SEYMOUR, Circuit Judge.
Steve Alan Daniels requests leave to file an application for a writ of habeas corpus pursuant to 28 U.S.C. § 2255, arguing the Supreme Court’s recent decision in Apprendi v. New Jersey, 530 U.S. 466, 120 S.Ct. 2348, 147 L.Ed.2d 435 (2000), renders his sentence unconstitutional. Because Mr. Daniels has filed previous applications for habeas relief, we construed his motion as a request to file a second or successive application. We initially consolidated his pro se application with that of Edwin Browning, another prisoner seeking permission to file a second or successive habe-as application based upon Apprendi. We appointed joint counsel and heard argument on both applications en banc. We then separated Mr. Browning’s application for disposition, holding in that case that Apprendi’s new rule has not yet been “made retroactive to cases on collateral review by the Supreme Court,” as required by section 2255, and thus may not be used as a basis for second or successive habeas applications. Browning v. United States, 241 F.3d 1262 (10th Cir.2001).
That holding does not entirely dispose of Mr. Daniels’ case, however. Because this is his first habeas application following the amendment of section 2255 by the Antiter-rorism and Effective Death Penalty Act of 1996 (AEDPA), Pub.L. No. 104-132, 110 Stat. 1214 (1996), he raises an additional question for our consideration: whether, assuming he can meet the pre-AEDPA requirements for raising an Apprendi claim in a successive petition under section 2255, applying AEDPA’s standards to his application is impermissibly retroactive. After a detailed consideration of the requirements for raising new claims based on new rules of criminal procedure pre- and post-AEDPA, we conclude that no impermissible retroactive result will arise [1185]*1185from the application of AEDPA’s rules to Mr. Daniels’ request.
I
BACKGROUND
Steve Daniels was convicted in 1988 on several counts of racketeering and heroin distribution, receiving a twenty-one-year sentence for the drug charges and concurrent sentences of five to twenty years for the other crimes. He appealed, asserting, inter alia, that his sentence on the drug distribution charges exceeded the period authorized by the statute under which he was indicted. After this court upheld his conviction and sentence on appeal, United States v. Ware, 897 F.2d 1538 (10th Cir.1990), Mr. Daniels filed two habeas petitions disputing other aspects of his sentence. Each petition was denied. See United States v. Daniels, No. 91-6333, 1992 WL 97997 (10th Cir. May 6, 1992); Daniels v. United States, No. 94-6289, 1995 WL 139398 (10th Cir. Mar.31, 1995).
In seeking permission to file this latest habeas petition, Mr. Daniels contends his conviction and sentence are unconstitutional under Apprendi, which establishes that “any fact that increases the penalty for a crime beyond the prescribed statutory maximum must be submitted to a jury, and proved beyond a reasonable doubt.” 120 S.Ct. at 2362-63. Although Apprendi did not explicitly include indictment practices in this rule, it suggested that the relevant facts must be included in an indictment as well. See id. at 2356-57. We so held in United States v. Jones, 235 F.3d 1231, 1236 (10th Cir.2000) (holding drug quantity must be included in indictment after Apprendi). Accord United States v. Keith, 230 F.3d 784, 786-87 (5th Cir.2000) (same); United States v. Aguayo-Delgado, 220 F.3d 926, 933 (8th Cir.2000) (same).
The amount of drugs underlying the charges against Mr. Daniels was not stated in his indictment nor presented to the jury for determination as required after Ap-prendi. He was, however, indicted for and convicted of distributing an identifiable quantity of heroin, which is prohibited under 21 U.S.C. § 841(b)(1)(C). This statutory subsection sets forth a maximum sentence of twenty years, one year shorter than the twenty-one-year sentence Mr. Daniels received. Thus, if Apprendi were applied to his case on collateral review, Mr. Daniels would be resentenced within the lower twenty-year maximum on the drug charge. See Jones, 235 F.3d at 1237 (conviction for drug offense proper despite Apprendi violation, but resentencing required because sentence exceeded maximum of section 841(b)(1)(C)).
Under AEDPA, the rule of Apprendi is not applicable to second or successive habeas motions unless and until it has been “made applicable to cases on collateral review by the Supreme Court,” as required by 28 U.S.C. § 2255. In Browning, we interpreted the phrase “made applicable ... by the Supreme Court” to require a specific collateral application by the Court, or words to that effect, and noted the Court has been silent with regard to collateral application of Apprendi. 241 F.3d at 1265-66. Under current habeas law, therefore, applications to file second or successive habeas petitions based on Apprendi will be dismissed until such time as the Supreme Court chooses specifically to declare the new rule applicable to cases on collateral review. Id. at 1266-67.
If AEDPA’s requirements are applied to Mr. Daniels’ application, his petition to assert an Apprendi claim must be denied. He argues, however, that because this is his first section 2255 habeas application after AEDPA’s enactment, and because he could meet the pre-AEDPA standard for raising an Apprendi claim in a section 2255 second or successive habeas petition, ap[1186]*1186plying AEDPA’s new standard to deny this application would render the standard impermissibly retroactive.
