concurring.
Recently, in United States v. Antonakopoulos, 399 F.3d 68 (1st Cir.2005), a panel of this court explained for the first time our standards for review of unpreserved claims of sentencing error in the aftermath of United States v. Booker, 543 U.S. -, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005). That decision is binding on subsequent panels. Eulitt v. Me. Dep’t of Educ., 386 F.3d 344, 349 (1st Cir.2004) (in a multi-panel circuit, newly constituted panels are bound by prior panel decisions). I agree with the result of the application of Antonakopoulos to this case. I write separately, however, to explain why, if I were free to do so, I would take a different approach to reviewing unpreserved claims of Booker error.
Before explaining my differing views, however, I want to stress that I agree with much in Antonakopoulos. I agree with its description of Booker error as inhering in the mandatory nature of the sentencing guidelines, regardless of whether the sentence was premised on any judge-found *57facts. I agree, too, that the defendant has preserved a claim of Booker error if he argued below that his sentence violated Apprendi v. New Jersey, 530 U.S. 466, 120 S.Ct. 2348, 147 L.Ed.2d 435 (2000), or Blakely v. Washington, 542 U.S. —, 124 S.Ct. 2531, 159 L.Ed.2d 403 (2004), or that the guidelines were unconstitutional. I agree that we should not remand eases automatically, either because a defendant’s sentence was enhanced on the basis of judge-found facts or because the sentence was imposed on the basis of mandatory guidelines.
My one disagreement with Antonako-poulos is crucial, however. I do not believe that we should require defendants invoking unpreserved Booker error to make a specific showing of prejudice (the reasonable probability of a different outcome) to satisfy the third step of plain-error review. Rather, such error should entitle the defendant to a presumption of prejudice, which the government can then try to rebut. This approach, adopted by a panel of the Sixth Circuit in United States v. Barnett, 398 F.3d 516 (6th Cir.2005), is well grounded in Supreme Court precedent and has,been applied by our sister circuits in other contexts “where the inherr ent nature of the error made it exceptionally difficult for the defendant to demonstrate that the outcome of the lower court proceeding would have been different had the error not occurred.” Id. at 526-27. I wish to explain further my reasons for preferring this approach.
1. The difficulty of reconstructing a hypothetical sentencing decision
Several courts of appeals considering unpreserved claims of Booker error have emphasized the difficulty of speculating about what a district-court judge would have done differently under an advisory guidelines regime. The Fourth Circuit admitted that: “We simply do not know how the district court would have sentenced Hughes had it been operating under the regime established by Booker.” United States v. Hughes, 396 F.3d 374, 381 n. 8 (4th Cir.2005). Similarly, the Sixth Circuit wrote that “even if we conclude that the evidence is ‘overwhelming and essentially uncontroverted’ we cannot know the length of imprisonment that the district court judge would have imposed pursuant to this evidence following Booker.” United States v. Oliver, 397 F.3d 369, 380 n. 3 (6th Cir.2005) (quoting United States v. Cotton, 535 U.S. 625, 633, 122 S.Ct. 1781, 152 L.Ed.2d 860 (2002)). The Eleventh Circuit, too, after wondering what the district court might have done differently, answered its own question:
The obvious answer is that we don’t know. If the district court judge in this case had the liberty of increasing or decreasing Rodriguez’s sentence above or below the guidelines range, he might have given Rodriguez a longer sentence, or he might have given a shorter sentence, or he might have given the same sentence. The record provides no reason to believe any result is more likely than the other. We just don’t know.
United States v. Rodriguez, 398 F.3d 1291, 1301 (11th Cir.2005).1
Indeed, the Second Circuit found this problem so vexing that it chose to ask the district courts directly whether,' in their judgment, “a nontrivially different sentence would haye been imposed” under advisory guidelines. United States v. Crosby, 397 F.3d 103, 118 (2d Cir.2005). The Second Circuit noted that
in many cases, it will be impossible to tell whether the judge would .have imposed the same sentence had the judge not . felt compelled to impose a Guidelines sentence. It will also be impossible to tell what considerations counsel *58for both sides might have brought to the sentencing judge’s attention had they known that they could urge the judge to impose a non-Guidelines sentence.
