BY THE COURT:

A petition for rehearing having been filed and a member of this Court in active service having requested a poll on whether this case should be reheard by the Court sitting en banc, and a majority of the judges in active service on this Court having voted against granting a rehearing en banc, it is ORDERED that this case will not be reheard en banc.

WILLIAM PRYOR, Circuit Judge, joined by JULIE CARNES, Circuit Judge, respecting the denial of rehearing en banc:

A majority of the Court has voted not to rehear en banc our decision in this appeal, United States v. Matchett, 802 F.3d 1185 (11th Cir. 2015), which held that the advisory sentencing guidelines cannot be challenged as void for vagueness. As members of the panel (and coincidentally the only members of this Court to have served on the United States Sentencing Commission), we write to explain why we agree with that decision.

We divide our discussion in two parts. First, we explain that Matchett is correct because the vagueness doctrine applies only to laws that regulate the primary conduct of private citizens. Advisory sentencing guidelines regulate judges, not private individuals; they guide judicial discretion within a statutory range. Advisory sentencing guidelines do not define crimes or fix punishments. Second, we explain that Matchett is not worthy of en banc rehearing.

A. Matchett Is Correct.

Our opinion held that advisory sentencing guidelines cannot be void for vagueness under the Due Process Clause of the Fifth Amendment. See id. at 1193-96. We reaffirm that holding. To explain why, we begin with a brief history of the federal sentencing guidelines.

Before the Civil War, Congress enacted very few criminal laws and "crime control was left largely to the states." Sara Sun Beale, Federalizing Crime: Assessing the Impact on the Federal Courts, 543 Annals Am. Acad. Pol. & Soc. Sci. 39, 40 (1996). The states "uniformly followed the common-law practice of making death the exclusive and mandatory sentence for certain specified offenses." Woodson v. North Carolina, 428 U.S. 280, 289, 96 S.Ct. 2978, 49 L.Ed.2d 944 (1976); see also Williams v. New York, 337 U.S. 241, 247-48, 69 S.Ct. 1079, 93 L.Ed. 1337 (1949). All crimes "had a defined punishment," and "the period of incarceration was generally prescribed with specificity by the legislature." Ilene H. Nagel, Structuring Sentencing Discretion: The New Federal Sentencing Guidelines, 80 J. Crim. L. & Criminology 883, 892 (1990).

After the Civil War, this system of fixed sentencing was gradually replaced with individualized sentencing. See id. at 893-95; United States v. Grayson, 438 U.S. 41, 45-46, 98 S.Ct. 2610, 57 L.Ed.2d 582 (1978). Legislatures enacted broad statutory ranges, and judges began tailoring sentences to individual defendants by considering "the circumstances of the offense together with the character and propensities of the offender." Pennsylvania ex rel. Sullivan v. Ashe, 302 U.S. 51, 55, 58 S.Ct. 59, 82 L.Ed. 43 (1937); see also Pepper v. United States, 562 U.S. 476, 487-88, 131 S.Ct. 1229, 179 L.Ed.2d 196 (2011). "So long as sentencing judges stayed within the statutory boundaries, they had unbridled discretion to arrive at any sentence they pleased." United States v. Irey, 612 F.3d 1160, 1180 (11th Cir. 2010) (en banc). As Judge Marvin Frankel described the state of sentencing in 1973, "The sentencing powers of the judges [were] ... so far unconfined that, except for frequently monstrous maximum [statutory] limits, they [were] effectively subject to no law at all." Marvin E. Frankel, Criminal Sentences: Law Without Order 8 (1973).

Although individualized sentencing was less draconian than fixed sentencing, it produced new problems. Because sentencing judges had unbridled discretion and no real standards to guide them, "[s]erious disparities in sentences ... were common." Mistretta v. United States, 488 U.S. 361, 365, 109 S.Ct. 647, 102 L.Ed.2d 714 (1989). "[J]udges of widely varying attitudes on sentencing, administering statutes that confer[red] huge measures of discretion, mete[d] out widely divergent sentences where the divergences [were] explainable only by the variations among the judges, not by material differences in the defendants or their crimes." Frankel, supra, at 21. And because sentencing judges focused on the individual history and characteristics of each offender, "the offender's race, sex, religion, income, education, occupation and other status characteristics were found to influence judicial outcomes." Nagel, supra, at 895.

Despite its flaws, individualized sentencing remains perfectly constitutional. "[L]egislatures remain free to decide how much discretion in sentencing should be reposed in the judge or jury in noncapital cases...." Lockett v. Ohio, 438 U.S. 586, 603, 98 S.Ct. 2954, 57 L.Ed.2d 973 (1978). The Supreme Court has "never doubted the authority of a judge to exercise broad discretion in imposing a sentence within a statutory range." United States v. Booker, 543 U.S. 220, 233, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005).

Because of the disparities associated with individualized sentencing, see Koon v. United States, 518 U.S. 81, 92, 116 S.Ct. 2035, 135 L.Ed.2d 392 (1996), and following a massive expansion of federal criminal law in the twentieth century, see Beale, supra, at 41-43, Congress enacted the Sentencing Reform Act of 1984, which created the United States Sentencing Commission and empowered it to promulgate federal sentencing guidelines, see 28 U.S.C. §§ 991, 994. The first guidelines came into effect in 1987. See United States Sentencing Guidelines Manual (Nov. 1987). They calculated a narrow sentencing range by assigning numeric values to the type of offense, the characteristics of the offense, the offender's criminal history, and other factors. See id. The initial guidelines were also mandatory: district judges were required to impose a sentence within the guideline range, subject to limited departures, see 18 U.S.C. § 3553(b)(1), and appellate courts reviewed departures from the guidelines de novo, see id. § 3742(e).

The mandatory guidelines were quickly challenged as unconstitutional. The Supreme Court rebuffed a separation-of-powers challenge to the guidelines in Mistretta v. United States, 488 U.S. 361, 109 S.Ct. 647, 102 L.Ed.2d 714 (1989). The Court concluded that the mandatory guidelines were not an impermissible exercise of the legislative power because they do not regulate primary conduct:

Id. at 396, 109 S.Ct. 647 (emphasis added). The mandatory guidelines were challenged again in United States v. Booker, 543 U.S. 220, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005) — this time under the Fifth and Sixth Amendments. A majority of the Supreme Court held that the mandatory guidelines were unconstitutional because they allowed judges to find facts, by a preponderance of the evidence, that increased a defendant's sentence. See id. at 230-44, 125 S.Ct. 738. Other than prior convictions, "[a]ny fact that, by law, increases the penalty for a crime" must be admitted by the defendant or found by a jury beyond a reasonable doubt because such facts are not mere sentencing provisions; they are elements of the crime. Alleyne v. United States, ___ U.S. ___, 133 S.Ct. 2151, 2155, 186 L.Ed.2d 314 (2013); see also Apprendi v. New Jersey, 530 U.S. 466, 476-90, 120 S.Ct. 2348, 147 L.Ed.2d 435 (2000). To remedy this constitutional defect, a different majority of the Booker Court invalidated the statutory provisions that made the guidelines mandatory. Booker, 543 U.S. at 245, 125 S.Ct. 738. What remained were guidelines that are "effectively advisory." Id.

Now that the guidelines are advisory, they continue to play an important role in sentencing, but they do not and cannot play a decisive one. On the one hand, the guidelines are the "starting point" for sentencing and the "lodestar." Molina-Martinez v. United States, ___ U.S. ___, 136 S.Ct. 1338, 1346, 194 L.Ed.2d 444 (2016). District courts must calculate the guideline range correctly, Gall v. United States, 552 U.S. 38, 51, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007), and the failure to do so can constitute plain error because "[i]n the usual case ... the selected Guidelines range will affect the sentence," Molina-Martinez, 136 S.Ct. at 1346. And appellate courts can presume that a sentence within the guideline range is reasonable, Rita v. United States, 551 U.S. 338, 347, 127 S.Ct. 2456, 168 L.Ed.2d 203 (2007), although this Court does not do so, United States v. Hunt, 526 F.3d 739, 746 (11th Cir. 2008). On the other hand, the guidelines are just one factor among many to be balanced against six other statutory sentencing factors, 18 U.S.C. § 3553(a). See Pepper, 562 U.S. at 490, 131 S.Ct. 1229; see also 18 U.S.C. § 3661 ("No limitation shall be placed on the information concerning the background, character, and conduct of a person convicted of an offense which a court of the United States may receive and consider for the purpose of imposing an appropriate sentence."). District courts commit reversible error if they "treat[] the Guidelines as mandatory," Gall, 552 U.S. at 51, 128 S.Ct. 586, and they cannot "presume that a sentence within the applicable Guidelines range is reasonable," Nelson v. United States, 555 U.S. 350, 352, 129 S.Ct. 890, 172 L.Ed.2d 719 (2009). Appellate courts cannot presume that a deviation from the guideline range is unreasonable. Rita, 551 U.S. at 354-55, 127 S.Ct. 2456. A district judge may even refuse to follow the guidelines "based on [a] policy disagreement" with the Sentencing Commission, Spears v. United States, 555 U.S. 261, 264, 129 S.Ct. 840, 172 L.Ed.2d 596 (2009) (citing Kimbrough v. United States, 552 U.S. 85, 128 S.Ct. 558, 169 L.Ed.2d 481 (2007)).

The question presented in this appeal was whether the advisory guidelines can be void for vagueness. The vagueness doctrine states that "a statute which either forbids or requires the doing of an act in terms so vague that men of common intelligence must necessarily guess at its meaning and differ as to its application violates the first essential of due process of law." Connally v. Gen. Constr. Co., 269 U.S. 385, 391, 46 S.Ct. 126, 70 L.Ed. 322 (1926). As the decisions of the Supreme Court make clear, the vagueness doctrine applies only to laws that regulate private conduct — laws that define crimes, e.g., City of Chicago v. Morales, 527 U.S. 41, 119 S.Ct. 1849, 144 L.Ed.2d 67 (1999); Kolender v. Lawson, 461 U.S. 352, 103 S.Ct. 1855, 75 L.Ed.2d 903 (1983); Colautti v. Franklin, 439 U.S. 379, 99 S.Ct. 675, 58 L.Ed.2d 596 (1979); Smith v. Goguen, 415 U.S. 566, 94 S.Ct. 1242, 39 L.Ed.2d 605 (1974); Papachristou v. City of Jacksonville, 405 U.S. 156, 92 S.Ct. 839, 31 L.Ed.2d 110 (1972); Winters v. New York, 333 U.S. 507, 68 S.Ct. 665, 92 L.Ed. 840 (1948); Lanzetta v. New Jersey, 306 U.S. 451, 59 S.Ct. 618, 83 L.Ed. 888 (1939); Cline v. Frink Dairy Co., 274 U.S. 445, 47 S.Ct. 681, 71 L.Ed. 1146 (1927); United States v. L. Cohen Grocery Co., 255 U.S. 81, 41 S.Ct. 298, 65 L.Ed. 516 (1921); Int'l Harvester Co. of Am. v. Kentucky, 234 U.S. 216, 34 S.Ct. 853, 58 L.Ed. 1284 (1914); laws that fix sentences, e.g., Johnson v. United States, ___ U.S. ___, 135 S.Ct. 2551, 192 L.Ed.2d 569 (2015); laws that restrict speech, e.g., FCC v. Fox Television Stations, Inc., ___, U.S. ___, 132 S.Ct. 2307, 183 L.Ed.2d 234 (2012); Gentile v. State Bar of Nevada, 501 U.S. 1030, 111 S.Ct. 2720, 115 L.Ed.2d 888 (1991); Keyishian v. Bd. of Regents of Univ. of State of N.Y., 385 U.S. 589, 87 S.Ct. 675, 17 L.Ed.2d 629 (1967); Baggett v. Bullitt, 377 U.S. 360, 84 S.Ct. 1316, 12 L.Ed.2d 377 (1964); Cramp v. Bd. of Pub. Instruction of Orange Cty., 368 U.S. 278, 82 S.Ct. 275, 7 L.Ed.2d 285 (1961); and laws that regulate businesses, e.g., A.B. Small Co. v. Am. Sugar Ref. Co., 267 U.S. 233, 45 S.Ct. 295, 69 L.Ed. 589 (1925).

The vagueness doctrine reflects two "connected but discrete" concerns: notice and arbitrary enforcement. Fox Television Stations, 132 S.Ct. at 2317. Notice means that a law does not "give the person of ordinary intelligence a reasonable opportunity to know what is prohibited, so that he may act accordingly." Grayned v. City of Rockford, 408 U.S. 104, 108, 92 S.Ct. 2294, 33 L.Ed.2d 222 (1972). Arbitrary enforcement means that a law leaves government actors "free to decide, without any legally fixed standards, what is prohibited and what is not in each particular case." Giaccio v. Pennsylvania, 382 U.S. 399, 402-03, 86 S.Ct. 518, 15 L.Ed.2d 447 (1966). The guidelines implicate neither of these concerns. See United States v. Hurlburt, 835 F.3d 715, 726, No. 14-3611, 2016 WL 4506717 (7th Cir. Aug. 29, 2016) (en banc) (Hamilton, J., dissenting) ("After all, how can non-binding advice be unconstitutionally vague?").