Mr. Daniels filed his prior section 2255 petitions in 1991 and 1993, before section 2255 was amended by AEDPA in 1996. According to Mr. Daniels, when those earlier petitions were filed a prisoner could present a new legal claim in a second or successive petition under section 2255 if he could show both sufficient “cause” for his failure to raise the claim, i.e., “that the factual or legal basis for [the] claim was not reasonably available” earlier, and “actual prejudice” as a result of the claimed legal error. McCleskey v. Zant, 499 U.S. 467, 494, 111 S.Ct. 1454, 113 L.Ed.2d 517 (1991). Mr. Daniels contends that he can meet this “cause and prejudice” test. In contrast, the new standard imposed by aAEDPA requires reliance on a “new rule of constitutional law, made retroactive to cases on collateral review by the Supreme Court, that was previously unavailable.” 28 U.S.C. § 2255 at para. 8. As discussed in Browning, applications seeking to raise an Apprendi claim must be denied under the new standard. Mr. Daniels therefore contends that applying the new standard will work a retroactive change which takes away his pre-AEDPA right to file a subsequent habeas petition based on Apprendi. As a result, he argues, AEDPA’s section 2255 amendments should not be applied to his case.
The government, on the other hand, asserts that applying the AEDPA standard will have no actual retroactive effect because Mr. Daniels could not have satisfied the pre-AEDPA cause and prejudice test to excuse his earlier failure to raise the claim. The government further contends AEDPA’s changes were procedural rather than substantive and thus constitute an exception to the usual retroactivity analysis. Finally, the government argues that, even assuming AEDPA altered the substantive standards under which we consider Mr. Daniels’ application, there can be no impermissible retroactive effect because Mr. Daniels did not reasonably rely on the continued validity of the old law at the time he filed his initial habeas petition.
We hold, first, that AEDPA’s purely procedural “gatekeeping” requirements are applicable to all petitioners seeking to file post-AEDPA second or successive ha-beas applications, regardless of when their initial habeas petitions were filed. We then assess the substantive requirements for raising claims based on new rules of criminal procedure before and after AED-PA. Ultimately, we conclude there is no retroactive effect in applying post-AEDPA substantive standards to Mr. Daniels.
II
POST-AEDPA PROCEDURAL REQUIREMENTS
The AEDPA amendments changed both procedural and substantive aspects of federal habeas law. See In re Minarik, 166 F.3d 591, 599-600 (3rd Cir.1999) (considering procedural and substantive retroactivity separately); Pratt v. United States, 129 F.3d 54, 57 (1st Cir.1997) (observing AED-PA contains both aspects). On a procedural level, a prisoner seeking to file a second or successive application for habeas relief must now apply directly to the court of appeals, which applies the proper AED-PA substantive standard and then grants or denies the prisoner permission to proceed in the district court. Before considering the post-AEDPA substantive standard, we recognize that, for the reasons set out below, there is no impermissible retroactivity in applying AEDPA’s procedural amendments to habeas petitions. This is true regardless of when a petitioner’s first habeas application was filed.
[1187]*1187In general, the Supreme Court has established a two-part analysis for determining statutory retroactivity:
When a case implicates a federal statute enacted after the events in suit, the court’s first task is to determine whether Congress has expressly prescribed the statute’s proper reach. If Congress has done so, of course, there is no need to resort to judicial default rules. When, however, the statute contains no such express command, the court must determine whether the new statute would have retroactive effect, i.e., whether it would impair rights a party possessed when he acted, increase a party’s liability for past conduct, or impose new duties with respect to transactions already completed. If the statute would operate retroactively, our traditional presumption teaches that it does not govern absent clear congressional intent favoring such a result.
Landgraf v. USI Film Prod., 511 U.S. 244, 280, 114 S.Ct. 1483, 128 L.Ed.2d 229 (1994). In other words, when a new statute is passed, the general rule is that courts will not apply the statute in ways that would create new legal consequences for events completed before the statute was enacted. Congress, of course, has the power to override this traditional presumption and direct the statute’s retroactive application, but it must do so explicitly. If Congress does not provide clear guidance, courts apply a case-by-case analysis to determine whether application of the new statute acts to “impair- rights a party possessed when he acted, increase a party’s liability for past conduct, or impose new duties with respect to transactions already completed.” Id. If so, the new statute is impermissibly retroactive and will not be applied to that case.
With regard to the potential retroactivity of AEDPA, the Supreme Court considered the first part of the retroactivity inquiry in Lindh v. Murphy, 521 U.S. 320, 117 S.Ct. 2059, 138 L.Ed.2d 481 (1997), that is, whether Congress expressly declared the reach of the statute. The Court discerned Congress’ implied (but not express) intent “that the new provisions ... generally apply only to cases filed after the Act became effective,” as opposed to cases pending at the time the Act was passed, id. at 336, 117 S.Ct. 2059. The Court therefore held that the AEDPA amendments to 28 U.S.C. § 2254(d) did not apply to the petitioner’s pending noncapital case. Id. The Court expanded this analysis in Slack v. McDaniel, 529 U.S. 473, 120 S.Ct. 1595, 146 L.Ed.2d 542 (2000), determining that an appeal of the denial of a habeas petition is a new “case” to which AEDPA’s procedural gatekeeping amendments apply, even if the original petition was filed before that act was passed, id. at 482, 120 S.Ct. 1595. The Court reasoned: “When Congress instructs us (as Lindh says it has) that application of a statute is triggered by the commencement of a ‘case,’ the relevant case for a statute directed to appeals is the one initiated in the appellate court.” Id. at 1603.
The holdings in Lindh and Slack were based on the Court’s determination, by negative implication, that Congress did not intend AEDPA to apply retroactively to cases pending at the time the Act was passed. Because that determination paralleled the traditional presumption against retroactivity, the Court relied upon basic rules of statutory interpretation to discern Congress’ intention, rather than seeking the explicit Congressional guidance required to apply a statute retroactively.