Id. at 115. The court chose to solve the problem by remanding any case with un-preserved Booker error to the district courts, which would then decide if resen-tencing was required. Although I do not favor the Second Circuit’s remand approach, I understand the concerns that motivated it. So, too, did a panel of the Sixth Circuit in the Barnett decision.
2. Barnett and presumed prejudice in Olano
In Barnett, the court emphasized the difficulty faced by defendants trying to demonstrate the prejudice that resulted from the district court’s application of mandatory guidelines, when “well established case law substantially undermined any need or incentive for sentencing courts pr e-Booker to note their objections and reservations in sentencing defendants under the then-mandatory Guidelines.” 398 F.3d at 529. Consequently,
[i]t would be improper for this Court now to require defendants such as Barnett to produce this type of evidence— that sentencing courts had no reason to provide under our pr e-Booker case law — in order to establish that their substantial rights have been affected.... Instead of speculating as to the district court’s intentions in the pr e-Booker world, and trying to apply those intentions to predict the same court’s sentence under the post-Booker scheme, we are convinced that the most prudent course of action in this case is to presume prejudice....
Id.
Barnett reads United States v. Olano, 507 U.S. 725, 113 S.Ct. 1770, 123 L.Ed.2d 508 (1993), as suggesting that if a defendant would have extraordinary difficulty making a specific showing of prejudice, then a presumption of prejudice may be appropriate. Barnett, 398 F.3d at 526. The government, of course, can try to rebut that presumption. Id. at 529. The dissent in Barnett maintains, however, that “the Supreme Court has never put its imprimatur on the idea that we may presume prejudice in plain error review,” id. at 536. It supported that assertion by observing that Olano devotes just a sentence to the issue and then refused to consider it further.
That is not a fair reading of Olano. After first positing a category of plain errors entitled to a presumption of prejudice, the Court found no reason to place the specific plain error cited by the defendant (the presence of alternate jurors during jury deliberations) in that category. 507 U.S. at 735, 113 S.Ct. 1770. True, the Court declined to offer a strict definition of the category, seeing no need to “address those errors that should be presumed prejudicial if the defendant cannot make a specific showing of prejudice.” Id. Later, however, the Court did confront the question of whether Olano’s error would belong to such a category, if it exists; it decided that “we see no reason to presume prejudice here.” Id. at 737, 113 S.Ct. 1770. At the same time, the Court allowed that “[tjhere may be cases where an intrusion should be presumed prejudicial.” Id. at 739, 113 S.Ct. 1770. In fact, the Court framed the basic inquiry of its decision as: “The question, then, is whether the instant violation of Rule 24(c) prejudiced respondents, either specifically or presumptively.” Id. (emphasis added). Ultimately, the Court did not think “that the mere presence of alternate jurors entailed a sufficient risk of ‘chill’ to justify a presumption of prejudice on that score.” Id. at 741, 113 S.Ct. 1770. Nevertheless, its *59analysis forthrightly explored the possibility that some kinds of errors .would justify stich a presumption.
Here, in appeals with unpreserved claims of Booker error, we have been presented with such errors.2
3. Presumed prejudice in other contexts
Courts have presumed prejudice for errors that, by their very nature, make a demonstration of prejudice exceptionally difficult. The Sixth Circuit has presumed prejudice where alternate jurors actually participated in the jury’s deliberations (unlike Olano, where they were merely present). See Manning v. Huffman, 269 F.3d 720, 726 (6th Cir.2001) (“[T]he Olano court made it quite clear that in some situations a presumption of prejudice is appropriate.”). After all, there are “strict eviden-tiary prohibitions against inquiring into the mental processes of the jury [that] would make it almost impossible for a defendant to show that an alternate juror in fact prejudiced his case.” Id. at 725 n. 2. In Booker cases, defendants face similarly forbidding speculation about the mental processes of a district-court judge if given the opportunity to apply advisory senténc-ing guidélines.