With respect to notice, the advisory guidelines cannot notify a defendant of what sentence he will receive because they are just that — advisory. Consider a person who is thinking about committing a federal crime and wants to know what punishment he will receive if he gets caught. He can identify a hard ceiling (the statutory maximum sentence) and a hard floor (the statutory minimum sentence). But he cannot identify the sentence he will receive within the statutory range. He could calculate his guideline range, but the guidelines are just one of seven sentencing factors that the sentencing judge will consider. The other six include factors like "the history and characteristics of the defendant"; the need to "promote respect for the law," to "provide just punishment," to "afford adequate deterrence," and to "provide the defendant with needed ... correctional treatment"; and the need to "avoid unwarranted sentence disparities among defendants with similar records who have been found guilty of similar conduct." 18 U.S.C. § 3553(a). Each of these factors is "vague and ... hopelessly open-ended." United States v. Dean, 414 F.3d 725, 729 (7th Cir. 2005). Even if a person could predict what a judge would find relevant about his "history and characteristics" or what a judge thinks makes a punishment "just," the sentencing statute makes "no attempt to give weights to the different factors." United States v. Siegel, 753 F.3d 705, 707 (7th Cir. 2014). "[W]ithout weighting its factors a multifactor test is not a test but a list, and cannot yield an objective result." Id.

The data collected by the Sentencing Commission reveal the difficulties of trying to predict a sentence based on the guidelines. On average, the odds of receiving a sentence within the guideline range are worse than a coin flip. See U.S. Sentencing Comm'n, 2015 Sourcebook of Federal Sentencing Statistics tbl. N (reporting that 47.3 percent of defendants received a sentence within the guideline range in fiscal year 2015). If a defendant is a career offender, then the odds are even worse — essentially 3:1 against. See U.S. Sentencing Comm'n, Report to the Congress: Career Offender Sentencing Enhancements 36 (2016) [hereinafter Career Offender Report] (reporting that 27.5 percent of career offenders received a sentence within the guideline range in fiscal year 2014).

The advisory nature of the guidelines explains why the Supreme Court held in Irizarry v. United States that "[a]ny expectation subject to due process protection... that a criminal defendant would receive a sentence within the presumptively applicable Guidelines range did not survive our decision in [Booker], which invalidated the mandatory features of the Guidelines." 553 U.S. 708, 713, 128 S.Ct. 2198, 171 L.Ed.2d 28 (2008). "The due process concerns that ... require notice in a world of mandatory Guidelines no longer" apply. Id. at 714, 128 S.Ct. 2198. Those concerns include the vagueness doctrine. See United States v. Wivell, 893 F.2d 156, 159 (8th Cir. 1990). The advisory guidelines do not implicate arbitrary enforcement either. Our dissenting colleagues criticize our decision for allegedly failing to consider arbitrary enforcement separately from notice, but these concerns are "connected," Fox Television Stations, 132 S.Ct. at 2317. And the panel opinion concluded that the guidelines implicate neither notice nor arbitrary enforcement for the same overarching reason. Both aspects of the vagueness doctrine are concerned only with laws that regulate the primary conduct of private individuals. See id. (describing the vagueness doctrine as governing "laws which regulate persons or entities"); Gen. Constr. Co., 269 U.S. at 391, 46 S.Ct. 126 (describing the vagueness doctrine as governing laws that "forbid[] or require[] the doing of an act"); United States v. Brierton, 165 F.3d 1133, 1139 (7th Cir. 1999) ("[T]he vagueness doctrine presumes a law that attempts to proscribe or prescribe conduct."); Wivell, 893 F.2d at 159 ("[B]oth theories supporting the vagueness doctrine presume a law that attempts to proscribe or prescribe conduct."); El-Shifa Pharm. Indus. Co. v. United States, 607 F.3d 836, 856 n.4 (D.C. Cir. 2010) (en banc) (Kavanaugh, J., concurring in the judgment) (describing the vagueness doctrine as governing laws that "regulat[e] private conduct"); Cass R. Sunstein, Problems with Rules, 83 Calif. L. Rev. 953, 968 (1995) ("[T]he `void for vagueness' doctrine requires the state to set forth clear guidance before it may punish private conduct.").

Arbitrary enforcement, for purposes of the vagueness doctrine, means arbitrary enforcement of laws that regulate the primary conduct of private individuals — hence the word "enforcement." See Papachristou, 405 U.S. at 170, 92 S.Ct. 839; Parker v. Levy, 417 U.S. 733, 775, 94 S.Ct. 2547, 41 L.Ed.2d 439 (1974). Vague laws allow for arbitrary enforcement because they do not meaningfully limit who the police can arrest, who prosecutors can prosecute, and who judges and juries can convict. See Lawson, 461 U.S. at 358-60, 103 S.Ct. 1855; Grayned, 408 U.S. at 108-09, 92 S.Ct. 2294. In the federal context, one of the main concerns behind arbitrary enforcement is the separation of powers: "In our system, ... defining crimes and fixing penalties are legislative, not judicial, functions," and vague statutes require courts to "fill[] gaps so large that doing so becomes essentially legislative." United States v. Evans, 333 U.S. 483, 486-87, 68 S.Ct. 634, 92 L.Ed. 823 (1948); accord Lawson, 461 U.S. at 358 n.7, 103 S.Ct. 1855 (citing United States v. Reese, 92 U.S. 214, 221, 23 L.Ed. 563 (1875)); Goguen, 415 U.S. at 575, 94 S.Ct. 1242.

Judge Martin's dissent contends that the vagueness doctrine is not limited to laws that regulate primary conduct, but her two counterexamples only support the opposite conclusion. First, Judge Martin's dissent points out that the vagueness doctrine applies to sentencing statutes, see United States v. Batchelder, 442 U.S. 114, 123, 99 S.Ct. 2198, 60 L.Ed.2d 755 (1979). But statutes that specify facts that increase a defendant's mandatory minimum or maximum sentence do regulate primary conduct because they define the elements of the crime. See Alleyne, 133 S.Ct. at 2158. After Booker, the advisory guidelines do no such thing. See Booker, 543 U.S. at 233, 125 S.Ct. 738. Second, Judge Martin's dissent asserts that the Supreme Court applied the vagueness doctrine to a statute that did not regulate primary conduct in Giaccio v. Pennsylvania, 382 U.S. 399, 86 S.Ct. 518, 15 L.Ed.2d 447 (1966). In Giaccio, the Supreme Court invalidated a Pennsylvania statute that required acquitted defendants to pay the costs of their prosecution. See id. at 403, 86 S.Ct. 518. A jury could award the costs as a "sentence" if it found that the defendant was "guilty of some misconduct less than the offense which is charged but nevertheless misconduct of some kind as a result of which he should be required to pay some penalty short of conviction (and) ... his misconduct has given rise to the prosecution." Id. at 403-04, 86 S.Ct. 518. This statute as construed by the Pennsylvania courts plainly regulated primary conduct: it imposed a penalty (costs, and possibly jail time) on an acquitted defendant (a private individual) based on his out-of-court conduct (the misconduct that led to his prosecution). Contrary to Judge Martin's characterization, the statute in Giaccio did not impose administrative court costs; it imposed a penalty for out-of-court conduct by a private citizen. See Schilb v. Kuebel, 404 U.S. 357, 370, 92 S.Ct. 479, 30 L.Ed.2d 502 (1971) (distinguishing the statute in Giaccio from "an administrative cost imposed upon ... guilty and innocent alike"). The Supreme Court has never applied the vagueness doctrine to a statute that did not regulate primary conduct.

We know that the federal sentencing guidelines do not regulate primary conduct. In Mistretta, the Supreme Court held that the guidelines "do not bind or regulate the primary conduct of the public." 488 U.S. at 396, 109 S.Ct. 647. The guidelines "fetter the discretion of sentencing judges"; they do not "establish[] minimum and maximum penalties" for crimes. Id.; accord Brierton, 165 F.3d at 1139 ("The Guidelines do not establish the illegality of any conduct. Rather, they ... are designed to assist and limit the discretion of the sentencing judge."); Wivell, 893 F.2d at 160 ("The Sentencing Guidelines do not define illegal conduct: they are directives to judges for their guidance in sentencing convicted criminals, not to citizens at large.").

The term "arbitrary enforcement" makes little sense in this context. If judges exercising their sentencing discretion are "enforcing" the law against individuals, then the former system of individualized sentencing should have been void for vagueness. After all, the former system of individualized sentencing imposed no standards on judges, provided no notice to individual defendants, and resulted in disparate, arbitrary, and discriminatory sentences. See Frankel, supra, at 5. If individualized sentencing, which gives judges unbridled discretion, is constitutional, see Lockett, 438 U.S. at 603, 98 S.Ct. 2954, then vague sentencing guidelines, which only guide that discretion, are too.

The decision of the Supreme Court in Peugh v. United States, ___ U.S. ___, 133 S.Ct. 2072, 186 L.Ed.2d 84 (2013), does not mean that the guidelines can be void for vagueness. Peugh held that the guidelines can violate the Ex Post Facto Clause, see id. at 2088, but it said nothing about the vagueness doctrine. In fact, a plurality of the Court explained that its decision did not implicate the holding in Irizarry that "a defendant does not have an `expectation subject to due process protection' that he will be sentenced within the Guidelines range." Id. at 2085 (plurality opinion) (quoting Irizarry, 553 U.S. at 713-14, 128 S.Ct. 2198). And for good reason: the Ex Post Facto Clause and the vagueness doctrine have different scopes. Then-Judge Sotomayor has explained in detail why, although the Ex Post Facto Clause and the vagueness doctrine "share a concern for notice," "they are not necessarily identical in scope." Sash v. Zenk, 439 F.3d 61, 65 (2d Cir. 2006). Unlike the vagueness doctrine, "the Ex Post Facto Clause does not merely protect reliance interests. It also reflects principles of `fundamental justice.'" Peugh, 133 S.Ct. at 2085 (plurality opinion) (quoting Carmell v. Texas, 529 U.S. 513, 531, 120 S.Ct. 1620, 146 L.Ed.2d 577 (2000)); see also Sash, 439 F.3d at 64-66. Specifically, the Ex Post Facto Clause is "broader" and "more expansive" because it "is concerned not just with notice, but with the inherent injustice associated with retroactivity itself." Sash, 439 F.3d at 64-65. And the Ex Post Facto Clause is not limited to laws that regulate primary conduct. It applies to any change in law that "creates a `significant risk' of a higher sentence," Peugh, 133 S.Ct. at 2088 (majority opinion) — including, for example, changes to the rules of evidence, see id. at 2081 (citing Calder v. Bull, 3 U.S. 3 (Dall.) 386, 390, 1 L.Ed. 648 (1798) (opinion of Chase, J.)). That the guidelines can be ex post facto laws does not mean that they can be void for vagueness. Different doctrines with different purposes should be assessed differently. See Sash, 439 F.3d at 65 n.2.

Nor does the decision of the Supreme Court in Johnson v. United States, ___ U.S. ___, 135 S.Ct. 2551, 192 L.Ed.2d 569 (2015), mean that the guidelines can be void for vagueness. Johnson held that the residual clause of the Armed Career Criminal Act — a criminal statute that regulates primary conduct — was void for vagueness. See id. at 2557. The Court did not address the constitutionality of the career-offender guideline or any other guideline for that matter. Of course, the residual clause of the career-offender guideline is "virtually identical" to the residual clause of the Armed Career Criminal Act. United States v. Archer, 531 F.3d 1347, 1352 (11th Cir. 2008) (quoting United States v. Rainey, 362 F.3d 733, 735 (11th Cir. 2004)). So, if the guidelines are subject to the vagueness doctrine, then the residual clause of the career-offender guideline is void for vagueness because its language is virtually identical to the statutory language that was invalidated in Johnson. But that's a big "if," for the reasons given above, and Johnson says nothing whatsoever about the issue. See In re Embry, 831 F.3d 377, 380, No. 16-5447, 2016 WL, 4056056 (6th Cir. July 29, 2016) ("Before a court could invoke Johnson to invalidate the residual clause of the Guidelines, it would have to resolve a threshold question: Do[es] the [vagueness doctrine] apply to the advisory Sentencing Guidelines? The answer to this gateway question is not self-evident." (citation omitted)); Donnell v. United States, 826 F.3d 1014, 1015 (8th Cir.2016) ("Whether an advisory sentencing guideline is susceptible to a vagueness challenge is an open question in this circuit ... and the answer is not dictated by Johnson."); United States v. Gonzalez-Longoria, 831 F.3d 670, 679-80, No. 15-40041, 2016 WL 4169127 (5th Cir. Aug. 5, 2016) (en banc) (Jones, J., concurring) ("Johnson itself did not specifically treat or address the Guidelines," and "[n]o other Supreme Court case has directly confronted a vagueness challenge to the Guidelines or implied the propriety of such a challenge.").

The dissents are troubled by the possibility that judges must apply a guideline with language that Johnson held was vague, but the task is not as unusual or as daunting as they suggest. Johnson held that the language of the residual clause was too vague to be included in the Armed Career Criminal Act, a law that regulates primary conduct. See Johnson, 135 S.Ct. at 2557. But the advisory guidelines are directed to judges, not private citizens, and we tolerate much more vagueness in laws that regulate government actors than we do in laws that regulate private citizens. See Mahler v. Eby, 264 U.S. 32, 40-41, 44 S.Ct. 283, 68 L.Ed. 549 (1924) (distinguishing between vague statutes that provide discretion to government actors, which cannot be void for vagueness, and vague statutes that define crimes, which can); see also Nat'l Endowment for the Arts v. Finley, 524 U.S. 569, 587-88, 118 S.Ct. 2168, 141 L.Ed.2d 500 (1998) ("[T]he Government may allocate competitive funding according to criteria that would be impermissible were direct regulation of speech or a criminal penalty at stake."); Robert C. Post, Reconceptualizing Vagueness: Legal Rules and Social Orders, 82 Calif. L. Rev. 491, 492 (1994) ("[T]he doctrine of undue discretion is most often used when laws are addressed to official decisionmakers, whereas vagueness doctrine is typically used when legal rules directly constrain the conduct of ordinary citizens....").