We agree with those courts of appeals holding that reasoning based solely on Lindh and Slack does not reflect the unambiguous Congressional intent required to overcome the presumption against statutory retroactivity if retroactive effect is present in a particular case. See Mueller [1188]*1188v. Angelone, 181 F.3d 557, 566-68 (4th Cir.1999); Minarik, 166 F.3d at 598-99; United States v. Ortiz, 136 F.3d 161, 165 (D.C.Cir.1998); In re Hanserd, 123 F.3d 922, 934 & n. 22 (6th Cir.1997); see also In re Jones, 226 F.3d 328, 331 (4th Cir.2000) (noting holdings of Lindh and Slack but applying Landgraf analysis). In the case of AEDPA, no unambiguous guidance exists that Congress intended to apply the statute retroactively. See, e.g., Minarik, 166 F.3d at 598-99 (discussing Landgraf s requirements in AEDPA context and concluding “Congress did not clearly express its intent”).
Where the “statute contains no [explicit Congressional] command, the court must determine whether the new statute would have retroactive effect” for a particular case under the standard set out in Landgraf, 511 U.S. at 280, 114 S.Ct. 1483. If so, its terms will not be applied to that case. See Lindh, 521 U.S. at 326, 117 S.Ct. 2059 (“if the application of a term would be retroactive as to Lindh, the term will not be applied, even if, in the absence of retroactive effect, we might find the term applicable”). Since AEDPA contains no express command declaring its amendments to be retroactive, we turn to the second step in the Landgraf retroactivity analysis and consider whether AEDPA’s amendments could “impair rights a party possessed when he acted, increase a party’s liability for past conduct, or impose new duties with respect to transactions already completed.” Landgraf, 511 U.S. at 280, 114 S.Ct. 1483.1 In doing so, we consider separately the Act’s procedural and substantive aspects. See Minarik, 166 F.3d at 599-600; Pratt, 129 F.3d at 57.
It is clearly proper to apply AEDPA’s procedural framework to all second or successive habeas applications filed after the Act’s effective date. The presumption against retroactivity contains a well-established exception for rules of procedure, since procedural rules “regulate secondary rather than primary conduct” and avoid attaching new legal consequences to earlier actions. Landgraf, 511 U.S. at 275, 114 S.Ct. 1483; see also Slack, 529 U.S. at 482, 120 S.Ct. 1595; Minarik, 166 F.3d at 599 (AEDPA’s establishment of courts of appeals as gatekeepers “is a change in procedural law which falls within the firmly established ‘procedural change’ category described in Landgraf that may be retrospectively applied”). That a would-be petitioner must apply to a gatekeeping court of appeals for permission to file a subsequent habeas petition, rather than directly to a district court, does not affect the petitioner’s underlying legal lights; it merely speaks to which court will consider his application.' See Graham v. Johnson, 168 F.3d 762, 782 (5th Cir.1999) (“Substituting the court of appeals for the district court as the gatekeeper against abusive or procedurally defaulted claims would seem to raise no retroactivity concerns. A litigant has no reasonable expectation that a particular tribunal will adjudicate his claims.”). Consequently, regardless of when their first petitions were filed, prisoners must turn to our court as a gatekeeper under section 2244 of AEDPA. See Minarik, 166 F.3d at 599-600; In re Hanserd, 123 F.3d at 934; accord United States v. Gallegos, 142 F.3d 1211, 1212 (10th Cir.1998) (per curiam) (“Because [petitioner’s third § 2255] motion was filed after ... the effective date of AEDPA, he [1189]*1189was required to comply with the Act and obtain prior authorization from this court before filing in the district court.”).
Ill
PRE-AEDPA SUBSTANTIVE REQUIREMENTS
The Landgraf retroactivity analysis focuses on whether a change in law “would impair rights a party possessed when he acted, increase a party’s liability for past conduct, or impose new duties with respect to transactions already completed.” 511 U.S. at 280, 114 S.Ct. 1483. Although we have just determined that AEDPA’s procedural amendments work no retroactive ■ change to a party’s rights and duties, we must now compare the substantive requirements for raising new rules of criminal procedure before and after AEDPA to determine whether any changes impair Mr. Daniels’ ability to raise an Apprendi claim. We turn first to the pre-AEDPA requirements.
The traditional pre-AEDPA test for legal claims raised in a second or successive habeas encompassed two separate inquiries, asking first whether the new claim was procedurally defaulted as an “abuse of the writ” because it was not raised in earlier petitions, and then, if the claim was not defaulted because it was based on a new and hence previously unavailable rule of law, asking whether the new rule of law would be retroactively applied to that claim under Teague v. Lane, 489 U.S. 288, 109 S.Ct. 1060, 103 L.Ed.2d 334 (1989).2 See, e.g., Andrews v. Deland, 943 F.2d 1162, 1180 (10th Cir.1991) (claim dismissed as abuse of writ, or “additionally, and alternatively” unavailable because new legal rule not applied retroactively); Moore v. Zant, 885 F.2d 1497, 1503 n. 8 (11th Cir.1989) (“Both parties concede that they conceive of the abuse of the writ and retroactivity issues as being two separate analyses which must be conducted pursuant to a petitioner’s filing of a successful federal habeas petition”); see also McCleskey, 499 U.S. at 495, 111 S.Ct. 1454 (application of cause and prejudice standard “does not mitigate the force of Teague v. Lane”).
