S. 493 (1984), and (C) apply that framework. There was no double-jeopardy violation. A. “No person shall be . . . subject for the same offence to be twice put in jeopardy of life or limb.” U.S. CONST. amend. V. To violate that clause, the initial prosecution must have “put [the defendant] in jeopardy.” Id. That inquiry becomes important where the initial prosecution gets derailed. If a trial gets derailed, it still puts the defendant in jeopardy if jeopardy (1) attached and (2) terminated.[5] Attachment refers to the “point in criminal proceedings at which [double-jeopardy] purposes and policies are implicated.” Serfass v. United States, 420 U.S. 377, 388 (1975). For instance, in a jury trial, attachment occurs “when the jury is empaneled and sworn.” Crist v. Bretz, 437 U.S. 28, 38 (1978). Termination means that double jeopardy does not bar a second prosecution where “criminal proceedings against an accused have not run their full course.” Justs. of Bos. Mun. Ct. v. Lydon, 466 U.S. 294, 308 (1984) (quotation marks and citation omitted). For instance, a mistrial for a deadlocked jury does not terminate jeopardy, see Richardson v. United States, 468 U.S. 317, 323–24 (1984), but an acquittal does, see Lydon, 466 U.S. at 308. Where a guilty plea gets derailed, the Supreme Court has neither identified a precise moment of attachment[6] nor applied the concept of termination.[7] That reticence left lower courts to fill in the gaps. Because acceptance of a guilty plea is arguably analogous to a jury verdict, courts initially intuited that jeopardy attaches upon acceptance of a guilty plea.[8] For instance, in our first foray into the issue, in United States v. Sanchez, 609 F.2d 761, 762 (5th Cir. 1980), we agreed that “[j]eopardy attaches with the acceptance of a guilty plea.” Relying solely on a now-abrogated, out-of-circuit case,[9] we provided no reasoning for that conclusion. It is no surprise, then, that four years later, the Supreme Court “effectively reject[ed] the double jeopardy concerns expressed . . . in Sanchez.”[10] In Johnson, 467 U.S. at 494, the government charged the defendant with two sets of greater and lesser-included offenses. Johnson pleaded guilty—over the government’s objection—of the two lesser-included offenses, then moved to dismiss the greater offenses on double-jeopardy grounds. Id. In rejecting that claim, the Court applied two concepts that relate to attachment and termination. First, although attachment occurs where double-jeopardy “purposes and policies are implicated,” Serfass, 420 U.S. at 388, Johnson, 467 U.S. at 501, concluded that no double-jeopardy interest “is implicated” in the “acceptance of a guilty plea to lesser included offenses while charges on the greater offenses remain pending.” Thus, although Sanchez had suggested the opposite,[11] double jeopardy did not bar prosecution of a greater offense after a plea of a lesser-included offense. Second, the Court applied a rationale reminiscent of termination’s requirement that proceedings “run their full course” before a defendant can successfully invoke double jeopardy. Lydon, 466 U.S. at 308 (quotation marks and citation omitted). Specifically, in Johnson, 467 U.S. at 502, the Court sought to ensure that the government has “one full and fair opportunity to convict those who have violated its laws.” Thus, instead of expressly determining whether jeopardy attached and terminated, the Court analyzed (1) “finality” and (2) “prevention of prosecutorial overreaching,” concluding that “[n]o interest . . . protected by the Double Jeopardy Clause [was] implicated” in that situation. Id. at 501. Although that framework differs from the attachment and termination bookends that the Court employs when examining a trial, Johnson recognized differences between guilty pleas and trials.[12] Specifically, guilty pleas provide prosecutors no “opportunity to marshal [their] evidence and resources.” Id. A defendant can plead guilty even over the government’s objection. Id. at 494. That distinction undermines the assumption of lower courts—which infected Sanchez—that guilty pleas are relevantly analogous to jury verdicts and that jeopardy, therefore, attaches upon acceptance of a guilty plea. Nonetheless, our opinions have continued to recite Sanchez‘s rule that jeopardy attaches upon acceptance of a guilty plea.[13] We must decide whether Johnson abrogated that statement. Brune contends that, under Sanchez, jeopardy attaches upon acceptance of a guilty plea. The government contends that Johnson rejected Sanchez‘s double-jeopardy concerns, so jeopardy does not always attach upon acceptance of a guilty plea. We agree because (1) Johnson abrogated Sanchez‘s statement about attachment, (2) Brune’s counterarguments are not persuasive, and (3) the rule of orderliness does not preclude that conclusion. 