Vague standards regulate government officials, including judges, all the time — searches must be "reasonable," U.S. Const. amend. IV; trials must be "speedy," id. amend. VI; and regulations must be "in the public interest," 47 U.S.C. § 201(b). Consider the statutory sentencing factors: District judges must impose a sentence that is "sufficient, but not greater than necessary," to "promote respect for the law," to "provide just punishment," and to "afford adequate deterrence," among other things. 18 U.S.C. § 3553(a). And appellate judges must review whether a sentence is "unreasonable." Booker, 543 U.S. at 261, 125 S.Ct. 738. All of these standards might be void for vagueness if they appeared in a statute that regulated private individuals. Cf. L. Cohen Grocery Co., 255 U.S. at 89, 41 S.Ct. 298 (explaining that the phrase "detrimental to the public interest" would be void for vagueness in a criminal statute); Finley, 524 U.S. at 588, 118 S.Ct. 2168 (explaining that the word "respect" is "undeniably opaque" and "could raise substantial vagueness concerns" if it "appeared in a criminal statute or regulatory scheme"); EPA v. EME Homer City Generation, L.P., ___ U.S. ___, 134 S.Ct. 1584, 1616, 188 L.Ed.2d 775 (2014) (Scalia, J., dissenting) ("One could hardly overstate the capaciousness of the word `adequate'...."); Frink Dairy Co., 274 U.S. at 465, 47 S.Ct. 681 (holding that the phrase "reasonable profit" was void for vagueness in a criminal statute). But federal judges apply these standards every day and do so without a second thought. The dissents appear to think that there is a due-process right of judges to be free from interpreting vague laws in the exercise of their judicial duty. That "right" does not exist.

Furthermore, judges who must apply the residual clause of the career-offender guideline are not hopelessly adrift. Johnson held that the residual clause is vague in many of its applications, but it acknowledged that "there will be straightforward cases under the residual clause" and that "there is some conduct that clearly falls within the provision's grasp." 135 S.Ct. at 2560-61. Judges will continue to see examples of "obviously risky crimes" that "clearly pose a serious potential risk of physical injury to another." Id. Indeed, federal circuit and district judges interpreted this language thousands of times before Johnson. The Supreme Court did so four times as well. That caselaw is still on the books, and it can guide judges as they continue to do what they have always done when interpreting vague provisions directed at government officials: "liquidat[ing] and ascertain[ing]" their meaning through "a series of particular discussions and adjudications," The Federalist No. 37, at 236 (James Madison) (Carl Van Doren ed., Easton Press, collector's ed. 1979).

Of course, a guideline that is too vague is not a good guideline. Fortunately, mechanisms short of constitutional invalidation already exist to deal with bad guidelines. Most notably, the Sentencing Commission can and does repeal guidelines that are difficult to apply. In fact, it did so here.

As of August 1, 2016, the residual clause of the career-offender guideline no longer exists. See U.S. Sentencing Comm'n, Amendment to the Sentencing Guidelines 2 (Jan. 21, 2016). The Commission concluded that the residual clause should be repealed "as a matter of policy" based in part on the "considerable application difficulties" that the Supreme Court outlined in Johnson. Id. We have no reason to doubt that the Commission will continue to fulfill its ongoing duty to "periodically ... review and revise ... the guidelines" in the light of new data and commentary from all sectors of the federal criminal justice system. 28 U.S.C. § 994(o).

In addition to the Sentencing Commission, individual judges can vary from vague guidelines on policy grounds. See Spears, 555 U.S. at 264-66, 129 S.Ct. 840. That a guideline cannot be applied consistently could be a legitimate reason for a district judge to conclude that the guideline does not, for example, "avoid unwarranted sentence disparities among defendants with similar records who have been found guilty of similar conduct," 18 U.S.C. § 3553(a)(6). We do not mean to suggest that every problem can be solved by the "noblesse oblige" of sentencing commissioners and district judges, Fox Television Stations, 132 S.Ct. at 2318 (quoting United States v. Stevens, 559 U.S. 460, 480, 130 S.Ct. 1577, 176 L.Ed.2d 435 (2010)); but as the author of Johnson was fond of reminding us, "not every problem was meant to be solved by the United States Constitution," Herrera v. Collins, 506 U.S. 390, 428 113 S.Ct. 853, 122 L.Ed.2d 203 (1993) (Scalia, J., concurring).

Judge Rosenbaum's concurrences on this subject, appended to her dissent, fret that, unless the guidelines can be challenged as unconstitutionally vague, nothing would prevent the Sentencing Commission from issuing a nonsensical guideline about "cheese," see Rosenbaum Dissent app. at 22, but this concern is unfounded. For one, Judge Rosenbaum's concurrences fail to appreciate the difference between vagueness and unintelligibility. Vagueness is "[u]ncertain breadth of meaning; unclarity resulting from abstract expression." Vagueness, Black's Law Dictionary (10th ed. 2014). Unintelligibility is "[i]ncomprehensibility; the quality of being incapable of being understood." Unintelligibility, Black's Law Dictionary. "Vagueness ... is often intentional, as general terms (reasonable time, best efforts, equal protection) are adopted to cover a multitude of situations that cannot practicably be spelled out in detail or even foreseen." Antonin Scalia & Bryan A. Garner, Reading Law 32-33 (2012). But "[a]n unintelligible text is inoperative." Id. at 134. Judges would not apply a gibberish-filled guideline about "cheese" because such a guideline would be unintelligible. See id. at 134-39. The vagueness doctrine would have nothing to do with it. See Johnson, 135 S.Ct. at 2568 n.3 (Thomas, J., concurring in the judgment); Ralph W. Aigler, Legislation in Vague or General Terms, 21 Mich. L. Rev. 831, 834 (1923). For another, the members of the United States Sentencing Commission take their oaths seriously and strive to craft guidelines that "provide certainty and fairness in meeting the purposes of sentencing," 28 U.S.C. § 991(b)(1)(B). We know of no instance in which the Commission has intentionally or inadvertently constructed a "word salad," Rosenbaum Dissent app. at 30.

Although we have little to lose if the guidelines cannot be void for vagueness, the converse is not true. Many provisions of the federal guidelines might not withstand scrutiny under the vagueness doctrine. See Matchett, 802 F.3d at 1196. See also Hurlburt, 835 F.3d at 729, No. 14-3611, (Hamilton, J., dissenting) (suggesting that provisions for "sophisticated means," "vulnerable victim," "otherwise extensive" criminal activity, "relevant conduct," and several departures could be found to be unconstitutionally vague) (citing United States Sentencing Guidelines Manual §§ 2B1.1(b)(1)(10), 3A1.1(b), 3B1.1, 1B1.3). Twenty states also have sentencing guidelines, see generally Neal Kauder & Brian Ostrom, Nat'l Ctr. for State Courts, State Sentencing Guidelines: Profiles and Continuum (2008), and any interpretation of the Due Process Clause of the Fifth Amendment applies to the Due Process Clause of the Fourteenth Amendment, see Hurtado v. California, 110 U.S. 516, 534-35, 4 S.Ct. 292, 28 L.Ed. 232 (1884). A decision holding that the federal guidelines can be void for vagueness could lead to the invalidation of many state guidelines as well.

Further on the horizon, a decision holding that the advisory guidelines can be void for vagueness would be in considerable tension with the rulings that have upheld the constitutionality of the sentencing guidelines. Mistretta held that the guidelines do not violate the separation of powers because they "do not bind or regulate the primary conduct of the public or ... establish[] minimum and maximum penalties for every crime." Mistretta, 488 U.S. at 396, 109 S.Ct. 647. But if the guidelines can be void for vagueness because they are laws that fix punishments and regulate primary conduct, then the guidelines look more like an exercise of the legislative power that the Constitution vests exclusively in Congress, see Evans, 333 U.S. at 486, 68 S.Ct. 634. Booker held that the Fifth and Sixth Amendments require the guidelines to be advisory. But if the guidelines can be void for vagueness because they are the be-all and end-all of sentencing, then the guidelines look more presumptive and less advisory. See Kimbrough, 552 U.S. at 113-14, 128 S.Ct. 558 (2007) (Scalia, J., concurring) ("If there is any thumb on the scales[,] ... then the `advisory' Guidelines would, over a large expanse of their application, entitle the defendant to a lesser sentence but for the presence of certain additional facts found by judge rather than jury. This, as we said in Booker, would violate the Sixth Amendment."). True, applying the Ex Post Facto Clause to the advisory guidelines did not undermine Booker because "the Sixth Amendment and Ex Post Facto Clause inquiries are analytically distinct." Peugh, 133 S.Ct. at 2088. But the Sixth Amendment and the Due Process Clause are not so distinct; the Supreme Court often has linked them together. See, e.g., Irizarry, 553 U.S. at 713, 128 S.Ct. 2198; Yu Cong Eng v. Trinidad, 271 U.S. 500, 518, 46 S.Ct. 619, 70 L.Ed. 1059 (1926); L. Cohen Grocery Co., 255 U.S. at 92, 41 S.Ct. 298. See also Hurlburt, 835 F.3d at 727, No. 14-3611, (Hamilton, J., dissenting) ("If the Supreme Court extends the rationale of Peugh ... that result would be difficult to reconcile with the Booker remedy, which spared the Guidelines from Sixth Amendment challenges by making them advisory.").

If advisory guidelines can be void for vagueness, then the task for sentencing commissions just got harder. Advisory guidelines are not always drafted with the precision of laws that regulate primary conduct; vagueness can be a virtue in the case-by-case world of sentencing. And a decision undermining the constitutionality of advisory guidelines would be unfortunate because, although they are not perfect, advisory guidelines strike a much better balance between consistency, predictability, and flexibility than purely individualized sentencing and rigid fixed sentencing. "So to treat the due-process clause would hinder if not preclude ... progressive efforts to improve the administration of criminal justice." Williams, 337 U.S. at 251, 69 S.Ct. 1079. See also Hurlburt, 835 F.3d at 729, No. 14-3611, (Hamilton, J., dissenting) ("[T]his pervasive vagueness in Guideline provisions is not a bug in the system. It is a feature. It is intended to provide sentencing judges with needed flexibility.").

B. Matchett Is Not Worthy of En Banc Review.

This appeal also does not meet the traditional criteria for rehearing en banc. As we have explained, our decision is correct. Correct decisions are never worthy of en banc review.

Judge Martin's dissent portrays Matchett as a legal pariah, a decision that supposedly contradicts the "uniform view" of ten other circuits, see Martin Dissent at 1133-34 & n.1, but her math is hard to follow. Four circuits have held, in a published opinion, that the advisory guidelines can be void for vagueness. See Hurlburt, No. 14-3611; United States v. Calabretta, 831 F.3d 128, No. 14-3969, 2016 WL 3997215 (3d Cir. July 26, 2016); United States v. Pawlak, 822 F.3d 902 (6th Cir. 2016); United States v. Madrid, 805 F.3d 1204 (10th Cir. 2015). The other circuits cited in Judge Martin's dissent merely assumed, based on the government's concession, that the advisory guidelines could be void for vagueness. Such assumptions are not holdings and do not create precedent. See Casey v. United States, 343 U.S. 808, 808, 72 S.Ct. 999, 96 L.Ed. 1317 (1952). In fact, binding precedent in two of those circuits currently states that the guidelines cannot be void for vagueness. See Wivell, 893 F.2d at 160; United States v. Pearson, 910 F.2d 221, 223 (5th Cir. 1990). And several distinguished judges have explained why those precedents should still be good law after Johnson and Peugh. See United States v. Hurlburt, 835 F.3d at 726-30, No. 14-3611, (Hamilton, J., joined by Posner, J., Flaum, J., and Easterbrook, J., dissenting) (describing Matchett as "careful and persuasive"); Gonzalez-Longoria, 831 F.3d at 679-86, No. 15-40041, (en banc) (Jones, J., joined by Smith, J., concurring); United States v. Lee, 821 F.3d 1124, 1133-35 (9th Cir. 2016) (Ikuta, J., dissenting); United States v. Taylor, 803 F.3d 931, 933-35 (8th Cir. 2015) (Colloton, J., dissenting). In short, when Judge Martin's dissent says "ten circuits," it really means "four circuits and the Department of the Justice." And Judge Martin's dissent ignores the numerous state courts that have rejected the possibility that their sentencing guidelines could be void for vagueness. See, e.g., State v. Rourke, 773 N.W.2d 913, 922 (Minn. 2009); State v. Baldwin, 150 Wn.2d 448, 78 P.3d 1005, 1010-12 (2003) (en banc); Hall v. State, 767 So.2d 560, 562-63 (Fla. 4th DCA 2000), approved, 826 So.2d 268 (Fla. 2002); People v. McLeod, 143 Mich.App. 262, 372 N.W.2d 526, 528 (1985); State v. Mushrush, 135 Ohio App.3d 99, 733 N.E.2d 252, 259 (1999) (op. of Winkler, J.).

By my count, the circuits are split 5-3 and the states generally agree with Matchett. Although conflict with other courts can be a compelling reason to rehear an appeal, that concern is less pressing when rehearing will, at best, move us from one side of the conflict to the other. The Supreme Court of the United States, not this Court, is best suited to resolve such disagreements. Indeed, the Supreme Court will soon consider whether the guidelines can be vague in Beckles v. United States, ___ U.S. ___, 136 S.Ct. 2510, ___ L.Ed.2d ___ (2016), and its decision could settle the issue once and for all. Rehearing Matchett en banc would not be a wise use of our resources.