Abuse of the Writ
Prior to AEDPA’s amendments, 28 U.S.C. § 2244 provided that prisoners could raise only new legal or factual claims in a second or subsequent habeas petition and then only if “the applicant [had] not on the earlier application deliberately withheld the newly asserted ground or otherwise abused the writ.” McCleskey, 499 U.S. at 486, 111 S.Ct. 1454 (quoting statute). According to the established common-law test for procedural default, which survived enactment of the statutory re-[1190]*1190gimp., id. at 487, 111 S.Ct. 1454, any claim not raised in an earlier petition was presumptively an abuse of the writ3 unless the petitioner could demonstrate “cause” for his earlier failure to raise the claim and “actual prejudice” resulting from the claimed error.4 Id. at 493, 111 S.Ct. 1454; Andretus, 943 F.2d at 1171. Complicating this cause-and-prejudice inquiry is the fact that the Supreme Court has never “attempted to establish conclusively the contours of the standard.” Amadeo v. Zant, 486 U.S. 214, 221, 108 S.Ct. 1771, 100 L.Ed.2d 249 (1988); see also Reed v. Ross, 468 U.S. 1, 13, 104 S.Ct. 2901, 82 L.Ed.2d 1 (1984) (Court has provided no “precise content” defining “cause”); United States v. Frady, 456 U.S. 152, 168, 102 S.Ct. 1584, 71 L.Ed.2d 816 (1982) (exact understanding of “prejudice” remains “open question” in most contexts). Nevertheless, certain general guidelines have emerged from the Court’s jurisprudence.
A petitioner must first show “cause” for his failure to raise the legal claim in an earlier petition. Habeas petitioners may not choose to withhold claims for disposition in later proceedings, and even negligence in discovering a claim will not excuse a delay. Instead, a petitioner must show that his efforts to raise the claim at earlier stages were “impeded” by “some objective factor external to the defense,” for example, where “the factual or legal basis for a claim was not reasonably available to counsel” during earlier proceedings. McCleskey, 499 U.S. at 493-94, 111 S.Ct. 1454. The standard is an objective one, asking not what a particular attorney or pro se petitioner actually knew but whether the claim was “reasonably available” upon diligent inquiry. Id. at 494, 496, 111 S.Ct. 1454.
When controlling law changes after earlier review has been completed, a petitioner may be able to demonstrate that the new legal claim was “not reasonably available” under the old prevailing standard. Changes in law do not automatically constitute cause, however: “even if the law has changed since the earlier petition [a petitioner] also must show that he or his , attorney reasonably should not have been aware of the possibility or legal basis for such a claim.” Coleman v. Saffle, 869 F.2d 1377, 1380-81 (10th Cir.1989). Cause is present if “a constitutional claim is so novel that its legal basis is not reasonably available to counsel” prior to the change in law. Reed, 468 U.S. 1 at 16, 104 S.Ct. 2901, 82 L.Ed.2d 1.
It can be difficult to define exactly when a claim becomes “available” to counsel. As the Court explained in Reed:
It is in the nature of our legal system that legal concepts, including constitutional concepts, develop slowly, finding partial acceptance in some courts while meeting rejection in others. Despite the fact that a constitutional concept may ultimately enjoy general acceptance, ... when the concept is in its embryonic stage, it will, by hypothesis, be rejected by most courts.
[1191]*1191Id. at 15, 104 S.Ct. 2901. This strong likelihood of rejection led to the Court’s decision that there is no “functional purpose” in any rule “requiring a defendant to raise a truly novel issue.” Id.
While the cause inquiry continues to analyze whether a claim was “reasonably available” prior to a change in law, the Supreme Court narrowed the broad Reed “novelty” test in Bousley v. United States, 523 U.S. 614, 118 S.Ct. 1604, 140 L.Ed.2d 828 (1998). There, the court held that a claim was not “novel” where “Federal Reporters were replete with cases involving challenges” to the legal regime at issue. Id. at 622, 118 S.Ct. 1604. Even if it appears “futile” to attempt a particular legal argument, that perceived futility “cannot constitute cause if it means simply that a claim was ‘unacceptable to that particular court at that particular time.’ ” Id. “[T]he question is not whether subsequent legal developments have made counsel’s task easier, but whether at the time of the default the claim was ‘available’ at all.” United States v. Richards, 5 F.3d 1369, 1371 (10th Cir.1993) (quoting Smith v. Murray, 477 U.S. 527, 537, 106 S.Ct. 2661, 91 L.Ed.2d 434 (1986)). Thus, even a futile claim may be “reasonably available” for “cause” purposes prior to a change in the law.
If a petitioner succeeds in showing cause for failure to raise the new legal claim, he must then demonstrate “ ‘actual prejudice’ resulting from the errors of which he complains.” McCleskey, 499 U.S. at 494, 111 S.Ct. 1454. In the context of jury instructions that are later determined to be unconstitutional, the Supreme Court has explained that the prejudice inquiry asks “not merely whether ‘the instruction is undesirable, erroneous, or even universally condemned;’ ” instead, a court must determine “ ‘whether the ailing instruction by itself so infected the entire trial that the resulting conviction violates due process.’ ” Frady, 456 U.S. at 168, 102 S.Ct. 1584. This in itself is by no means a bright-line rule, and the Court noted in Frady that the definition of prejudice in every other context remains an “open question.” Id. Again, however, there is a general guideline that directs our inquiry.
The burden of showing prejudice is not an easy one. As a rule, the petitioner “must shoulder the burden of showing, not merely that the errors at his trial created a possibility of prejudice, but that they worked to his actual and substantial disadvantage, infecting his entire trial with error of constitutional dimensions.” Id. at 170, 102 S.Ct. 1584. In other words, it is not enough to assert that an error “might have changed the outcome of the trial.” Strickler v. Greene, 527 U.S. 263, 289, 119 S.Ct. 1936, 144 L.Ed.2d 286 (1999). Instead, a petitioner “must convince [a court] that ‘there is a reasonable probability’ that the result of the trial would have been different.” Id. at 289-91, 119 S.Ct. 1936 (explaining lower court wrong to consider “possibility” rather than “probability”). This “reasonable probability” standard does not require that a petitioner demonstrate he “would more likely than not have received a different verdict” without the claimed error. Id. at 289, 119 S.Ct. 1936. Rather, the question is whether, considering the error, “he received a fair trial, understood as a trial resulting in a verdict worthy of confidence.” Id. at 290, 119 S.Ct. 1936. Applying these standards, we have concluded a petitioner did not establish prejudice where the jury did not consider newly required elements of a crime but its verdict “necessarily embraced the missing elements.” United States v. McDonald, 150 F.3d 1301, 1304 (10th Cir.1998).