Courts have presumed prejudice where the defendant has been denied his right to allocution, that is, his opportunity to present mitigating circumstances to the court before being sentenced. In those cases, too, courts have presumed that the defendant was prejudiced because of the extraordinary difficulty in discerning the error’s prejudicial effect. See United States v. Reyna, 358 F.3d 344, 352-53 (5th Cir.2004) (en banc), cert. denied, — U.S. —, 124 S.Ct. 2390, 158 L.Ed.2d 966 (2004). The Fifth Circuit noted that this approach “avoids speculation as to what the defendant might have said or argued to mitigate his sentence.” Id. at 352. In United States v. Adams, 252 F.3d 276 (3d Cir.2001), the Third Circuit stated that showing prejudice
would be an onerous burden for Adams [i.e., the defendant] to meet. In order to prove that the error actually “affected the outcome of the district court proceedings,” Adams would have to point to statements that he would have made at sentencing, and somehow show that these statements would have changed the sentence imposed by the District Court.... But as the Supreme Court explained in Olano, there may be some errors “that should be presumed prejudicial if the defendant cannot make a specific' showing of prejudice.” Olano, 507 U.S. at 735, 113 S.Ct. 1770. Thus the question for us becomes: should we presume prejudice when a district court violates a defendant’s right of allocution?
252 F.3d at 287. The Third Circuit’s answer was yes.3 In United States v. De Alba Pagan, 33 F.3d 125 (1st Cir.1994), *60although we did not explicitly use Olano’s four-part test, we observed that relieving a defendant of his burden to show prejudice can compensate for cases where it would be extraordinarily difficult for the defendant to do so. Consequently, we wrote that the denial of right to allocution “ordinarily requires vacation of the sentence imposed without a concomitant inquiry .into prejudice. This is so precisely because the impact of the omission on a discretionary decision is usually enormously difficult to ascertain.” Id. at 130.
4. Other reasons for presuming prejudice
In addition to the difficulty of reconstructing a hypothetical sentencing-court decision, there are three other reasons that make the presumption of prejudice a sensible choice for addressing unpreserved Booker error.
a. Difficulty of anticipating the error
Although some defendants preserved their claims of Booker error in the way described in Antonakopoulos,4 others did not. Perhaps they can be faulted for not doing what other defendants did. Yet the fact remains that the status of the Guidelines was uncertain until Booker was decided, and the adoption of the remedy chosen in Booker for the Sixth Amendment violation (converting the Guidelines from mandatory to advisory) surprised many in the legal profession. Placing the burden on defendants to establish prejudice for un-preserved errors ordinarily makes sense because they ignored existing law that they could have invoked to avoid the error. As the Supreme Court has explained, the “burden should not be too easy” because the prejudice standard helps “to encourage timely objections and reduce wasteful reversals by demanding strenuous exertion to get relief for unpreserved error.” United States v. Dominguez Benitez, 542 U.S. 74, 124 S.Ct. 2333, 2340, 159 L.Ed.2d 157 (2004).
Here, however, the “existing” law (the availability of advisory guidelines) is an artifice of our rule that “a new rule for the conduct of criminal prosecutions is to be applied retroactively to all cases ... pending on direct review ..., with no exception for cases in which the new rule constitutes a ‘clear break’ with the past.” Griffith v. Kentucky, 479 U.S. 314, 328, 107 S.Ct. 708, 93 L.Ed.2d 649 (1987); see also Johnson v. United States, 520 U.S. 461, 467, 117 S.Ct. 1544, 137 L.Ed.2d 718 (1997). In actuality, a demand by defendants pre-Booker that district-court judges treat the guidelines as advisory would have been rejected by most of them as an incorrect statement of the law. Frankly, “it seems unfair to fault [the defendant] for failing to raise at [sentencing] an objection based upon a rule that was not announced until after the [sentencing] was concluded.” United States v. Barone, 114 F.3d 1284, 1294 (1st Cir.1997) (citing United States v. Collins, 60 F.3d 4, 7 (1st Cir.1995)).