Judge Martin's dissent advocates rehearing Matchett based on the plight of career offenders who were sentenced in the Eleventh Circuit, but of course there are costs on the other side. According to a recent study by the Sentencing Commission, one in ten federal prisoners are career offenders — or 20,000 total offenders. See Career Offender Report, supra, at 24. Of those career offenders, 80 percent are potentially eligible to benefit from an invalidation of the residual clause of the career-offender guideline because they have a conviction for a crime of violence. See id. at 28. But once they are released from prison, career offenders with at least one conviction for a crime of violence recidivate at a rate of almost 70 percent — much higher than the recidivism rate for other federal prisoners, even though career offenders are older when released from prison after serving long sentences. See id. at 40-41. Their recidivating crimes of choice are most often violent — assault (28.6 percent) and robbery (35.3 percent). Id. at 42. Undoubtedly, smaller percentages of these offenders commit even more serious crimes like murder, rape, and kidnapping. Although Judge Martin's dissent empathizes with the prisoners who were sentenced in the Eleventh Circuit under the career-offender guideline, they are not the only "victims" our decision might affect.

Any cost-benefit analysis is better left to the Sentencing Commission, an expert agency that gathers data on disparities and recidivism and amends or repeals the guidelines accordingly. Our job is to get the law right. In our view, the panel opinion in Matchett correctly assessed the longstanding limits on the vagueness doctrine and correctly held that it does not apply to advisory sentencing guidelines. We concur in the denial of rehearing en banc.

WILSON, Circuit Judge, joined by JILL PRYOR, Circuit Judge, dissenting from the denial of rehearing en banc:

It is a violation of due process for a court to rely on a criminal sentencing scheme "so vague that it fails to give ordinary people fair notice ... or so standardless that it invites arbitrary enforcement."1 See Johnson v. United States, 576 U.S. ___, ___, 135 S.Ct. 2551, 2556-57, 192 L.Ed.2d 569 (2015). Post-Booker,2 the touchstone of appellate review of sentences is reasonableness — an inquiry that turns on a district court's application of the United States Sentencing Guidelines (Guidelines). That is to say, Supreme Court precedent establishes an expectation that defendants will receive reasonable sentences, and we rely on the Guidelines to determine reasonableness. Because the Guidelines drive appellate review under this sentencing scheme, fatally vague Guidelines provisions necessarily result in both "arbitrary enforcement by [courts]" and denial of "fair notice." See id. at 2557. Therefore, vague Guidelines provisions violate the due process clause's void-for-vagueness doctrine. The Matchett panel's decision to the contrary is erroneous.

And importantly, given the "central," "significant role" that the Guidelines play in sentencing, see Molina-Martinez v. United States, 578 U.S. ___, ___, 136 S.Ct. 1338, 1341-42, 194 L.Ed.2d 444 (2016), Matchett's holding is unworkable. Appellate judges like myself must now review sentences that were imposed based on language that the Supreme Court has deemed "hopeless[ly] indetermina[te]" — the text of the residual clause in § 4b1.2(a) of the Guidelines. See Johnson, 135 S.Ct. at 2558.

For these reasons, as well as those set forth by Judge Martin and Judge Rosenbaum in their thoughtful dissents, our court should reconsider Matchett. Accordingly, I respectfully dissent from the denial of the request to rehear Matchett en banc.

I

Under our post-Booker sentencing regime, appellate courts must review all sentences for reasonableness, and the Guidelines direct each step of that review. See Gall v. United States, 552 U.S. 38, 51, 128 S.Ct. 586, 597, 169 L.Ed.2d 445 (2007). Consequently, "[t]he Sentencing Guidelines provide the framework for the tens of thousands of federal sentencing proceedings that occur each year." See Molina-Martinez, 136 S.Ct. at 1342.

We are required to assess the reasonableness of a sentence in two steps. See Gall, 552 U.S. at 51, 128 S.Ct. at 597. We "must first ensure that the district court committed no significant procedural error, such as failing to calculate (or improperly calculating) the Guidelines range ... or failing to adequately explain the chosen sentence — including an explanation for any deviation from the Guidelines range." Id. Next, we "consider the substantive reasonableness of the sentence." Id. In doing so, we "take into account the totality of the circumstances, including the extent of any variance from the Guidelines range." Id. Indeed, the Supreme Court has held that appellate courts may "apply a presumption of reasonableness" to a sentence within the Guidelines range, see Rita v. United States, 551 U.S. 338, 355, 127 S.Ct. 2456, 2467, 168 L.Ed.2d 203 (2007), and our court has concluded that a within-Guidelines-range sentence is ordinarily reasonable, see United States v. Docampo, 573 F.3d 1091, 1101 (11th Cir. 2009).

A

Considering the "central role" of the Guidelines in this analysis, see Molina-Martinez, 136 S.Ct. at 1341, an impossibly vague Guidelines provision guarantees arbitrary enforcement of the law and denial of fair notice to the public as to what constitutes a reasonable sentence. For example, given that the first step of reasonableness review requires us to determine whether the district court properly calculated the defendant's Guidelines range, arbitrary enforcement is a fait accompli when the defendant's range is based on a vague provision. A defendant's Guidelines range is dictated by which Guidelines provisions apply to the defendant. Hence, if the district court relied on a vague provision in calculating a defendant's range, then we must interpret that provision to decide whether it actually applies to the defendant. A vague provision, however, provides no "generally applicable test" for determining its reach. See Johnson, 135 S.Ct. at 2559. Confronted with such a provision, we will be forced to rely on "guesswork and intuition," see id. and our decisions regarding to which defendants it properly applies will be arbitrary. This means that our enforcement of the "reasonable sentence" requirement will be arbitrary. If two similar defendants simultaneously challenge a district court's application of the provision to them, one defendant's appeals panel could find her sentence unreasonable and vacate her sentence, while — by mere bad luck — the other defendant's panel could affirm his sentence. Thus, despite the Guidelines being advisory, a vague Guidelines provision can give rise to arbitrary enforcement of the law and dictate the treatment of defendants.3

B

Likewise, when a Guidelines provision is vague, it denies the public fair notice of the consequences of breaking the law. Because defendants have a clearly established expectation that they will receive reasonable sentences and the Guidelines serve as the foundation of the reasonableness analysis, the Guidelines provide notice as to the scope of an acceptable federal sentence. The public must look to the Guidelines to discern the types of sentence that can reasonably be imposed on them. Therefore, Matchett's rationale for holding that the void-for-vagueness doctrine does not apply to the Guidelines — that defendants cannot "look to the Guidelines for notice" — is unconvincing. See 802 F.3d at 1194 (quoting United States v. Tichenor, 683 F.3d 358, 365 (7th Cir. 2012), overruled by United States v. Hurlburt, 835 F.3d 715, Nos. 14-3611, 15-1686, 2016 WL 4506717 (7th Cir. August 29, 2016) (en banc)).

* * *

In sum, the Supreme Court has held that a sentencing scheme that either "invites arbitrary enforcement" or denies "fair notice" is unconstitutional. See Johnson, 135 S.Ct. at 2556-57. As demonstrated here, a fatally vague Guidelines provision does both. For this reason, vague Guidelines provisions must be treated no differently under the due process clause than vague criminal statutes. To hold otherwise ignores the reality of sentencing post-Booker: the Guidelines are enmeshed in our case law and we heavily depend on them to fulfill our judicial duties. See Molina-Martinez, 136 S.Ct. at 1346 ("[T]he Guidelines are not only the starting point for most federal sentencing proceedings but also the lodestar. The Guidelines inform and instruct the district court's determination of an appropriate sentence.").

II

Turning to the specific impact of Matchett on our appellate review process, the "hopeless indeterminacy" of the residual clause in § 4b1.2(a) makes our charge to review the reasonableness of sentences based on that clause all but impossible. See Johnson, 135 S.Ct. at 2558. As discussed above, our first step when reviewing a sentence is to determine whether the defendant's Guidelines range was properly calculated. Accordingly, when faced with an appeal in which the district court found that the defendant qualified for a particular sentence under the residual clause, we must decide whether that frustratingly opaque clause applies to the defendant. As made clear in Johnson, this is a futile inquiry. See id. at 2560. In the face of such an unworkable task, appellate review of the defendant's sentence is not only impracticable but also "does not comport with the Constitution's guarantee of due process." See id.

I respectfully dissent.

MARTIN, Circuit Judge, joined by JILL PRYOR, Circuit Judge, dissenting from the denial of rehearing en banc:

The United States Sentencing Guidelines generally call for longer prison sentences for defendants who have a history of criminal convictions. Calvin Matchett's sentence was based, in part, on a guideline provision that calls for a harsher punishment for people whose earlier crime "involves conduct that presents a serious potential risk of physical injury to another." USSG § 4B1.2(a)(2). The Supreme Court has told us that these 13 words, referred to as the "residual clause" and also found in the Armed Career Criminal Act (ACCA), are so vague that prison sentences based on them violate the Due Process Clause of the Fifth Amendment. See Johnson v. United States, ___ U.S. ___, 135 S.Ct. 2551, 192 L.Ed.2d 569 (2015). In explaining its ruling, the Court analyzed cases applying these 13 words from the Sentencing Guidelines as well as cases applying the same words from ACCA. This review led to the Court's conclusion that these words have "proved nearly impossible to apply consistently." Id. at 2560 (quotation omitted).

The criticisms the Supreme Court leveled at this language in ACCA apply equally to the identical words found in the Sentencing Guidelines. Indeed, every other Court of Appeals (ten total) has either held or assumed that Johnson makes these 13 words unconstitutionally vague in both ACCA and the Sentencing Guidelines.1 The government argues the same. But in the face of this uniform view to the contrary, a panel of this court ruled that the vagueness doctrine does not apply to the Sentencing Guidelines, so neither does Johnson. See United States v. Matchett, 802 F.3d 1185 (11th Cir. 2015). By doing so, the panel has taken away all ability of judges on this court to give relief to people serving prison sentences that were calculated based on language that the Supreme Court said resulted in "more unpredictability and arbitrariness than the Due Process Clause tolerates."2 Johnson, 135 S.Ct. at 2558.

Over eleven months ago Mr. Matchett asked this court to rehear his case. Only now do we issue our ruling denying his petition for rehearing. In the intervening months, the Supreme Court granted certiorari in a case that will allow it to evaluate the panel opinion in Matchett.3 In doing so, the Court will be able to address the questions presented by Mr. Matchett's case. But in the meantime the panel's Matchett opinion will continue to dictate how this Circuit treats hundreds of prisoners who were sentenced in Alabama, Florida and Georgia. I had very much hoped that the outlier position that now constitutes binding precedent for our Circuit would be reconsidered by our court as a whole. It is with disappointment therefore that I dissent to the decision by the majority of this court not to rehear the appeal of Calvin Matchett.

I.

The panel's opinion in Matchett affects a lot of people. In establishing the framework for federal sentencing, the Sentencing Guidelines routinely set harsher sentencing ranges for people convicted of a "crime of violence" earlier in their lives. The Guidelines define this term "crime of violence" in three ways, one of which is the 13 words the Supreme Court ruled unconstitutional in Johnson: an offense that "involves conduct that presents a serious potential risk of physical injury to another." USSG § 4B1.2(a)(2). Perhaps the most severe sentencing enhancement that incorporates this definition is known as the "career offender" guideline, which applies to defendants with "at least two prior felony convictions of either a crime of violence or a controlled substance offense." USSG § 4B1.1. "Pursuant to that Guideline, each defendant who qualifies for career offender status is automatically placed in criminal history `Category VI,' the highest available under the Guidelines." United States v. LaBonte, 520 U.S. 751, 754, 117 S.Ct. 1673, 1675, 137 L.Ed.2d 1001 (1997). No matter how low a defendant's criminal history score otherwise might be, once he is found to be a career offender, he is placed in the worst class of offenders. In fiscal year 2014 alone 2,269 people around the country were sentenced using the "career offender" guideline.4

The 13 word definition of "crime of violence" found in § 4B1.2(a)(2) is also used in other guidelines, where it triggers harsher guideline ranges for people being sentenced under those guidelines. See, e.g., id. § 2K1.3 (the Guidelines section for crimes involving explosives, which requires a 12-level increase for two prior "crimes of violence"); id. § 2K2.1 (the section for firearm crimes, which was the basis for Mr. Matchett's sentence, and can double a defendant's guideline range for one "crime of violence" or triple or quadruple it based on two5); id. § 2S1.1 (the section for laundering crimes); id. § 7B1.1 (the section for probation and supervised release violations).

The decision of this court to leave the Matchett panel opinion in effect — even in light of the uniform rejection of its ruling by other federal courts of appeal — results in harsher treatment for prisoners who were sentenced in the Eleventh Circuit. Calvin Matchett's case demonstrates this point. Mr. Matchett was sentenced in Florida but is serving his sentence outside the Eleventh Circuit. See http://www.bop.gov/inmateloc. He is likely in prison with other inmates who committed the same federal crime he did. The other inmates were prosecuted by the same federal government and sentenced by federal judges who all took the same oath. But if he is the only inmate sentenced in the Eleventh Circuit, Mr. Matchett alone will serve out a prison term based on a standard that the Supreme Court has found results in "unavoidable uncertainty and arbitrariness of adjudication." Johnson, 135 S.Ct. at 2562. His fellow inmates sentenced outside the Eleventh Circuit have likely been back before a judge for a new sentence that does not violate Johnson.