Teague Retroactivity Inquiry
Under the pre-AEDPA standard, a petitioner who manages to surmount the [1192]*1192substantial requirement of cause and prejudice will be allowed to proceed with raising a new constitutional claim. If that claim is based upon new law decided after his conviction has become final, however, he faces a second, nearly insurmountable, hurdle: as established by the Supreme Court in Teague v. Lane,5 new constitutional rules of criminal procedure will not be applied retroactively on collateral review unless they meet one of two narrow exceptions.
The first step in the Teague analysis is to ask whether the holding of a particular case constitutes a “new rule” for purposes of retroactivity. The Court has explained that
[i]t is admittedly often difficult to determine when a case announces a new rule, and we do not attempt to define the spectrum of what may or may not constitute a new rule for retroactivity purposes. In general, however, a case announces a new rule when it breaks new ground or imposes a new obligation on the States or the Federal Government.
Teague, 489 U.S. at 301, 109 S.Ct. 1060. In essence, “a case announces a new rule if the result was not dictated by precedent existing at the time the defendant’s conviction became final.” Id. (emphasis in original). We recognized in Browning, 241 F.3d at 1266, that Apprendi constitutes a new rule of constitutional law.
Teague held that, in general, “new constitutional rules of criminal procedure will not be applicable to those cases which have become final before the new rules are announced.” 489 U.S. at 310, 109 S.Ct. 1060. The new rule of Apprendi then, is not available for collateral review of Mr. Daniels’ case unless it fits one of two narrow exceptions to Teague’s general nonretroac-tivity rule. The first exception states that “a new rule should be applied retroactively if it places ‘certain kinds of primary, private individual conduct beyond the power of the criminal law-making authority to proscribe.’ ” Id. at 311, 109 S.Ct. 1060. Apprendi does not do so. The second exception allows for retroactive application of “watershed rules of criminal procedure,” those that “require[ ] the observance of those procedures that are implicit in the concept of ordered liberty.” Id. at 311, 109 S.Ct. 1060. This exception is further restricted to “those new procedures without which the likelihood of an accurate conviction is seriously diminished.” Id. at 313, 109 S.Ct. 1060. Not every constitutional error will serve as sufficient basis for a habeas petition, but the second Teag-ue exception recognizes that “errors that undermine confidence in the fundamental fairness of the ... adjudication certainly justify the issuance of the federal writ.” Williams v. Taylor, 529 U.S. 362, 375, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000) (Stevens, J., writing for four justices)6.
[1193]*1193Because the second exception, in particular, applies only when a rule is “central to an accurate determination of innocence or guilt,” the Court has predicted it is “unlikely that many such components of basic due process have yet to emerge.” Teague, 489 U.S. at 313, 109 S.Ct. 1060. The examples Justice O’Connor gave in Teague of such “watershed” procedures were those designed to preclude criminal proceedings “dominated by mob violence” or where “the prosecutor knowingly made use of perjured testimony” or “the conviction was based on a confession extorted from the defendant by brutal methods.” Id. We have held that a new rule fit this exception when it involved “a misdescription of the burden of proof’ in jury instructions. Tillman v. Cook, 215 F.3d 1116, 1122 (10th Cir.2000). We noted Teague’s emphasis on the accuracy of convictions and explained that incorrect standards regarding the burden of proof “vitiate[ ] all the jury’s findings.” Id. (quoting Sullivan v. Louisiana, 508 U.S. 275, 281, 113 S.Ct. 2078, 124 L.Ed.2d 182 (1993)). Because this Teague inquiry is applied both pre- and post-AED-PA, as explained below, its application cannot work retroactive effect in Mr. Daniels’ case. Therefore, we need not decide at this time whether Apprendi fits the second Teague exception.7
Mr. Daniels contends Teague is inapplicable to his case because it was decided in the context of a challenge to state confinement under section 2254 and neither the Supreme Court nor this court has ever applied Teague’s antiretroactivity rule to a federal prisoner’s petition under section 2255. See Talk, 158 F.3d at 1071 n. 5 (declining to address the question). It is true that the Teague rule was based in large part upon the need for federal courts to respect the decisions made by state courts. See Teague, 489 U.S. at 308-10, 109 S.Ct. 1060. Nonetheless, the rule against retroactive application of new laws supports important interests of finality as well as comity. See id. at 309, 109 S.Ct. 1060 (“Without finality, the criminal law is deprived of much of its deterrent effect.”); see also McCleskey, 499 U.S. at 492, 111 S.Ct. 1454 (“Perpetual disrespect for the finality of convictions disparages the entire criminal justice system.”). “[T]he Federal Government, no less than the States, has an interest in the finality of its criminal judgments.” Frady, 456 U.S. at 166, 102 S.Ct. 1584. In fact, concerns of finality carry particular weight in the context of a section 2255 motion, since a “federal prisoner ... has already had an opportunity to present his federal claims in federal trial and appellate forums.” Id. at 166, 102 S.Ct. 1584.