Perhaps defendants should be grateful that our rule of retroactivity for cases pending on direct review allows any possibility at all for resentencing. But the fact remains that a greater willingness to acknowledge the likelihood of prejudice from a dramatic change in the law like Booker does not reward “sandbaggers” who hoard their objections to hedge against a result not to their liking. There was no game-playing with a rule of law not yet known.5
*61b. Administrative burden
The administrative burdens of increased remands for sentencing (a likely result of the presumption of prejudice approach), while not insubstantial, would certainly be manageable, given the limited universe of cases at issue and the relatively low cost of correcting errors in sentencing. For example, the Second Circuit estimated that, when Booker was decided, it had about 200 cases pending on direct review with sentences that might be erroneous under the Supreme Court’s new teaching:
Many of these will likely be remanded.... Some of the remands will likely result in resentencing. We do not regard that prospect as an undue burden on the proper functioning of the criminal justice system in the federal courts of this Circuit. On the contrary, we consider it far preferable to leaving some materially erroneous sentences in place simply because we cannot guess what sentencing judges would have done.
United States v. Williams, 399 F.3d 450, 461, 2005 WL 425212, at *8 (2d Cir.2005). These cases are a closed set. As soon as the Supreme Court issued its opinion in Booker, district courts knew to stop sentencing defendants under mandatory guidelines, thus ensuring that no more cases will be' tainted with Booker error.
Besides the simple numbers involved, we must considér that resentencing in cases still pending on direct review does not undermine the judicial system’s high stakes in finality. By definition, the cases we address here are not final. Moreover, resentencing does not pose the burdén of a new trial, with its considerable costs in time, money, and other resources. As the Second Circuit observed in Williams, “the cost of correcting a sentencing error is far less than the cost of a retrial. A resen-tencing is'a brief event, normally taking less than a day and requiring the attendance of only the defendant, counsel, and court personnel.” Williams, 399 F.3d at 461, 2005 WL 425212, at *8. Given what is at stake in sentencing decisions — the potential for additional months or even years in prison — I believe that the increased administrative burdens are a tolerable price to pay.
*62c. Possibility of rebuttal
A presumption of prejudice still permits rebuttal by the government, as Barnett acknowledges. 398 F.3d at 529. Thus, there would be no automatic remands, whether based on the presence of judge-found facts and the use of mandatory guidelines or on the simple fact that mandatory guidelines were used as the basis for sentencing after a jury’s findings or a defendant’s admissions. Similarly, this approach also avoids treating Booker error as a structural error that “undermin[es] the fairness of a criminal proceeding as a whole.” Dominguez Benitez, 124 S.Ct. at 2339; see also Olano, 507 U.S. at 735, 113 S.Ct. 1770; Arizona v. Fulminante, 499 U.S. 279, 309-10, 111 S.Ct. 1246, 113 L.Ed.2d 302 (1991) (providing examples of structural error). As in cases of preserved error that we review for harmlessness, the government will be able to argue the absence of prejudice based on the entirety of the existing record.
Indeed, this case provides an apt example of that prospect. Here, Serrano faced a statutory mandatory minimum of 60 months’ imprisonment. After taking into account his role in the offense and his criminal history, the guidelines required (now, advise) a sentence within the range of 63 to 78 months, leaving little room for modification between the mandatory minimum of the statute and the guidelines minimum. The judge sentenced Serrano to 63 months. Even now, the district court “must consult those Guidelines and take them into account when sentencing.” Booker, 125 S.Ct. at 767. The court could have sentenced him below the 63 months of the guidelines only by totally ignoring the sentencing factors that it is still required to consider. Thus, the government could show that Serrano was not harmed by the mandatory nature of the guidelines.