II.

The Matchett panel opinion rests on two flawed assertions. It limits application of the vagueness doctrine and it makes an unjustified distinction between ACCA and the Sentencing Guidelines. Then Matchett stands on this foundation to put the Guidelines beyond the reach of the vagueness doctrine. Specifically, the panel says the vagueness doctrine applies only to "laws that regulate the primary conduct of private individuals." Pryor Op. at 10.6 With that, the Matchett panel reasons that the provisions of ACCA are subject to the vagueness doctrine because ACCA regulates private conduct. And even though sentencing judges administer the residual clause in the Sentencing Guidelines by exactly the same process that they used to administer the identical provision in ACCA, Matchett says the Guidelines "are directed to judges, not private citizens,"7 so the vagueness doctrine cannot apply. Pryor Op. at 17.

Neither the panel's characterization of the Sentencing Guidelines nor its characterization of the vagueness doctrine accurately reflects the state of the law. I will address each in turn.

A.

In explaining why the Supreme Court's holding in Johnson doesn't apply to the Guidelines, the panel relied on United States v. Tichenor, 683 F.3d 358 (7th Cir. 2012). The Seventh Circuit decided Tichenor before the Supreme Court's ruling in Johnson, and held in Tichenor that the Guidelines are not susceptible to challenge on vagueness grounds. The Matchett panel followed the lead of the Seventh Circuit, quoting from its opinion as follows: "`[S]ince the Guidelines are merely advisory, defendants cannot rely on them to communicate the sentence that the district court will impose. Defendants' inability to look to the Guidelines for notice underscores why ... they cannot bring vagueness challenges against the Guidelines.'" 802 F.3d at 1194 (quoting Tichenor, 653 F.3d at 365). But the Seventh Circuit's path in Tichenor is no longer there for us to follow. In light of the Supreme Court's decisions in Johnson and Peugh v. United States, ___ U.S. ___, 133 S.Ct. 2072, 186 L.Ed.2d 84 (2013), the Seventh Circuit has abandoned Tichenor. See Hurlburt, No. 14-3611. Indeed that court recognized that Johnson and Peugh "have fatally undermined [Tichenor's] reasoning." Id. at 3.

In Peugh, the Supreme Court applied the Ex Post Facto Clause to the Sentencing Guidelines. 133 S.Ct. at 2078. Like the vagueness doctrine, the Ex Post Facto Clause imposes a constitutional requirement of "fair notice." Weaver v. Graham, 450 U.S. 24, 30, 101 S.Ct. 960, 965, 67 L.Ed.2d 17 (1981). This right to fair notice, the Supreme Court concluded, is violated "when a defendant is sentenced under Guidelines promulgated after he committed his criminal acts and the new version provides a higher applicable Guidelines sentencing range than the version in place at the time of the offense." Peugh, 133 S.Ct. at 2078.

The notion that the Sentencing Guidelines do not regulate private conduct — or, in the words of Matchett, do not "fix punishments," 802 F.3d at 1189 — rests on the idea that the Guidelines play a minimal or optional role in the sentencing process. As Matchett put it, the Guidelines merely "`assist ... the sentencing judge' in determining a sentence." Id. at 1194 (quoting Tichenor, 653 F.3d at 364). But Peugh has put this misconception to rest. Peugh told us that although the Guidelines are advisory, they "remain the starting point for every sentencing calculation in the federal system." 133 S.Ct. at 2083. In fact, the Eleventh Circuit has long required district judges to calculate the guideline range for the person being sentenced. United States v. Crawford, 407 F.3d 1174, 1179 (11th Cir. 2005) (citing United States v. Shelton, 400 F.3d 1325, 1332 n.9 (11th Cir. 2005)). So while judges can choose a sentence outside what the Guidelines recommend, the Supreme Court has for good reason emphasized "the centrality of the Guidelines in the sentencing process." Molina-Martinez v. United States, ___ U.S. ___, 136 S.Ct. 1338, 1346, 194 L.Ed.2d 444 (2016). In Molina-Martinez, the Supreme Court again reminded us that the Guidelines are "the starting point for the district court's decision and anchor the court's discretion in selecting an appropriate sentence." Id. at 1349; see also id. at 1342 ("The Sentencing Guidelines provide the framework for the tens of thousands of federal sentencing proceedings that occur each year."); Peugh, 133 S.Ct. at 2087 ("District courts must begin their sentencing analysis with the Guidelines ... and use them to calculate the sentencing range correctly; and those Guidelines will anchor both the district court's discretion and the appellate review process.").8

The Guidelines have this "anchor" effect even when judges depart from them. "Even if the sentencing judge sees a reason to vary from the Guidelines, if the judge uses the sentencing range as the beginning point to explain the decision to deviate from it, then the Guidelines are in a real sense the basis for the sentence." Peugh, 133 S.Ct. at 2083 (quotation omitted). "In less than one-fifth of cases since 2007 have district courts imposed above- or below-Guidelines sentences absent a Government motion. Moreover, the Sentencing Commission's data indicate that when a Guidelines range moves up or down, offenders' sentences move with it." Id. at 2084 (citations omitted). Indeed, a vague guideline can wreak harm on a defendant even before he is convicted of any crime. See Peugh, 133 S.Ct. at 2085 (plurality opinion) (citation omitted) ("[A] defendant charged with an increased punishment for his crime is likely to feel enhanced pressure to plead guilty. This pressure does not disappear simply because the Guidelines range is advisory; the defendant will be aware that the range is intended to, and usually does, exert controlling influence on the sentence that the court will impose.").9 Peugh recognizes that the Sentencing Guidelines govern both how a judge decides a sentence, as well as the private conduct of the person who will be sentenced.

Peugh outright rejected the idea the Matchett panel relied on — that the Guidelines need not give notice because they are "purely advisory." Id. at 2087 (majority opinion). The Court wrote: "[i]t is simply not the case that the Sentencing Guidelines are merely a volume that the district court reads with academic interest in the course of sentencing." Id. Rather, the Guidelines are "the Federal Government's authoritative view of the appropriate sentences for specific crimes." Id. at 2085 (plurality opinion). And they announce "the most recent views of the agency charged by Congress with developing sentencing policy." Id. at 2087 (majority opinion). For these reasons the Guidelines must "give fair warning of their effect and permit individuals to rely on their meaning." Miller v. Florida, 482 U.S. 423, 430, 107 S.Ct. 2446, 2451, 96 L.Ed.2d 351 (1987) (quotation omitted) (applying the Ex Post Facto Clause to Florida's advisory sentencing guidelines); see also Peugh, 133 S.Ct. at 2085 (plurality opinion) ("The [Ex Post Facto] Clause ensures that individuals have fair warning of applicable laws.").

The vagueness doctrine must apply to the Sentencing Guidelines. Just as the Ex Post Facto Clause ensures "fair warning," the vagueness doctrine says no law can be "so vague that it fails to give ordinary people fair notice of the conduct it punishes." Johnson, 135 S.Ct. at 2556. The Matchett panel rejected Johnson, saying "advisory guidelines that inform a sentencing judge's discretion ... cannot violate the notice requirement." 802 F.3d at 1195. Peugh tells us the opposite.

The panel's idea that notice is not relevant to the Sentencing Guidelines seems to be based on its misreading of Irizarry v. United States, 553 U.S. 708, 128 S.Ct. 2198, 171 L.Ed.2d 28 (2008). See 802 F.3d at 1194; Pryor Op. at 10, 14. Irizarry held that Rule 32 of the Federal Rules of Criminal Procedure does not require a judge to say in advance of the sentence hearing what sentence she may impose. Id. at 709, 128 S.Ct. at 2200. In contrast, neither Peugh nor Johnson has anything to do with what a judge must say in any single case. Peugh and Johnson tackle the larger question of what notice is due the entire public about what punishment can be expected for a given offense. This type of notice is required from the "sentencing process long before the district court imposes the sentence." Molina-Martinez, 136 S.Ct. at 1342. That's why Peugh says that the Guidelines must give notice of what conduct the courts will punish, even if a particular judge is not required to give advance warning about how she will use her discretion in a given case.

The Matchett opinion ignores these lessons from Peugh, Johnson, and Molina-Martinez. Worse, the Matchett panel relies on decisions of the other courts of appeal that those courts have themselves recognized are no longer good law in light of Peugh, Johnson, and Molina-Martinez. For example, the panel purported to "join" the Sixth Circuit "insofar as we reject Matchett's argument that advisory guidelines can be unconstitutionally vague." 802 F.3d at 1196 (citing United States v. Smith, 73 F.3d 1414, 1418 (6th Cir. 1996)). As it happens, the Sixth Circuit has itself considered this same argument and held "that the rationale of Johnson applies equally to the residual clause of the Guidelines." United States v. Pawlak, 822 F.3d 902, 911 (6th Cir. 2016) (overruling Smith, 73 F.3d 1414). The Matchett panel also twice quoted from United States v. Wivell, 893 F.2d 156 (8th Cir. 1990), and so does today's Statement. See 802 F.3d at 1194-95, 1196; Pryor Op. at 11, 13. But the Eighth Circuit has recognized that "[t]he reasoning in Wivell that the guidelines cannot be unconstitutionally vague because they do not proscribe conduct is doubtful after Johnson." Taylor, 803 F.3d at 933. And as I mentioned, the Tichenor opinion, quoted by the Matchett panel four times, has now been overruled by the Seventh Circuit sitting en banc. Hurlburt, No. 14-3611. This leaves only the Fifth Circuit among the courts relied upon by the Matchett panel, and it has also vacated § 4B1.2 sentences in light of Johnson. See, e.g., Estrada, No. 15-40264. The Eleventh Circuit is now all alone on this.

Finally, in his Statement Judge Pryor tells us that federal judges depart from the Sentencing Guidelines so often that the "guidelines cannot notify a defendant of what sentence he will receive." Pryor Op. at 8. In fact, he says "the odds of receiving a sentence within the guideline range are worse than a coin flip." Id. at 9. While he cites statistics that make this seem true, the fact is that district judges give the Sentencing Guidelines much more deference than he lets on.

It is true that in 52.7 percent of all sentencings in 2015 the judge did not sentence the defendant within the guideline range. U.S. Sentencing Comm'n, 2015 Sourcebook of Federal Sentencing Statistics tbl. N. However, the statistics also tell us that for all sentences imposed during 2015, in which a judge imposed a sentence below the guideline range, 58 percent of the time the judge did so based on the government's motion.10 Id. This means that in the majority of cases in which a judge sentenced a defendant below the guideline range, she did so not because of her distaste or disregard for the Guidelines. Instead she sentenced outside of the guideline range because the prosecutor asked her to. In other words, everyone who is in court for the sentence hearing recognizes that, unless the government asks for a below-guideline sentence, the district judge is often going to sentence the defendant within the guideline range. Also important, it was in only 2.2 percent of all cases that a judge sentenced a defendant above the guideline range. Id. So while a person may not be able to rely on the Guidelines to predict whether he will receive a sentence even lower than the bottom of the guideline range, he can reliably use the Guidelines to predict the harshest sentence he'll get.

B.

This is how the Matchett panel got the Sentencing Guidelines wrong. But the panel got the vagueness doctrine wrong as well. It said "[t]he vagueness doctrine applies only to laws that prohibit conduct and fix punishments." 802 F.3d at 1189. Or as the Statement now says, the doctrine "applies only to laws that regulate the primary conduct of private citizens." Pryor Op. at 12. I say federal policy that causes certain conduct to be punished by more years in prison "prohibit[s] conduct and fix[es] punishments" and regulates "private citizens." But even if I am wrong on this, we know that the Supreme Court's application of the vagueness doctrine is not limited to things that "prohibit conduct and fix punishments."

For example, in Giaccio v. State of Pennsylvania, 382 U.S. 399, 86 S.Ct. 518, 15 L.Ed.2d 447 (1966), the Supreme Court held unconstitutionally vague a Pennsylvania law that allowed juries, after acquitting a defendant, to decide whether to charge that defendant the costs of prosecution. Specifically, the jury could impose costs if it found the defendant "guilty of some misconduct less than the offense which is charged but nevertheless misconduct of some kind as a result of which he should be required to pay some penalty[.]" Id. at 404, 86 S.Ct. 518. The statute said that, if a jury decides to impose costs, the trial judge "`shall forthwith pass sentence to that effect, and order him (defendant) to be committed to the jail of the county' there to remain until he either pays or gives security for the costs." Id. at 403, 86 S.Ct. 518. The Supreme Court recognized that the law was not a penal "statute which imposed forfeitures, punishments or judgments for costs." Id. at 404, 86 S.Ct. 518. Yet while the statute neither prohibited conduct nor fixed punishment, "[t]he Court did not hesitate in striking down the statute on vagueness grounds." Pac. Mut. Life Ins. Co. v. Haslip, 499 U.S. 1, 49, 111 S.Ct. 1032, 1059, 113 L.Ed.2d 1 (1991) (O'Connor, J., dissenting). The Statement argues that the Pennsylvania statute, unlike the Sentencing Guidelines, "imposed a penalty for out-of-court conduct by a private citizen" and thus "regulated primary conduct." Pryor Op. at 12-13. Not so. Rather, the law at issue in Giaccio is closely analogous to the Sentencing Guidelines. Contrary to Judge Pryor's description in the Statement, the statute did not "require[]" a jury to impose costs. Id. at 12, 111 S.Ct. 1032, 1059. Rather it authorized the jury to impose the "sentence" in its discretion. Giaccio, 382 U.S. at 401, 86 S.Ct. 518. It was, in effect, an advisory sentencing guideline. Giaccio thus makes clear that, while the vagueness doctrine might more commonly apply to laws that directly govern private conduct, the doctrine also applies to laws that govern judges' or juries' decisions about whether and how to punish private conduct. And this is true even where that decision-making is fully discretionary.11

The application of the vagueness doctrine to rules that guide discretionary sentencing goes to the very heart of the fairness interests that the doctrine is designed to protect. Johnson reminded us of two ways in which vague laws can violate the Fifth Amendment's guarantee of due process of law: by being "so vague that [the law] fails to give ordinary people fair notice of the conduct it punishes, or so standardless that it invites arbitrary enforcement." 135 S.Ct. at 2556. The Court then held that the 13 words of ACCA's residual clause were unlawful in both these ways: "We are convinced that the indeterminacy of the wide-ranging inquiry required by the residual clause both denies fair notice to defendants and invites arbitrary enforcement by judges." Id. at 2557.