The Supreme Court has described several practical considerations supporting the need for finality in federal criminal cases. For instance, “when a habeas petitioner succeeds in obtaining a new trial, the erosion of memory and dispersion of witnesses that occur with the passage of time prejudice the government and diminish the chances of a reliable criminal adjudication.” McCleskey, 499 U.S. at 491, 111 S.Ct. 1454 (citation omitted). Also, “[federal collateral litigation places a heavy burden on scarce federal judicial resources, and threatens the capacity of the system to [1194]*1194resolve primary disputes.” Id. This is particularly true in the case of second or successive federal collateral challenges because “[i]f reexamination of a conviction in the first round of federal habeas stretches resources, examination of new claims raised in a second or subsequent petition spreads them thinner still.” Id. at 492, 111 S.Ct. 1454.
Those courts of appeals that have addressed the relationship between Teague and section 2255 articulate further reasons for applying the nonretroactivity rule to federal petitions. First, although Teague arose in the context of a section 2254 petition, its analysis relied on earlier opinions by Justice Harlan that considered “cases presenting collateral challenges to federal convictions under § 2255” and “did not ‘propose to make any distinction, for retroactivity purposes, between state and federal prisoners seeking collateral relief ” Van Daalwyk v. United States, 21 F.3d 179,180-81 (7th Cir.1994) (quoting Mackey v. United States, 401 U.S. 667, 681 n. 1, 91 S.Ct. 1160 (1971) (Harlan, J. concurring in part and dissenting in part)). Furthermore, on a basic level it would be “simply inequitable” to allow retroactive application of new rules for federal prisoners but apply Teague to bar application of the same rules for state prisoners. Id. at 183. As the Second Circuit has explained:
[T]he reason the Supreme Court expressly clarified the principles governing retroactivity on collateral review was not to limit the doctrine to state convictions, but to develop a principle that would allow more consistency in results. Injecting a federal/state dichotomy into the picture would defeat rather than further the goal of consistency.
Gilberti v. United States, 917 F.2d 92, 94-95 (2d Cir.1990) (citations omitted). The Fourth Circuit has also held that Teague bars application of new rules in section 2255 petitions, based upon the same considerations set forth by the Seventh and Second Circuits. See United States v. Martinez, 139 F.3d 412, 416 (4th Cir.1998). We agree with these circuits and hold that Teague’s, nonretroactivity doctrine applies equally to habeas petitions brought under sections 2254 and 2255.
Overlap Between Two Inquiries
As explained above, prior to AEDPA a petitioner like Mr. Daniels who wished to raise a new rule of constitutional law in a second or successive habeas petition first had to overcome the “cause and prejudice” test for abuse of the writ in order to raise his claim, and then had to show the new rule could be applied retroactively to that claim under Teague. While we have discussed each test separately, there is a significant amount of overlap between the two inquiries.
A showing of cause to excuse procedural default requires that a new rule be so novel that it was not reasonably available for argument in the past. See Reed, 468 U.S. at 16, 104 S.Ct. 2901. In contrast, Teague bars retroactive application of any new rule “not dictated by prior precedent.” 489 U.S. at 301, 109 S.Ct. 1060. Any claim declared sufficiently novel to meet the cause test, by definition, could not have been dictated by prior precedent. In other words, if one has cause for not raising a constitutional claim in earlier petitions because it is sufficiently “novel,” that same novelty ensures the claim is barred from application on collateral review as a new rale under Teague (unless one of two exceptions applies). See Hopkinson v. Shillinger, 888 F.2d 1286, 1290 (10th Cir.1989) (en banc) (“a holding that a claim is so novel that there is no reasonably available basis for it, thus establishing cause, must also mean that the claim was too novel to be dictated by past precedent”), overruled on other grounds, [1195]*1195Sawyer v. Smith, 497 U.S. 227, 245, 110 S.Ct. 2822, 111 L.Ed.2d 193 (1990); see also Selvage v. Collins, 975 F.2d 131, 136 (5th Cir.1992) (because of Teague, “a rule found novel for cause purposes ... will not be [retroactively] available in a federal ha-beas proceeding”); Prihoda v. McCaughtry, 910 F.2d 1379, 1386 (Teague “nullifies” effect of allowing new rules as basis for cause); Clark v. Dugger, 901 F.2d 908, 912 (11th Cir.1990) (rule novel for cause but barred by Teague).
Before AEDPA, therefore, a new rule would be applied collaterally to a second or successive habeas petition only if it met one of Teague’s two exceptions. If pre-AEDPA law is applied to Mr. Daniels’ habeas application, he could succeed in raising his Apprendi claim in only one situation: if the new rule is novel enough to constitute “cause” under the cause and prejudice test and also fits one of the two exceptions to Teague’s presumption against retroactivity. In any other situation, either the claim itself would be procedurally defaulted or the new rule would be inapplicable as Teague-barred.
IV
POST-AEDPA SUBSTANTIVE REQUIREMENTS
AEDPA’s amendments to 28 U.S.C. § 2244 made no mention of applying either the cause-and-prejudice test or a Teague analysis to applications to file second or subsequent federal habeas petitions. Instead, section 2244(b)(2)(A) now provides that such petitions will not be granted unless they rely upon a “new rule of constitutional law, made applicable to cases on collateral review by the Supreme Court, that was previously unavailable.” As explained in Part I, under this standard, Mr. Daniels cannot succeed in filing a successive habeas petition based upon Apprendi. He argues, therefore, that applying the AEDPA standard takes away his earlier right to raise an Apprendi claim. While we determine here that the AEDPA standard is indeed a new inquiry, rather than a codification of the previous two-part test, we conclude this change in substantive requirements works no impermissible retroactive effect because the result is the same under either standard.