1. Johnson abrogated Sanchez‘s statement regarding attachment. The First[14] and Third[15] Circuits agree that, under Johnson, jeopardy does not always attach upon acceptance of a guilty plea. The Second Circuit implies that jeopardy attaches upon acceptance of a guilty plea[16] and treats Johnson as an exception to that rule, which applies only where (1) the prosecutor objects to a plea of a lesser-included offense and (2) and the charge on the greater offense remains pending.[17] The Sixth, Eighth, Ninth, Tenth, and Eleventh Circuits either largely ignore Johnson[18] or skirt the issue.[19] We conclude that jeopardy does not always attach upon acceptance of a guilty plea. Two lines of reasoning support that conclusion. a. Sanchez‘s rule about attachment is inconsistent with Johnson. Under Sanchez, jeopardy attaches upon acceptance of a guilty plea. Sanchez, 609 F.2d at 762. Moreover, lesser-included and greater offenses constitute the “same offense” for double-jeopardy purposes. Brown v. Ohio, 432 U.S. 161, 166 n.6 (1977). Thus, if Sanchez were correct that jeopardy attaches upon acceptance of a guilty plea of a lesser offense, then—under Supreme Court precedent[20]—a successive prosecution for a greater offense would implicate double-jeopardy interests.[21] Consequently, if jeopardy had attached upon acceptance of Johnson’s guilty plea to the lesser-included offense, then prosecution of him for the greater offense would have had double-jeopardy consequences.[22] But the government’s prosecution of Johnson for the greater offense did not invoke double-jeopardy interests.[23] It necessarily follows that jeopardy did not attach upon the court’s acceptance of Johnson’s guilty plea of the lesser-included offense. Thus, under Johnson and contrary to Sanchez, jeopardy does not always attach upon acceptance of a guilty plea. The Third Circuit came to a similar conclusion. Although the court had previously held that “jeopardy . . . attached with the acceptance of [a] guilty plea,” Jerry, 487 F.2d at 606, the court later concluded that Jerry‘s statement about attachment was “only an assumption,” which “is inconsistent with . . . Johnson,” Gilmore, 793 F.2d at 571. Importantly, Jerry was the sole basis for Sanchez‘s statement that jeopardy attaches upon acceptance of a guilty plea. See Sanchez, 609 F.2d at 762 (citing Jerry, 487 F.2d at 606). In sum, Sanchez‘s rule about attachment is inconsistent with Johnson. Moreover, Johnson abrogated Sanchez‘s sole buttress—Jerry. It is no surprise, then, that Brune concedes that Sanchez and Johnson are in conflict.[24] b. Johnson‘s holding has the hallmarks of attachment, not some other facet of double jeopardy. For instance, attachment occurs at the “point in criminal proceedings at which [double-jeopardy] purposes and policies are implicated.”[25] Consistently with that concept, Johnson, 467 U.S. at 501, held that “[n]o interest . . . protected by the Double Jeopardy Clause is implicated” in that situation. Johnson‘s focus on the threshold inquiry—whether double jeopardy is even implicated in the first place—thus sounds in attachment. Conversely, Johnson does not resemble an exception to the double-jeopardy prohibition.[26] For instance, where the Supreme Court applies such an exception, it first “assume[s] that jeopardy attached” and then asks whether an exceptional circumstance “removed the double jeopardy bar.” Ricketts, 483 U.S. at 8. Johnson neither assumed that jeopardy attached nor used language about removing the double-jeopardy bar. Finally, although Johnson‘s holding appears to implement the policy behind termination, see Part II.A, supra, it does not appear to engraft termination’s legal test onto plea proceedings. For instance, to establish that jeopardy did not terminate after a trial, a prosecutor must show a “manifest necessity” to “retry the defendant.” Richardson, 468 U.S. at 323–24 (quotation marks and citation omitted). Johnson did not, however, employ anything akin to the manifest-necessity standard.[27] In fact, the First and Third Circuits, which had attempted to graft that standard onto pleas, concluded that Johnson was inconsistent with those precedents. See Soto, 825 F.2d at 619; Gilmore, 793 F.2d at 571. In short, Johnson‘s holding has more to do with attachment than with termination or an exception to the double-jeopardy bar. 2. Brune raises two counterarguments.