The Matchett panel gave no heed to these admonitions against "arbitrary enforcement." Zero. Instead, the panel addressed only Johnson's "notice" rationale, without ever mentioning the Court's concern about "arbitrary enforcement by judges." This matters because we have been instructed that the "arbitrary enforcement" concern is "the more important aspect of vagueness doctrine." Kolendar v. Lawson, 461 U.S. 352, 358, 103 S.Ct. 1855, 1858, 75 L.Ed.2d 903 (1983). Perhaps reflecting this lesson, every time Johnson told us why the residual clause is not lawful, it underscored the problem that the vague language of the clause led different judges to give similarly situated defendants widely varying sentences.12 But again, the panel made no effort to address this concern about arbitrariness, which the Supreme Court told us is "the more important aspect of vagueness doctrine." Id.

If the panel had been willing to evaluate how the residual clause in the Sentencing Guidelines leads to "arbitrary enforcement by judges," then the case would have easily resolved in Mr. Matchett's favor.13 The Supreme Court has told us that overly vague laws violate our Constitution because they "delegate[] basic policy matters to policemen, judges, and juries for resolution on an ad hoc and subjective basis, with the attendant dangers of arbitrary and discriminatory application." Grayned v. City of Rockford, 408 U.S. 104, 108-09, 92 S.Ct. 2294, 2299, 33 L.Ed.2d 222 (1972). Or as Johnson put it, "the indeterminacy of the wide-ranging inquiry required by the residual clause" makes the clause unconstitutional because it "invites arbitrary enforcement by judges." 135 S.Ct. at 2557.

The risks of "discriminatory application" and "arbitrary enforcement" here should be obvious. Two judges who are sentencing defendants with identical records can arrive at different sentences based on each judge's personal sense of what seems like a crime of violence. Judges who must review sentences imposed under USSG § 4B1.2 will certainly try to apply pre-Johnson residual clause opinions correctly.14 But Justice Scalia once said that he worried that, in the end, judges will have little choice but to "simply throw the opinions into the air in frustration, and give free rein to their own feelings as to what offenses should be considered crimes of violence." Derby v. United States, 564 U.S. 1047, 131 S.Ct. 2858, 2859, 180 L.Ed.2d 904 (2011) (Scalia, J., dissenting from denial of certiorari).15 Of course we expect that judges will not act so ignobly. But "the due process protection against vague regulations does not leave the public at the mercy of noblesse oblige." FCC v. Fox Television Stations, Inc., ___ U.S. ___, 132 S.Ct. 2307, 2318, 183 L.Ed.2d 234 (2012) (quotation omitted). Instead it bans any regulation that is "so standardless that it authorizes or encourages seriously discriminatory enforcement." United States v. Williams, 553 U.S. 285, 304, 128 S.Ct. 1830, 1845, 170 L.Ed.2d 650 (2008).

III.

Also worrisome, the Matchett panel opinion forces this court to continue to apply and even add to the body of law that Johnson discredited when it reviews the sentences of individuals who were sentenced under § 4B1.2(a)(2)'s residual clause. The panel recognizes that in Johnson, the Court "abrogated the previous decisions of the Supreme Court interpreting the residual clause." 802 F.3d at 1195. But the panel nevertheless instructs courts that we "must still adhere to the reasoning of [these] cases" when interpreting § 4B1.2. Id. at 1195-96. Our court will thus continue to apply cases the Supreme Court derided as "anything but evenhanded, predictable, or consistent." Johnson, 135 S.Ct. at 2563. Indeed Johnson referenced several § 4B1.2 cases to illustrate how the residual clause "produces more unpredictability and arbitrariness than the Due Process Clause tolerates." Id. The Supreme Court even gave this court an unwanted tip of the hat, when it cited one of our § 4B1.2 opinions to declare that "[t]his Court is not the only one that has had trouble making sense of the residual clause." Id. at 2559-60 (quotation omitted) (citing United States v. Whitson, 597 F.3d 1218 (11th Cir. 2010)). By my reading, the Supreme Court treated the problem with the residual clause in the statute as identical to the problem with the residual clause in the Sentencing Guidelines.16

IV.

The panel warned that applying the vagueness doctrine to the residual clause in the Guidelines "would upend our sentencing regime" since "many [Guidelines] provisions could be described as vague." Id. at 1196. Now in his Statement, Judge Pryor expands the point. He describes an even more forbidding scene in which all sentencing guidelines, in both the federal and state systems, could be invalidated. Pryor Op. at 22. I am sure he does not mean to advocate abandoning our Constitution because the effect of enforcing it would be too disruptive. This approach would seem, as Justice Brennan put it, "to suggest a fear of too much justice." McCleskey v. Kemp, 481 U.S. 279, 339, 107 S.Ct. 1756, 1791, 95 L.Ed.2d 262 (1987) (Brennan, J., dissenting). And in any event, while I appreciate Judge Pryor's full-throated defense of the Sentencing Guidelines, the Guidelines are not under attack. Johnson did not invalidate any or all guidelines that "could be described as vague." Instead the Court singled out a far more distinct problem: laws that require judges to apply overly vague standards "to a judicially imagined `ordinary case' of a crime, not to real-world facts." 135 S.Ct. at 2557; see also id. at 2558 ("It is one thing to apply an imprecise "serious potential risk" standard to real-world facts; it is quite another to apply it to a judge-imagined abstraction.").

The Guidelines determine punishment based almost exclusively on a defendant's actual conduct.17 Consider the two provisions that the Matchett panel warned could next be deemed vague. See 802 F.3d at 1196. The first asks if the defendant used "sophisticated means" in committing the actual crime. USSG § 2B1.1(b)(10). The other asks if she played a "minor" role in the actual crime. Id. § 3B1.2(b). Johnson expressly condones standards that assess actual conduct in this way. See 135 S.Ct. at 2561 ("[W]e do not doubt the constitutionality of laws that call for the application of a qualitative standard such as `substantial risk' to real-world conduct."). What the panel describes as a slippery slope in fact contains a clear, well-defined "constitutional toehold." Walz v. Tax Comm'n of City of New York, 397 U.S. 664, 699-700, 90 S.Ct. 1409, 1427, 25 L.Ed.2d 697 (1970) (Harlan, J., concurring).

But even if applying Johnson to the identical residual clause in the Guidelines were to lead to a future holding that some other provision of the Guidelines is also void for vagueness, this possibility is no basis for refusing to uphold the Constitution here. The prospect that there may be other provisions of the Guidelines that are unconstitutionally vague "may be dismaying, but it does not justify complete abdication of our judicial role. The Constitution was framed fundamentally as a bulwark against governmental power, and preventing the arbitrary administration of punishment is a basic ideal of any society that purports to be governed by the rule of law." McCleskey, 481 U.S. at 339, 107 S.Ct. 1756 (Brennan, J., dissenting). Our court and the people of Alabama, Florida, and Georgia would have been well served by vacatur of the Matchett panel opinion. I dissent from the court's vote to leave it intact.

ROSENBAUM, Circuit Judge, joined by JILL PRYOR, Circuit Judge, dissenting from the denial of en banc rehearing:

My colleague Judge William Pryor takes issue with my concurrences in In re Hunt, 835 F.3d 1277, 1280-85, No. 16-14756-J, 2016 WL 3895246, at *4-7 (11th Cir. July 18, 2016) (Rosenbaum, J., concurring),1 and In re Clayton, 829 F.3d 1254, 1267-71, No. 16-14556-J, 2016 WL 3878156, at *10-13 (11th Cir. July 18, 2016) (Rosenbaum, J., concurring), which criticize our opinion in United States v. Matchett, 802 F.3d 1185 (11th Cir. 2015). Judge Pryor's opinion both expressly assails my Hunt/Clayton concurrences and more generally attacks my concurrences when his opinion refers to all of the Hunt/Clayton concurrences as a whole. I respectfully disagree with his points.

Judge Pryor's criticism of my specific Hunt/Clayton concurrences is not substantive in nature and is belied by what my concurrences actually say. As for my colleague's more general condemnation of all of the Hunt/Clayton concurrences' points about the arbitrary-enforcement problem that the career-offender guideline's residual clause creates, a close review of Judge Pryor's analysis reveals why the Hunt/Clayton concurrences have the better argument.

I.

Judge Pryor's opinion makes two complaints expressly about my particular Hunt/Clayton concurrences: (1) that I "fret" that the Sentencing Commission might issue a "nonsensical guideline about `cheese,'" see Pryor Op. at 22, and (2) that I have mistaken vagueness for unintelligibility.

I feel silly addressing Judge Pryor's first point. But since he relies on it to incorrectly insinuate that I attribute ill intentions to the Sentencing Commission,2 I think I must respond.

Surely Judge Pryor does not truly believe that the introductory paragraph of my Hunt/Clayton concurrences somehow suggests that we should worry that the Sentencing Commission might issue a "nonsensical guideline about `cheese.'"3 But just to reassure my colleague, I note that I do not now nor have I ever lived in fear that the Sentencing Commission might issue a "nonsensical guideline about `cheese.'"4

I used what I described as the "hypothetical" cheese guideline as an analogy. An analogy, of course, is a literary device that is a "comparison made between one thing and another for the purpose of explanation or clarification." Analogy, OXFORD ENGLISH DICTIONARY, Definition 3.b., http://www.oed.com/view/Entry/7030?redirectedFrom=analogy#eid (last visited Aug. 26, 2016). In this case, I employed the fictional cheese guideline to help explain the problem of trying to apply a guideline, like the residual clause of the career-offender guideline, that is essentially unintelligible.

That brings me to my second point. Judge Pryor offers a vocabulary lesson in the differences between "vagueness" and "unintelligibility," suggesting that although the career-offender guideline's residual clause is vague, it is not unintelligible. See Pryor Op. at 23.

But the problem with the career-offender guideline's residual clause is that it is so vague as to be essentially unintelligible. Indeed, that's exactly how Justice Scalia characterized the identical residual clause of the Armed Career Criminal Act ("ACCA") — as "unintelligible." See James v. United States, 550 U.S. 192, 230-31, 127 S.Ct. 1586, 167 L.Ed.2d 532 (2007) (Scalia, J., dissenting) (describing the residual clause of the ACCA as "an unintelligible criminal statute"), overruled by Johnson v. United States, ___ U.S. ___, 135 S.Ct. 2551, 2563, 192 L.Ed.2d 569 (2015); see also Sykes v. United States, 564 U.S. 1, 33, 131 S.Ct. 2267, 180 L.Ed.2d 60 (2011) (Scalia, J., dissenting) (same). I think we can safely assume that Justice Scalia understood the meanings of "vague" and "unintelligible." And his viewpoint that the identically phrased ACCA residual clause is "unintelligible" ultimately prevailed in Johnson.

II.

I now turn to Judge Pryor's more general criticism of all of the Hunt/Clayton concurrences' point that the residual clause of the career-offender guideline is unconstitutionally vague because it allows for arbitrary enforcement. First, Judge Pryor embarks on a half-hearted defense of the idea that arbitrary enforcement of the career-offender guideline's residual clause may be avoided. Then, in apparent recognition of the problems with this position, he focuses his fire power on the idea that even if arbitrary enforcement is a problem, it is not a cognizable concern when it comes to the residual clause of the career-offender guideline. I respectfully disagree.

A.

For the reasons described in my Hunt/Clayton concurrences, arbitrary enforcement of the residual clause of the career-offender guideline represents a serious problem. In response to these concerns, Judge Pryor wishfully offers that "judges who must apply the residual clause of the career-offender guideline are not hopelessly adrift." Pryor Op. at 20. But the support for my colleague's conclusion rests on quotations from Johnson which, when read in context, suggest the opposite. Judge Pryor explains,

Pryor Op. at 20-21.

The actual sections from which Judge Pryor selectively picked his quotations leave a very different impression. They suggest that to the extent that a universe of "obviously risky crimes" exists, it is exceedingly small:

Johnson, 135 S.Ct. at 2560-61 (bold emphasis added) (regular italics original) (citations omitted). So, at least in the Supreme Court's view, we cannot take comfort in the idea that some significant number of crimes "clearly" fall within the bounds of the residual clause.

Nor, for this same reason, is Judge Pryor's proposal that pre-Johnson caselaw "guide judges" in construing the residual clause satisfactory. See Pryor Op. at 21. Judge Pryor notes that "federal circuit and district judges interpreted [the residual clause] thousands of times before Johnson[,] [and] [t]he Supreme Court did so four times as well." Id. He suggests that this fact means that nothing is wrong with continuing to construe the residual clause like it was interpreted before the Supreme Court issued Johnson.