AEDPA Replaces Both Teague and Cause and Prejudice Test
As a rule, “[i]t is not unusual for Congress to codify earlier precedent in the habeas context.” Williams, 529 U.S. at 380 n. 11, 120 S.Ct. 1495 (Stevens, J.). Accordingly, various courts have addressed whether AEDPA codifies the standards set out in Teague or the cause-and-prejudice test, although no court has considered both. Those courts that have analyzed whether AEDPA codifies Teague note that AEDPA “imports an antiretroac-tivity principle” like that of Teague, but conclude that the statutory amendments fall short of specifically adopting the Teag-ue standards. Green v. French, 143 F.3d 865, 873-74 (4th Cir.1998), overruled on other grounds by Williams, 529 U.S. 362, 120 S.Ct. 1495, 146 L.Ed.2d 389; see also Larry W. Yackle, A Primer on the New Habeas Corpus Statute, 44 Buff. L.Rev. 381, 416-17 (1996) (“It is accurate only to say that Congress has legislated in the field in which Teague has operated and has thus reformulated the way in which the federal habeas courts are to deal with the question that Teague previously answered”).
On one hand, AEDPA places an emphasis upon nonretroactivity resembling Teag-ue’s bar on retroactive application. For example, the post-AEDPA requirements for review of state convictions command reliance upon “clearly established law,” which Justice Stevens has suggested is the “functional equivalent” of Teague’s antire-[1196]*1196troactivity principle. Williams, 529 U.S. at 379, 120 S.Ct. 1495.
Because there is no reason to believe that Congress intended to require federal courts to ask both whether a rule sought on habeas is “new” under Teag-ue — which remains the law — and also whether it is “clearly established” under AEDPA, it seems safe to assume that Congress had congruent concepts in mind.
Id. at 379-80, 120 S.Ct. 1495. Although the “clearly established” language never appeared in the Teague jurisprudence, Justice Stevens concluded that the invocation of Teague’s “new rule” phraseology in several other AEDPA subsections supports his “impression that Congress had Teague ... specifically in mind in amending the habeas statute. These provisions, seen together, make it impossible to conclude that Congress was not fully aware of, and interested in codifying into law, that aspect of this Court’s habeas doctrine.” Id. at 380 n. 12, 109 S.Ct. 1060; but see Green, 143 F.3d at 874 n. 1 (fact that Congress used Teague language in some sections strengthens conclusion that, where Congress chose different language, it did not intend to codify Teague). Justice O’Connor, writing -for the Court, seems to agree with Justice Stevens that while Teague did not mention “clearly established federal law” as such, it established the same principle as the opposite of “new rule.” Thus, Justice O’Connor said, “[wjhatever would qualify as an old rule under our Teague jurisprudence will constitute ‘clearly established federal law as determined by the Supreme Court of the United States’ under § 2254(d)(1).” Id. at 412.
On the other hand, AEDPA clearly did not adopt the Teague doctrine wholesale. More specifically, AEDPA tightens application of the presumption against retroac-tivity of new constitutional rules by limiting it to situations in which the Supreme Court itself has declared that the new law should apply. See 28 U.S.C. § 2254(d)(1) (in state context, applying clearly established federal law “as determined by the Supreme Court of the United States”); 28 U.S.C. § 2255 at para. 8 (in federal context, requiring that new rules of constitutional law be “made retroactive to cases on collateral review by the Supreme Court”); see also Browning, 241 F.3d at 1265. This additional requirement “extends the principle of Teague by limiting the source of doctrine on which a federal court may rely.” Williams, 529 U.S. at 381, 120 S.Ct. 1495 (Stevens, J.); id. at 412, 120 S.Ct. 1495 (O’Connor, J.) (“ § 2254(d)(1) restricts the source of clearly established law to this Court’s jurisprudence.”); see also Rodriguez v. Bay State Corr’l Ctr., 139 F.3d 270, 274 (1st Cir.1998) (concluding “AEDPA does not codify Teague” because of addition of this requirement), abrogated on other grounds by Bousley v. United States, 523 U.S. 614, 622-24, 118 S.Ct. 1604, 140 L.Ed.2d 828 (1998).
Turning to the other of the two pre-AEDPA tests, courts have been quick to note that AEDPA does not explicitly mention cause and prejudice. The statute’s silence on cause and prejudice could lead to two very different conclusions, however: either the earlier test has disappeared completely in light of AEDPA or it survives AEDPA as an independent doctrine that must be applied in addition to the AEDPA standard. Those courts that have specifically addressed the relationship between AEDPA and the cause-and-prejudice test have assumed that AEDPA replaces the former test. See, e.g., In re Jones, 226 F.3d at 332 n. 1 (petitioner must satisfy AEDPA standard “rather than” showing no abuse of writ); Minarik, 166 F.3d at 595 (cause and prejudice test replaced by “more rigorous” AEDPA standard). Some have found AEDPA works an impermissibly retroactive effect where a petitioner could have succeeded under [1197]*1197the earlier cause-and-prejudice test but fails under AEDPA. See, e.g., Minarik, 166 F.3d at 600-01 (holding AEDPA im-permissibly retroactive whenever it mandates result different from that of earlier law); Hanserd, 123 F.3d at 931 (same). This is the argument made by Mr. Daniels.