[28] Neither is persuasive. a. The Supreme Court said—about a century ago, in a case that didn’t mention double jeopardy—that “[a] plea of guilty . . . is itself a conviction.” Kercheval v. United States, 274 U.S. 220, 223 (1927). And double jeopardy prohibits “a second prosecution for the same offense after conviction.” Johnson, 467 U.S. at 498 (quotation marks and citation omitted). Thus, if we myopically weld those precedents together, the argument goes, attachment must occur upon the plea of guilty, because that is when the conviction occurs. See Morris, 264 F.3d at 49. The Johnson dissenters made that sort of argument and lost.[29] It is no surprise, then, that, besides its cameo in Morris, that argument largely inhabits pre-Johnson concurrences[30] and student notes.[31] Brune raises it here, and it fails for three reasons. First, that theory is inconsistent with Johnson. If Johnson’s guilty plea of a lesser-included offense constituted his conviction for double-jeopardy purposes, then the government’s subsequent prosecution would have implicated double-jeopardy interests. But the subsequent prosecution did not do that. Johnson, 467 U.S. at 501. And, to the extent that Kercheval conflicts with Johnson, we must follow Johnson as the Court’s later pronouncement. Second, Brune cherry-picks Kercheval‘s language. In Kercheval, 274 U.S. at 225, the Court had nothing to say about double jeopardy; it concluded that evidence of a withdrawn guilty plea is inadmissible at trial. In any event, a more vigorous examination of Kercheval undermines Brune’s argument. Although Brune notes Kercheval‘s statement that a guilty plea is “conclusive” of guilt, he omits its observation that it would be wrong to “hold [a withdrawn] plea conclusive.” Id. at 224. Kercheval even noted that, in some circumstances, courts may “vacate a plea of guilty,” “substitute a plea of not guilty,” and “have a trial.” Id. Brune does not engage with any of that language. We thus decline to graft Kercheval‘s inapposite, cherry-picked statement onto the double-jeopardy context in a manner that undermines Johnson. Third, Brune’s reasoning places more weight on Kercheval‘s use of “conviction” than it can bear. See id. at 223. Because a judgment of conviction can occur only after sentencing, FED. R. CRIM. P. 32(k)(1), a “defendant ha[s] not been formally convicted” until “entry of judgment and sentencing on the accepted guilty plea.”[32] b. Some scholars suggest that jeopardy attaches where “the risks of injury” are sufficiently great. LAFAVE ET AL., supra, § 25.1(d) (quotation marks and citation omitted). Moreover, a guilty plea exposes a defendant to risks, because it constitutes an admission of material facts along with a waiver of myriad rights. Combs, 634 F.2d at 1300 (McKay, J., concurring in part and dissenting in part). Thus, Brune contends that, “[w]ith the risk of a determination of guilt, jeopardy attaches.” That argument fails for two reasons. First, Brune confuses necessary and sufficient conditions. Citing Ser-fass, 420 U.S. at 391–92, he contends that a risk of a determination of guilt is sufficient to show attachment.[33] But he shrewdly omits any quotation, because Serfass really says that, “[w]ithout risk of a determination of guilt, jeopardy does not attach.” Id. Thus, Serfass said that risk is necessary—not sufficient—to show attachment. That makes sense, because, before trial, a defendant must conduct motions and jury selection, which “may decide the defendant’s case.” Crist, 437 U.S. at 49 (Powell, J., dissenting). Yet, even with exposure to those risks, jeopardy does not attach until later. Id. Thus, pointing to some risk of a determination of guilt is not enough to show that jeopardy attaches. Second, Brune’s single-minded focus on risks ignores the government’s interest in completing its prosecution. Yet, Johnson, 467 U.S. at 502, considered it relevant that “ending prosecution now would deny the State its right to one full and fair opportunity to convict those who have violated its laws.” A singular focus on risks would allow a defendant to place himself in jeopardy—by pleading guilty—and thus “use the Double Jeopardy Clause as a sword to prevent the State from completing its prosecution.” Id. Johnson, therefore, bars us from focusing on risk to the exclusion of the government’s interest in completing its prosecution. 3. Given those considerations, jeopardy does not always attach upon acceptance of a guilty plea.