This argument buries its head in the pre-Johnson landscape's sand. After Johnson, that argument is stranded in a legal desert, devoid of usable caselaw. As the Supreme Court explained in Johnson, it could revisit its earlier residual-clause decisions under stare decisis only because "experience with [the] application [of those earlier decisions] reveal[ed] that [they were] unworkable." Johnson, 135 S.Ct. at 2562. If the Supreme Court found its own decisions and other pre-Johnson caselaw "unworkable," it requires no leap of logic to conclude that the lower courts also should not rely on that body of law.

Judge Pryor seems to implicitly recognize this problem, proposing two solutions: first, he advises sentencing judges that they can just "vary from [the career-offender guideline] on policy grounds." Pryor Op. at 22. And second, he states that "the Sentencing Commission can and does repeal ["bad"] guidelines that are difficult to apply. In fact, they did so here." Pryor Op. at 21 (noting the repeal of the residual clause of the career-offender guideline, as of August 1, 2016).

But disregarding the effect of the career-offender guideline's residual clause on policy grounds is no answer to the judge who has no policy qualm with the aim behind the career-offender guideline — that violent recidivists should receive substantially longer sentences than other defendants — and just wants to be able to ascertain properly whether, in a given case, the guideline should apply. What is that judge to do?

Plus, in view of Judge Pryor's second solution to the problems raised by the residual clause of the career-offender guideline — that the Sentencing Commission repealed it as of August 1, 2016 — my colleague's advice is the legal equivalent of closing the stable door after the horse has bolted. No one reading his opinion now will be required to decide whether and how to apply the career-offender guideline's residual clause at sentencing. And I am sure that Judge Pryor does not mean to suggest that appellate courts can decide to set aside a district court's application of the career-offender guideline's residual clause solely because of an appellate court's policy disagreement with the guideline under which a defendant was sentenced.

Nor are Judge Pryor's responses any answer to arbitrary enforcement in the many cases where sentencings occurred before the residual clause of the career-offender guideline was revoked. Indeed, although the Supreme Court issued Johnson on June 26, 2015, the residual clause remained in effect — and arbitrary enforcement continued to occur — for more than a year after that, until August 1, 2016. And it's too late now for district courts that sentenced defendants during that period to take my colleague's advice to vary downward from the Guidelines range based on a policy disagreement with the residual clause, even if they wanted to do so.5

B.

With no satisfactory answer to the problem of arbitrary enforcement, Judge Pryor turns his attention to arguing that arbitrary enforcement is not a cognizable reason to invalidate a guideline. He invokes two major reasons why: (1) in Judge Pryor's view, the vagueness doctrine applies to only laws that "regulate the primary conduct of private citizens," Pryor Op. at 3, and the residual clause of the career-offender guideline does not fall into that category; and (2) "judges [do not have a due-process right] to be free from interpreting vague laws in the exercise of judicial duty," id. at 18 (emphasis in original). Judge Pryor's first contention is not borne out by the caselaw. And his second proceeds from an incorrect premise: of course, the Hunt/Clayton concurrers do not believe that judges have a due-process right to be free from interpreting vague laws. We do, however, believe that the public has a due-process right to be free from the necessarily arbitrary enforcement that judges must engage in when they must apply unconstitutionally vague laws.

1.

The central theme of Judge Pryor's argument is that the vagueness doctrine — including its concern for avoiding arbitrary enforcement — applies to only laws that "regulate the primary conduct of private citizens," and the Sentencing Guidelines do not fall into that category. Pryor Op. at 3. This "rule" — that the vagueness doctrine applies to only laws that "regulate the primary conduct of private citizens" — is a creation of Judge Pryor, based on his characterization of cases where the vagueness doctrine has been applied. The Supreme Court has not been so narrow in describing the types of laws to which the vagueness doctrine may apply. In fact, Supreme Court jurisprudence applying the vagueness doctrine defies my colleague's characterization of that body of caselaw.

First, the Supreme Court has never described the vagueness doctrine as applying solely to laws that regulate the primary conduct of private individuals. In Johnson, for example, the Supreme Court observed broadly that the vagueness doctrine invalidates a criminal law that is "so vague that it fails to give ordinary people fair notice of the conduct it punishes, or so standardless that it invites arbitrary enforcement." 135 S.Ct. at 2556 (emphasis added). Nothing more is required. So even if all existing Supreme Court jurisprudence on the vagueness doctrine could accurately be characterized as involving laws that regulate the primary conduct of private individuals — it can't (more on this later) — a law that does not regulate the primary conduct of private individuals could still be unconstitutionally vague if it were "so standardless that it invite[d] arbitrary enforcement."

Second, in supporting his thesis that the vagueness doctrine applies to only laws that regulate the primary conduct of private citizens, Judge Pryor incorrectly characterizes Giaccio v. Pennsylvania, 382 U.S. 399, 86 S.Ct. 518, 15 L.Ed.2d 447 (1966) — a decision on which Judge Martin relied in her Clayton concurrence — as a case where the Supreme Court invalidated a law that regulated the primary conduct of private citizens. Judge Pryor describes the law at issue in Giaccio as allowing "[a] jury [to] award the costs [of an acquitted defendant's prosecution] as a `sentence' if it found that the defendant was `guilty of some misconduct less than the offense which is charged but nevertheless misconduct of some kind as a result of which he should be required to pay some penalty short of conviction.'" Pryor Op. at 14 (quoting Giaccio, 382 U.S. at 403-04, 86 S.Ct. 518). But that description matches only part of what the Supreme Court held unconstitutionally vague in Giaccio: the judge's jury instructions.

Significantly, the Supreme Court also ruled unconstitutionally vague the actual law itself that was at issue in Giaccio — and that law did not even arguably regulate primary conduct. See Giaccio, 382 U.S. at 402-03, 86 S.Ct. 518. The law that the Supreme Court invalidated in Giaccio provided, in relevant part, only that "in all cases of acquittals by the petit jury on indictments for (offenses other than felonies), the jury trying the same shall determine, by their verdict, whether the county, or the prosecutor, or the defendant shall pay the costs...." Id. at 400-01, 86 S.Ct. 518. It required no assessment of fault of any type on the part of the acquitted defendant.

The Supreme Court specifically invalidated that law as written — and without regard to how the jury was instructed in Giaccio's case6 — as unconstitutionally vague:

Id. at 403, 86 S.Ct. 518. Judge Pryor concluded that the law at issue in Giaccio "plainly regulated primary conduct" because "it imposed a penalty (costs, and possibly jail time) on an acquitted defendant (a private individual) based on his out-of-court conduct (the misconduct that led to his prosecution)." Pryor Op. at 14-15.

That description of the Giaccio law is simply incorrect. The law allowed the imposition of costs on an acquitted defendant, even if the jury concluded that the defendant had engaged in no misconduct that led to his prosecution. A law that imposes a penalty on a private individual, without respect to his out-of-court conduct, cannot qualify as a law that regulates the primary conduct of private individuals. And since the vagueness doctrine was held applicable to such a law in Giaccio, Judge Pryor's thesis that the vagueness doctrine applies to only laws regulating primary conduct of private individuals cannot be correct.

2.

Judge Pryor also seems to think that the fact that judges do not enjoy a due-process right "to be free from interpreting vague laws in the exercise of their judicial duty" can somehow excuse a vague law like the residual clause of the career-offender guideline from constitutional compliance. See Pryor Op. at 20. He argues that "the advisory guidelines are directed to judges, not private citizens, and we tolerate much more vagueness in laws that regulate government actors than we do in laws that regulate private citizens." Id. at 19. In support of his position, Judge Pryor relies on Mahler v. Eby, 264 U.S. 32, 40-41, 44 S.Ct. 283, 68 L.Ed. 549 (1924), and National Endowment for the Arts v. Finley, 524 U.S. 569, 118 S.Ct. 2168, 141 L.Ed.2d 500 (1998) ("NEA").

Judge Pryor's argument suffers from at least two problems. First, determination of whether the career-offender guideline's residual clause applies to any given prior conviction is not discretionary. Nor is the requirement that every sentencing court begin the sentencing process by correctly calculating the defendant's applicable Guidelines range. So Mahler and NEA, which involve vagueness challenges to discretionary provisions, are not instructive. And second, Judge Pryor's argument fails to recognize that, regardless of to whom the Guidelines and the ACCA are addressed, the residual clause of each requires judges to engage in the same analysis, meaning that the arbitrary-enforcement problems that plagued ACCA's residual clause occur with equal force in the context of the career-offender guideline's residual clause.7 I address each problem below in more detail.

First, Mahler and NEA are not relevant to the question of whether the vagueness doctrine applies to the residual clause of the career-offender guideline. Significantly, correct determination of whether the residual clause of the career-offender guideline applies to a defendant should not involve discretion of any kind. That's because the residual clause is applied categorically, meaning that the sentencing court may not consider unique factual circumstances when determining whether the residual clause applies to a given prior conviction. If a specific crime qualifies as a violent crime under the residual clause, it qualifies for all defendants, regardless of the individual factual circumstances of a particular defendant's prior conviction for that crime. So, if the residual clause were susceptible of application in the way it was intended, a district judge's determination of whether the residual clause of the career-offender guideline applies in a specific case would be either objectively right or wrong. There would be no room for a range of correct answers.

That is simply not the case with the discretionary laws at issue in the cases Judge Pryor cites. In Mahler, for example, the Supreme Court rejected a vagueness challenge to the Secretary of Labor's statutory power to expel aliens, observing that the power was discretionary. 264 U.S. at 40-41, 44 S.Ct. 283.

Similarly, the law at issue in NEA, 524 U.S. 569, 118 S.Ct. 2168, required the Chairperson of the National Endowment for the Arts to ensure that "artistic excellence and artistic merit are the criteria by which [grant] applications are judged, taking into consideration general standards of decency and respect for the diverse beliefs and values of the American public." Id. at 572, 118 S.Ct. 2168 (quoting 20 U.S.C. § 954(d)(1)). The Supreme Court described this law as "vest[ing] the NEA with substantial discretion to award grants [by] identif[ying] only the broadest funding priorities...." Id. at 573, 118 S.Ct. 2168.

The laws at issue in Mahler and NEA, of course, are nothing like the residual clause of the career-offender guideline, the application of which, I have noted, was intended to result in a single objectively correct answer to the question of whether a prior conviction qualified as a violent crime. Because Mahler and NEA involved laws that did not purport to establish standards whose application results in a single correct answer, they are not helpful in appraising the residual clause of the career-offender guideline for vagueness.

Second, Judge Pryor's argument does not account for the fact that the residual clauses of both the career-offender guideline and the ACCA require judges to engage in the same analysis. Judge Pryor describes the Guidelines as "directed to judges, not private citizens," Pryor Op. at 19, suggesting that the vagueness in the residual clause of the career-offender guideline is somehow more tolerable than the exact same vagueness in the ACCA. See id.

But regardless of at whom the Sentencing Guidelines are directed, judges engage in the very same analysis when they apply the career-offender guideline's residual clause as they did when they applied the ACCA's residual clause. In both cases, judges must construe the same thirteen words, and in both cases, judges must decide whether a prior conviction categorically — not individually with respect to the details of a given defendant's prior crime — qualifies as a violent crime. That one definition appears in a guideline while the other is in a statute does not, as a practical matter, affect the way in which the courts go about analyzing whether the provision applies. And if the provision is too vague to avoid arbitrary enforcement under the ACCA, it is equally too vague to avoid arbitrary enforcement under the career-offender guideline.

III.

Finally, in a last-ditch effort to support his view that the vagueness doctrine should not apply to the residual clause of the career-offender guideline, Judge Pryor trots out the old parade of horribles. He predicts that most of the Sentencing Guidelines and the states' sentencing guidelines will necessarily be invalidated if the vagueness doctrine can apply to the residual clause of the career-offender guideline, and down the road, the Sentencing Guidelines themselves, as a whole, will be ruled an unconstitutional violation of the Separation of Powers. Pryor Op. at 23-24.

But ruling that the residual clause of the career-offender guideline is unconstitutionally vague would not mean the end of the sentencing world as we know it. Unlike the residual clause of the career-offender guideline, most guidelines are not intended to be categorically applicable. They are designed instead to, as Judge Pryor has explained, "strike a ... balance between consistency, predictability, and flexibility." Pryor Op. at 25. As a result, most guidelines are supposed to and do allow the judge some discretion in determining the relevant facts of a particular defendant's case in applying the guideline. In fact, at times, correct application of a single guideline can result in different, correct answers. Again, that's by design.

But the residual clause of the career-offender guideline is different. It is not intended to apply flexibly, depending on a particular defendant's unique factual circumstances. Because it is meant to be applied categorically, it is supposed to yield an objectively correct answer about whether any particular crime qualifies under it as a violent crime, regardless of a defendant's individual circumstances and a sentencing judge's view of the record. That it cannot be applied in this way, despite the intended design of the guideline, is the source of the vagueness problem with the residual clause of the career-offender guideline. Indeed, the Supreme Court held ACCA's identical residual clause unconstitutionally vague, in significant part, because of the law's failed categorical nature. See Welch v. United States, ___ U.S. ___, 136 S.Ct. 1257, 1262, 194 L.Ed.2d 387 (2016).