After careful consideration of the questions involved, we are persuaded AEDPA replaces the traditional two-part analysis for second and successive motions seeking to raise new constitutional claims with a single, clear requirement: the motion must rely on “a new rule of constitutional law, made applicable to cases on collateral review by the Supreme Court, that was previously unavailable.” 28 U.S.C. § 2255 at para. 8. Significantly, we find in 28 U.S.C. § 2244 partial support for the idea that Congress intended to replace the cause-and-prejudice test with the new requirements of the AEDPA regime. Under the former test, there may be cause for failure to raise a claim in one of two situations. The first is a new rule of law. The second is newly discovered evidence that reasonably could not have been known at the time of the first filing and that sheds important new light on the petitioner’s case.8 See McCleskey, 499 U.S. at 494, 111 S.Ct. 1454 (cause if “the factual or legal basis for a claim was not reasonably available” earlier). Compare those two possibilities to the AEDPA standard at 28 U.S.C. § 2244(b)(2), which states the following:
A claim presented in a second or successive habeas corpus application under section 2254 that was not presented in a prior application shall be dismissed unless—
(A) the applicant shows that the claim relies on a new rule of constitutional law, made retroactive to eases on collateral review by the Supreme Court, that was previously unavailable; or
(B)(i) the factual predicate for the claim could not have been discovered previously through the exercise of due diligence; and
(ii) the facts underlying the claim, if proven and viewed in light of the evidence as a whole, would be sufficient to establish by clear and convincing evidence that, but for constitutional error, no reasonable factfinder would have found the applicant guilty of the underlying offense.
This language arguably parallels the two options for establishing cause under the pre-AEDPA test. Section 2255 is similar, requiring that “a second or successive motion must be certified as provided in section 2244 by a panel of the appropriate court of appeals to contain” either newly discovered evidence or new constitutional law. 28 U.S.C. § 2255 at para. 8. Both sections differ from the traditional cause standard in that a new law must be both “constitutional” and “made applicable to cases on collateral review by the Supreme Court.” Still, the close parallel between the old and the new standards supports the argument that Congress intended the cause-and-prejudice analysis to be replaced by AEDPA.
Our determination that AEDPA’s new provisions wholly and intentionally replace the concept of cause is further bolstered by the often-overlooked requirement that a new rule of constitutional law made retroactive to cases on collateral review by the Supreme Court must have been previously unavailable. 28 U.S.C. § 2255 at para. 8. The “previously unavailable” language is often dropped from dis-[1198]*1198cussions of the section 2255 test, presumably because it appears to duplicate the “new rule” term. If a rule is “new,” one may generally conclude that the rule was “previously unavailable.” The duplicate language makes sense, however, when considered in light of the theory that Congress intended AEDPA to replace both Teague and the earlier test for cause. Each phrase has a different meaning as a term of art under pre-AEDPA law — the “new rule” term implicates Teague, while the “previously unavailable” term invokes the narrower requirements of the old test for cause. See, e.g., Reed, 468 U.S. at 16, 104 S.Ct. 2901 (cause present if “a constitutional claim is so novel that its legal basis is not reasonably available” prior to change in law) (emphasis added). ■
With respect to the “prejudice” portion of the test, it is fully embodied in section 2244(b)(2)(B)(ii), which governs newly raised factual claims. As for section 2244(b)(2)(A), governing new rules of constitutional law, we have already concluded that AEDPA incorporates a Teague analysis. The second Teague exception speaks of “watershed” new rules of criminal procedure. This Teague exception applies to “those new procedures without which the likelihood of an accurate conviction is seriously diminished.” Teague, 489 U.S. at 313, 109 S.Ct. 1060. When an error seriously affects the accuracy of a conviction, there will be a “reasonable probability” that results might have been different without the claimed error, thus meeting the prejudice test.
Despite our conclusion that AEDPA does indeed replace the old cause-and-prejudice and Teague standards with a ■new requirement that a rule be made applicable to cases on collateral review by the Supreme Court, the result for Mr. Daniels is the same under either regime. As explained below, there is no actual retroactive effect in applying the new AED-PA standard to Mr. Daniels.
No Retroactive Effect as Applied
This case involves a federal prisoner seeking to rely upon a constitutionally based new rule of criminal, procedure in a second or successive habeas petition under section 2255. Our discussion of the pre-AEDPA two-part test for abuse of the writ and retroactive application of law concluded with the observation that such petitioners would succeed only if the rule is novel enough to constitute “cause” and the new law fits the second Teague exception for “watershed” new rules of criminal procedure that seriously affect the fundamental accuracy of a proceeding. AEDPA’s new standard undeniably alters the legal landscape in the sense that only the Supreme Court may now declare a rule of law applicable to second or successive motions. Teague remains controlling law, however, see Williams, 529 U.S. at 379, 120 S.Ct. 1495, so presumably the Court will reach its retroactivity decision by applying a Teague analysis. Even after AEDPA, a new rule of criminal procedure must then meet the second Teague exception before the Court will declare it retroactively applicable to cases on collateral review. The second Teague exception controls the ret-roactivity analysis under either pre-or post-AEDPA law.9
[1199]*1199The current showing required to raise a claim based upon a new rule of criminal procedure thus remains fundamentally unchanged: both before and after AEDPA, the new rule must meet Teague’s, second exception. Because the question of whether Apprendi fits that exception for successive habeas applications is left to the exclusive province of the Supreme Court after AEDPA, we do not attempt to address it here. Instead, we hold that the new AEDPA substantive standards work no impermissibly retroactive effect as applied to Mr. Daniels’ application. Therefore, we must apply the post-AEDPA section 2255 standard to his habeas application. Because Apprendi has not yet been made applicable to cases on collateral review by the Supreme Court as required by section 2255, we conclude based upon our holding in Browning that this application must be dismissed.
y
CONCLUSION
For the foregoing reasons, Mr. Daniels’ request for permission to file a successive habeas petition under 28 U.S.C. § 2255 is DENIED.