[34] We thus join the majority of circuits that have analyzed the impact of Johnson and, in doing so, we affirm our only previous analysis of that issue, namely that Johnson “effectively reject[ed] the double jeopardy concerns expressed . . . in Sanchez.” Foy, 28 F.3d at 471 n.13. The rule of orderliness does not preclude that conclusion, because (a) Sanchez made pronouncements about attachment only in now-abrogated dicta, and (b) none of our later opinions reciting Sanchez‘s rule about attachment grappled with whether Johnson rejected Sanchez‘s rule. a. Sanchez made conclusions about attachment only in dicta. For instance, although Sanchez, 609 F.2d at 762, stated that “[j]eopardy attaches with the acceptance of a guilty plea,” the trial court accepted Sanchez’s guilty plea only conditionally, so there was no double-jeopardy violation, id. at 763. Because any successive prosecution occurred before acceptance of the guilty plea, we needed only to conclude—as then-Judge Gorsuch did in his analysis of that issue—that jeopardy does not attach at least until acceptance of a guilty plea.[35] Because the broader assumption about attachment was unnecessary to our decision, that “broad and unnecessary language of [Sanchez] [can]not be considered binding authority.”[36] Moreover, because Johnson “implicitly overrule[d]” Sanchez‘s statement about attachment, “we have the . . . obligation to declare and implement this change in the law.” Hines v. Quillivan, 982 F.3d 266, 271 (5th Cir. 2020) (quotation marks and citation omitted). In short, Sanchez‘s statements about attachment are abrogated dicta. b. We have repeated Sanchez‘s rule about attachment in three, post-Johnson, published cases.[37] None of those opinions, however, grappled with whether Johnson rejected Sanchez‘s rule.[38] And, under the rule of orderliness, “[a]n opinion restating a prior panel’s ruling does not sub silentio hold that the prior ruling survived an uncited Supreme Court decision.” Gahagan v. United States Citizenship & Immigr. Servs., 911 F.3d 298, 302 (5th Cir. 2018). Thus, “the rule of orderliness has little persuasive force when the prior panel decision at issue conflicts with a Supreme Court case to which the subsequent panel decision is faithful.” Kennedy v. Tangipahoa Par. Libr. Bd. of Control, 224 F.3d 359, 370 n.13 (5th Cir. 2000). Because our prior opinions did not analyze whether Johnson abrogated Sanchez‘s rule about attachment, we are not bound by their rote recitations of Sanchez‘s rule.[39] B. Because jeopardy does not always attach upon acceptance of a guilty plea, we next determine the test for ascertaining when it attaches. Johnson had to address that same question. Specifically, the Court had held that the Double Jeopardy Clause barred prosecution of a greater offense after a court accepted a guilty plea and sentenced for a lesser-included offense. Brown, 432 U.S. at 169, 162. Because Johnson also involved a successive prosecution of a greater offense after a plea of a lesser-included offense, the Court needed to distinguish Brown—which implicated double jeopardy—from Johnson—which did not. Johnson, 467 U.S. at 496. To do so, the Court determined that “the principles of finality and prevention of prosecutorial overreaching applied in Brown” were absent in Johnson. Id. at 501. Johnson thus “provided a framework . . . for determining whether jeopardy attaches when a defendant pleads guilty.”[40] Courts must examine “the twin aims of the Double Jeopardy Clause: protecting a defendant’s finality interests and preventing prosecutorial overreaching.” Patterson, 406 F.3d at 1097 (Kozinski, J., dissenting from denial of rehearing en banc). The Second Circuit disagrees, concluding that the lodestar of Johnson‘s analysis was the fact that the greater charge remained pending at the time Johnson pleaded guilty of the lesser-included offense.[41] Johnson, however, used “pending” once. Johnson, 467 U.S. at 501. Moreover, it did so only within its analysis of finality and prosecutorial overreach. See id. To be sure, whether a charge of a greater offense was pending may prove relevant, but that was not the Court’s guiding principle in distinguishing Brown. C. To determine whether jeopardy attached, we consider (1) Brune’s finality interest and (2) prevention of prosecutorial overreach. 