Allowing a successful vagueness challenge to a guideline that is intended to apply categorically but is incapable of actually working that way does not set a precedent for holding guidelines that are designed to apply with some flexibility and discretion — as opposed to categorically — vague. As I have previously noted, non-categorical guidelines "guide judicial discretion," Pryor Op. at 3, in a way that is substantively and materially different than categorical guidelines do. Whereas categorical guidelines should yield but a single correct answer applicable to every case without regard to an individual defendant's circumstances or a judge's view of the record, non-categorical guidelines should not. So an inability to categorically apply a guideline that is intended to be categorically applied necessarily results in arbitrary enforcement every time it is applied. As a result, it is vague in a way that a non-categorical guideline never can be.

Similarly, invalidating the career-offender guideline's residual clause as unconstitutionally vague has no implications for 18 U.S.C. § 3553(a) standards, see Pryor Op. at 19-20, since they also do not apply categorically. Rather, by design, judges consider the individual circumstances of a given defendant's case and are expected to exercise their discretion in applying the § 3553(a) standards. See United States v. Hurlburt, No. 14-3611, 2016 WL 4506717, at *7 (7th Cir. Aug. 29, 2016) (en banc) ("Johnson itself specially addressed this kind of objection [that holding the residual clause unconstitutional will open the floodgates to vagueness challenges to other sentencing provisions] and rejected it. The Court explained at length that the vagueness defect in the ACCA's residual clause is not just its use of indeterminate language; it's that the clause uses indeterminate language and must be applied categorically, without regard to real-world facts.").

Nor does declaring the residual clause of the career-offender guideline unconstitutionally vague have any implications for the constitutionality of the Sentencing Guidelines under the Separation of Powers. The reason that the residual clause of the career-offender guideline must be held unconstitutionally vague is not that the freestanding guideline results by itself in arbitrary enforcement — a circumstance that might cause Separation-of-Powers problems if it existed; it is instead because under 18 U.S.C. § 3553(a), "district courts must begin their analysis with the Guidelines and remain cognizant of them throughout the sentencing process." Gall v. United States, 552 U.S. 38, 50 & n.6, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007) (emphasis added).

As the Supreme Court has recognized, by congressional mandate in the form of § 3553(a), the Sentencing Guidelines "anchor... the district court's discretion."8 Molina-Martinez v. United States, ___ U.S. ___, 136 S.Ct. 1338, 1345, 194 L.Ed.2d 444 (2016). And it is that same congressional mandate of § 3553(a) that causes a district court's improper calculation of a defendant's Guidelines range to constitute a "significant procedural error," id. at 1346 (quoting Gall, 552 U.S. at 51, 128 S.Ct. 586) (quotation marks omitted) — so "particularly serious," id. that the error generally qualifies in its own right as having "affected the defendant's substantial rights." Id. Because the unconstitutional vagueness of the residual clause of the career-offender guideline results from the congressional mandate in § 3553(a) that requires courts to first correctly calculate the Guidelines range — not from any force that the freestanding career-offender guideline has in and of itself, allowing the residual clause of the career-offender guideline to be stricken as unconstitutionally vague does not implicate the Separation of Powers.

IV.

These issues are important ones. They potentially impact numerous defendants. So I would have granted en banc rehearing when the poll was initially taken in March 2016.

APPENDIX

ROSENBAUM, Circuit Judge, concurring, joined by WILSON and JILL PRYOR, Circuit Judges:

Imagine a sentencing guideline that read, "A defendant is a career offender if `[p]uddles do not ask for why not? It is cheese! Breath and wind. It is cheese.'" Boston Legal, "Word Salad Days" (2006), http://www.imdb.com/title/tt0770843/quotes (last visited Apr. 28, 2016). Now imagine that based on the Guidelines range that that indecipherable language required, a district court sentenced a defendant to twice as much time as it otherwise would have. How could the sentencing court know that the guideline applied? How could the reviewing court know that the correct Guidelines calculation included an enhancement under that guideline? Surely doubling a defendant's sentence based on nonsense would violate due process. But in United States v. Matchett, 802 F.3d 1185 (11th Cir. 2015), we allowed defendants to continue to be sentenced to much more severe sentences than they would otherwise receive, based on the residual clause of the career-offender guideline, a guideline that the Supreme Court has found hardly more scrutable than the hypothetical one above.

No doubt criminal defendants do not have a due-process right to a sentence within a particular Sentencing Guidelines range. But Congress can, and essentially has, required courts to begin the sentencing process by correctly calculating the Guidelines range. The question here is whether, when the Supreme Court strikes language from a statute because it is unconstitutionally vague language and that same language also appears in a guideline, we are constitutionally able to continue to apply that language in the sentencing process that Congress has mandated. The answer, unlike the challenged part of the career-offender guideline, is clear: we are not.

I concur in Sections I.A. and II of Judge Wilson's well-reasoned concurrence. I agree that the Supreme Court's decision in Johnson v. United States, 576 U.S. ___, 135 S.Ct. 2551, 192 L.Ed.2d 569 (2015), holding the Armed Career Criminal Act's ("ACCA") residual clause unconstitutionally vague renders the exact same language in the Sentencing Guidelines unconstitutional as well. So while we are bound by Matchett in deciding Hunt's Guidelines claim, I write separately to explain why I believe that Matchett was incorrectly decided.

I.

In Matchett, 802 F.3d 1185, the panel reached the opposite conclusion because it held that the vagueness doctrine does not apply to the Sentencing Guidelines. 802 F.3d at 1193-95. To reach that result, the panel first described the vagueness doctrine as "rest[ing] on [a] lack of notice." Id. at 1194 (quoting Maynard v. Cartwright, 486 U.S. 356, 361, 108 S.Ct. 1853, 1857, 100 L.Ed.2d 372 (1988)). Then, the panel construed Irizarry v. United States, 553 U.S. 708, 128 S.Ct. 2198, 171 L.Ed.2d 28 (2008), as precluding due-process challenges to, essentially, anything having to do with sentencing under the Guidelines, based on the Supreme Court's remark that that "[a]ny expectation subject to due process protection... that a criminal defendant would receive a sentence within the presumptively applicable Guidelines range did not survive our decision in United States v. Booker, 543 U.S. 220, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005), which invalidated the mandatory features of the Guidelines." Matchett, 802 F.3d at 1194 (quoting Irizarry, 553 U.S. at 713, 128 S.Ct. at 2202). Finally, the panel quoted the Eighth Circuit's decision in United States v. Wivell, 893 F.2d 156, 160 (8th Cir. 1990), for the proposition that "[b]ecause there is no constitutional right to sentencing guidelines... the limitations the Guidelines place on a judge's discretion cannot violate a defendant's right to due process by reason of being vague." Matchett, 802 F.3d at 1194-95 (quoting Wivell, 893 F.2d at 160).

A.

The problem with the first part of the panel's analysis — that the vagueness doctrine "rest[s] on [a] lack of notice" — is that it is incomplete. The vagueness doctrine also protects against arbitrary enforcement by judges. Indeed, in Johnson itself the Supreme Court held that the ACCA equivalent of the 13 words at issue here violated due process because it "both denies fair notice to defendants and invites arbitrary enforcement by judges." Johnson, 135 S.Ct. at 2557 (emphasis added).

B.

As for the second part of the panel's analysis — that Irizarry precludes due-process challenges to all forms of sentencing error under the Guidelines — I respectfully disagree. In Irizarry, under the advisory Guidelines, a defendant was sentenced above the correctly calculated Guidelines range. 553 U.S. at 712, 128 S.Ct. at 2201. He asserted that his due-process rights had been violated because the sentencing court varied upwards from the Guidelines range without providing him with prior notice. See id. The Supreme Court rejected his argument, explaining that under the advisory Guidelines, "neither the Government nor the defendant may place the same degree of reliance on the type of `expectancy' [of a given sentence] that gave rise to a special need for notice [when the Guidelines were mandatory and the sentencing court departed from them]." Id. 553 U.S. at 713-14, 128 S.Ct. at 2202.

Put simply, Irizarry stands for only the proposition that a defendant has no due-process interest in receiving a sentence within the Guidelines range. But Irizarry says nothing about whether a defendant has a due-process right to a correct and fair sentencing process under the Sentencing Guidelines. And while the Supreme Court has not expressly spoken to such a right, the Court's recent decisions strongly indicate that the right exists.

For starters, in Molina-Martinez v. United States, 578 U.S. ___, 136 S.Ct. 1338, 194 L.Ed.2d 444 (2016), the Supreme Court recently explained that a district court that "improperly calculat[es]" a defendant's Guidelines range makes a "significant procedural error," id. at 1346 (quoting Gall v. United States, 552 U.S. 38, 51, 128 S.Ct. 586, 597, 169 L.Ed.2d 445 (2007)) (quotation marks omitted) — so "particularly serious," id. in fact, that the error generally qualifies in its own right as having "affected the defendant's substantial rights." Id.1

And that is not surprising, given that the Supreme Court has established that a correct and fair sentencing process necessarily begins with the correct calculation of the Guidelines range. Gall v. United States, 552 U.S. 38, 50 & n.6, 128 S.Ct. 586, 596 & n.6, 169 L.Ed.2d 445 (2007). Indeed, the Supreme Court has instructed that under 18 U.S.C. § 3553(a), "district courts must begin their analysis with the Guidelines and remain cognizant of them throughout the sentencing process." Id. at 50 n.6, 128 S.Ct. at 596 n.6 (emphasis added); see Peugh v. United States, 569 U.S. ___, 133 S.Ct. 2072, 2083, 186 L.Ed.2d 84 (2013) (same). The correct Guidelines calculation "anchor[s] both the district court's [sentencing] discretion and the appellate review process." Peugh, 133 S.Ct. at 2087.

In other words, the Supreme Court has acknowledged that Congress has effectively legislated the requirement that a sentencing court start the sentencing process by first correctly calculating the Guidelines range. That makes § 3553(a) a "statute[ specifying the procedure for] fixing sentences." See Johnson, 135 S.Ct. at 2556-57 (holding that the vagueness doctrine applies to statutes fixing sentences). So to the extent that, as a part of the statutorily mandated sentencing process, § 3553(a) requires courts in calculating the Guidelines range to use a guideline that is "so standardless that it invites arbitrary enforcement," the guideline must be struck down. See id. Failure to do so would render the sentencing process that § 3553(a) requires — determining the correct calculation of the Guidelines range — violative of due process because no court could reliably ascertain the correct calculation of the Guidelines range.

That is exactly the problem that the challenged language of the career-offender guideline presents. How can a sentencing court correctly calculate the Guidelines range when it is forced to apply the "hopeless[ly] indetermina[te]" language of the career-offender guideline? Johnson, 135 S.Ct. at 2448. Courts had "trouble making sense" of the very same words when they tried to apply them under the ACCA's residual clause. Id. at 2559-60. The Supreme Court observed that "[n]ine years' experience trying to derive meaning from the residual clause convince[d it] that [it] ha[d] embarked upon a failed enterprise." Id. at 2560. This "`black hole of confusion and uncertainty' that frustrates any effort to impart `some sense of order and direction,'" id. at 2562 (quoting United States v. Vann, 660 F.3d 771, 787 (4th Cir. 2011) (Agee, J., concurring)), does not somehow magically become clearer or more meaningful because the words appear in the guideline, rather than in the ACCA.

Because of this muddle, a sentencing court cannot ascertain whether the challenged part of the career-offender guideline even applies when the guideline is raised, so the court necessarily cannot correctly calculate the Sentencing Guidelines range. As a result, the sentencing court cannot comply with the sentencing process's virtual statutory requirement that the sentencing court first correctly calculate the applicable Guidelines range.

And, as Judge Wilson notes, the confusion only grows on appeal. Determining whether a sentence imposed by a district court was procedurally reasonable requires appellate courts to first ascertain whether the district court correctly calculated the applicable Guideline range. But we are no more skilled in applying "hopeless[ly] indetermina[te]" language than district courts.

C.

Finally, with regard to the third part of the Matchett panel's analysis — that the Sentencing Guidelines cannot be challenged as vague because no constitutional right to sentencing guidelines exists — I again respectfully disagree. True, "legislatures remain free to decide how much discretion in sentencing should be reposed in the judge or jury in noncapital cases." Lockett v. Ohio, 438 U.S. 586, 603, 98 S.Ct. 2954, 2964, 57 L.Ed.2d 973 (1978). But legislatures cannot, as Matchett would apparently hold, cabin the sentencing discretion of judges by mandating that they calculate a defendant's sentence using unconstitutionally vague language.

This would be another case entirely if sentencing judges could choose to wholly disregard the unconstitutionally vague career-offender guideline in calculating sentences. They cannot. Instead, district courts must begin the sentencing process by correctly calculating a defendant's Guidelines range. Peugh, 133 S.Ct. at 2083. Under Matchett, that means that Congress has essentially required district courts to apply unconstitutionally vague language in sentencing defendants. But it could not do that. Due process may not require sentencing guidelines, but it does prohibit Congress from requiring judges to apply unconstitutionally vague language in correctly calculating a defendant's sentence under any guidelines it chooses to enact.

II.

At bottom, statutorily, courts are required to begin every sentencing by correctly calculating the Guidelines range. Yet the Supreme Court has recognized that courts cannot reliably know whether the challenged language of the career-offender Guideline applies in any given case. As a result, they cannot possibly know whether a correct calculation of the Guidelines range should or should not include such an enhancement. But in Matchett, we nonetheless required sentencing courts to impose the enhancement and ourselves to uphold it, anyway. Trying to divine meaning from the word salad that is the challenged portion of the career-offender guideline guarantees an arbitrary and unfair sentencing process in violation of due process. For this reason, I respectfully disagree with our holding in Matchett.