5. Conclusion
Applying a presumption of prejudice in cases of Booker error would not be an innovation. As Barnett explains, the concept is well grounded in Olano and other circuit decisions in a variety of contexts. Hence, this approach remains faithful to Booker’s directive to the appellate courts to review unpreserved claims for plain error; it avoids automatic remands; and it responds fairly to the unique problems left in Booker’s wake.
. The Eleventh Circuit decided that the difficulty of speculating about the district court’s sentencing under advisory guidelines prevented the defendant from showing prejudice. That approach would seem to doom most unpreserved claims of Booker error on appeal. Rodriguez, 398 F.3d at 1301, 2005 WL 272952, at *10 (citing Jones v. United States, 527 U.S. 373, 394-95, 119 S.Ct. 2090, 144 L.Ed.2d 370 (1999)). While I share the prem-*58ce of the Eleventh Circuit about the difficulty of the speculation on a hypothetical sentence, I disagree with the reasoning that followed.
. I do not consider this approach foreclosed by Justice Breyer's closing words commending to the courts of appeal "ordinary prudential doctrines," like plain error. Booker, 125 S.Ct. at 769. Olano discussed presumption of prejudice in the context of plain-error review.
. The Third Circuit has also presumed prejudice after a constructive amendment of the indictment:
Similar to the plight of a defendant who is denied the right of allocution, it is very difficult for a defendant to prove prejudice resulting from most constructive amendments to an indictment.... Therefore, we will apply in the plain error context a rebut-table presumption that constructive amendments are prejudicial (and thus that they satisfy the third prong of plain error review).
United States v. Syme, 276 F.3d 131, 154 (3d Cir.2002).
. "The argument that a Booker error occurred is preserved if the defendant below argued Apprendi or Blakely error or that the Guidelines were unconstitutional.” Antonakopoulos, 399 F.3d at 76.
. The Second Circuit has relied on an intervening change in law to justify shifting the *61prejudice burden from the defendant to the government: "In this Circuit, when the error results from an intervening change in the law, we have applied a modified version of the plain error doctrine whereby the burden is on the Government to show that the error did not affect substantial rights.” United States v. Williams, 399 F.3d 450, 458 n. 7, 2005 WL 425212, at *5 n. 7 (2d Cir.2005); see also United States v. Viola, 35 F.3d 37, 41-42 (2d Cir.1994), abrogated on other grounds by Salinas v. United States, 522 U.S. 52, 118 S.Ct. 469, 139 L.Ed.2d 352 (1997). At the same time, the Second Circuit has wondered whether that so-called "modified version” has been implicitly rejected by Johnson v. United States, 520 U.S. 461, 117 S.Ct. 1544, 137 L.Ed.2d 718 (1997). See Williams, 399 F.3d at 458 n. 7; United States v. Thomas, 274 F.3d 655, 668 n. 15 (2d Cir.2001) (en banc).
Although I respect the Second Circuit's prudence, -I find its speculation puzzling. In . Johnson, the district court had- decided for itself the issue of materiality in a perjury persecution, instead of submitting it to the . jury. After Johnson was convicted, and before her appeal, the Supreme Court decided United States v. Gaudin, 515 U.S. 506, 115 S.Ct. 2310, 132 L.Ed.2d 444 (1995), which required materiality to be submitted to the jury. The Supreme Court held that an appellate court can correct an error that was not plain at trial, but became so on appeal. Johnson confined its discussion of prejudice to addressing petitioner's theory that her error should be considered structural • error and hence outside the ambit of Rule 52(b) altogether. The Court was dubious. Ultimately, the Court did not need to decide the issue because her case failed anyway at thé fourth step of the plain-error’ analysis because of the overwhelming evidence of materiality. Johnson, 520 U.S. at 470, 117 S.Ct. 1544. The issue of presumed prejudice — whether justified by a change in law or by othér reasons— was simply not before the Court.