1. Johnson employed three considerations in ascertaining a defendant’s finality interest. None is present here. First, the Court asked whether the situation before it involved any of the “implications of an implied acquittal which results from a verdict . . . rendered by a jury.” Id. at 501–02 (cleaned up). “The mere acceptance of a guilty plea,” however, “does not carry the same expectation of finality and tranquility that comes with a jury’s verdict.” Soto, 825 F.2d at 620. Moreover, given a court’s ability to correct errors, an erroneous citation in an order accepting a plea does not imply an acquittal. FED. R. CRIM. P. 36. Second, Johnson asked whether “the State had the opportunity to marshal its evidence and resources more than once or to hone its presentation of its case through a trial.” Johnson, 467 U.S. at 501. The government had no opportunity to marshal evidence against Brune, however, because his plea proceedings “[did] not involve the ordeal of a trial.” Soto, 825 F.2d at 618. Third, Johnson, 467 U.S. at 501, noted that the charges of the greater offenses were pending, which was relevant presumably because “Johnson could have foreseen a prosecution on the pending charges,” Soto, 825 F.2d at 619. Evidence abounds that Brune foresaw a subparagraph B prosecution. His factual resume referenced subparagraph B five times, reciting its citation, penalty range, and 50-gram threshold. His waiver of indictment cited that subparagraph. At his arraignment, the government referenced that subparagraph’s elements and penalty range. Finally, Brune’s counsel conceded that “the intention of the parties was for Mr. Brune to enter a guilty plea” to subparagraph B. Given that evidence, Brune’s finality interest is even more miniscule than that of the defendant in Johnson: Johnson was merely on notice of his greater offense; Brune intended to plead guilty of his. In short, Brune intended to plead guilty to subparagraph B. Because the government botched its citations, he now seeks “an undeserved windfall by shaving” years off his sentence. Patterson, 406 F.3d at 1095 (Kozinski, J., dissenting from denial of rehearing en banc). In other words, Brune seeks to “use the Double Jeopardy Clause as a sword to prevent the State from completing its prosecution” of subparagraph B—a contingency that Johnson sought to avoid.[42] Brune’s finality interest is nil. 2. Johnson employed two considerations in identifying prosecutorial overreach. First, in Johnson, 467 U.S. at 501–02, there was no overreach where the charge for the greater offense remained “pending.” In contrast, some authorities suggest there might be overreach where the government charges a defendant “with a second crime after getting him to plead guilty” with a plea agreement.[43] The government did not bring new charges against Brune. Nor did it dupe him with a plea agreement—there never was one to begin with. Second, Johnson, 467 U.S. at 502, considered whether “ending prosecution now would deny the [government] its right to one full and fair opportunity to convict those who have violated its laws.” Because Brune’s case has not gone to trial, and the government has not dismissed the charge for subparagraph B, the government has not yet had one full opportunity to convict him of subparagraph B. And there’s nothing unfair to Brune about that result: The government seeks to prosecute him for the only charge to which Brune himself pleaded guilty. In sum, the Double Jeopardy Clause “was not written or originally understood to pose an insuperable obstacle to the administration of justice in cases”—like Brune’s—”where there is no semblance of . . . oppressive practices.” Currier v. Virginia, 138 S. Ct. 2144, 2149 (2018) (quotation marks and citation omitted). Correction of a typo isn’t oppressive. Because Brune’s finality interest is low, and there is no evidence of prosecutorial overreach, jeopardy did not attach upon the court’s acceptance of Brune’s guilty plea. There is no double-jeopardy violation. III. Brune avers that the district court clearly erred in applying an “importation” sentencing enhancement. We disagree. Under U.S.S.G. § 2D1.1(b)(5), a court can increase a defendant’s offense level by two if his offense “involved the importation of . . . methamphetamine.” “We review the district court’s factual determination that an offense involved the importation of methamphetamine for clear error.”[44] We will find no clear error so long as the court’s conclusion is “plausible in light of the record read as a whole.” United States v. Dinh, 920 F.3d 307, 310 (5th Cir. 2019) (quotation marks and citation omitted). Brune sold at least 50–75 pounds of meth over nine months. His supplier was a member of the Michoacán Cartel based in Dallas, and that cartel borrows its name from a Mexican state. Those facts support the inference that some of Brune’s drugs were imported.[45] The importation finding is “plausible in light of the record read as a whole,” so there is no clear error.[46] AFFIRMED.
