Appellant Brian Davis was sentenced to prison for crimes involving powder and crack cocaine before Congress and the Sentencing Commission took steps to reduce the disparity in sentencing ranges between the two. Unfortunately for Davis, these efforts were directed at crimes involving lesser amounts of cocaine than his. In a suit that seeks declaratory relief and possibly damages, Davis claims that these efforts violate the Equal Protection Clause because they do not reach his crimes. This appeal does not take up the merits of Davis’s claims, but their form. The district court dismissed his suit on the ground that the only relief available to Davis is in habeas. For the reasons set forth below, we reverse.
I
For years, the Sentencing Guidelines treated one gram of cocaine base, commonly known as “crack cocaine,” the same as one hundred grams of powder cocaine. See Dorsey v. United States, — U.S. -, 132 S.Ct. 2321, 2327-28, 183 L.Ed.2d 250 (2012). This 100-to-1 ratio came in for heavy criticism from many quarters, and both Congress and the Sentencing Commission took steps to reduce the sentencing disparities it created. Id. at 2328-29. In 2007, the Commission issued Amendment 706, which lowered base offense levels for crimes involving less than 4.5 kg of crack cocaine. U.S. Sentencing Guidelines Manual amend. 706 (2011). Then, in response to the Fair Sentencing Act of 2010, Pub.L. No. 111-220, § 2(a), 124 Stat. 2372, 2372, the Commission issued Amendment 750, which reduced the ratio to 18-to-l for crimes involving less *95than 8.4 kg of crack cocaine. U.S. Sentencing Guidelines Manual app. C, amend. 748 (temporarily reducing the ratio); id., amend. 750 (making Amendment 748 permanent). The Commission made both amendments retroactive, U.S.S.G. § lB1.10(c), allowing inmates convicted based on the old sentencing ranges to seek discretionary sentence reductions under 18 U.S.C. § 3582(c)(2).
In 1993, Davis was convicted of conspiracy to possess with intent to distribute and the distribution of powder and crack cocaine. The sentencing court assigned him a base offense level of 42, which at the time applied to offenses involving 15 kg or more of crack cocaine. The court sentenced Davis to life imprisonment. Davis has sought relief from this sentence, but because neither Amendment 706 nor Amendment 750 applies to offenses involving 15 kg or more of crack cocaine, they are of no help to him. In fact, a district court has twice denied his attempts to reduce his sentence under Amendment 706. He did find some relief elsewhere. In 2008, the district court reduced his sentence from life to 360 months based on an unrelated amendment to the Guidelines.
In 2011, Davis brought this pro se lawsuit seeking relief under the Declaratory Judgment Act, 28 U.S.C. § 2201(a), and Bivens v. Six Unknown Named Agents of the Federal Bureau of Narcotics, 403 U.S. 388, 91 S.Ct. 1999, 29 L.Ed.2d 619 (1971). This latest resort to the courts differs from his previous efforts. Davis does not request a sentence modification pursuant to 18 U.S.C. § 3582(c)(2). Instead, he seeks a declaration that Amendments 706 and 750 deny him equal protection of the laws because they fail to reduce the sentencing disparity for defendants convicted of crimes involving higher quantities of crack cocaine. Only if Davis succeeds on the merits will the sentencing reductions in Amendments 706 and 750 be made applicable to his offense, allowing him to seek a discretionary reduction of his sentence under 18 U.S.C. § 3582(c)(2).
The question before us is whether he has selected the proper vehicles for his equal protection challenges. Neither the Declaratory Judgment Act nor Bivens has carried Davis far. Even before the Commission had answered Davis’s complaint, the district court dismissed his claims for lack of jurisdiction under Federal Rule of Civil Procedure 12(h)(3). Davis v. U.S. Sentencing Comm’n, 812 F.Supp.2d 1, 1 (D.D.C.2011). Declaratory relief was unavailable, the court held, because “an adequate remedy is available by petitioning the sentencing court for a writ of habeas corpus.” Id. at 2 (citations omitted). And Davis’s Bivens action was “patently insubstantial” because he neither sued an individual nor requested damages. Id. (internal quotation marks omitted). Davis appealed, and we appointed an amicus to brief and argue the case on his behalf.1 We have jurisdiction under 28 U.S.C. § 1291, and review the district court’s dismissal de novo. Doe v. Metro. Police Dep’t, 445 F.3d 460, 465 (D.C.Cir.2006).
II
We first consider whether Davis must bring his equal protection challenge by means of a habeas petition. The answer turns on whether his claim for relief is at the “core of habeas.” The Supreme Court has held that Congress has channeled state prisoners’ claims for relief — however styled — into habeas alone if the prisoners seek a remedy that is at the “core of habeas.” See, e.g., Wilkinson v. Dotson, *96544 U.S. 74, 79, 125 S.Ct. 1242, 161 L.Ed.2d 258 (2005) (quoting Preiser v. Rodriguez, 411 U.S. 475, 487, 98 S.Ct. 1827, 36 L.Ed.2d 439 (1973)). In order to determine whether Davis, a federal prisoner, must bring his equal protection challenge by means of a habeas petition, this court must resolve two interrelated questions: Does the scope of the habeas-channeling rule differ for federal and state prisoners? And is the rule for federal prisoners so broad that it includes equal protection challenges to Guidelines amendments?
The modern habeas-channeling rule emerged in Preiser v. Rodriguez, when the Supreme Court held that a prisoner may not challenge “the fact or duration of his confinement” by means of an action brought under 42 U.S.C. § 1983. 411 U.S. at 489, 93 S.Ct. 1827. The state prisoners in Preiser alleged that their period of incarceration had been unlawfully extended when the New York State Department of Correctional Services revoked their good-conduct credits towards early release. Id. at 476-77, 93 S.Ct. 1827. The Court held that Congress set out the procedures prisoners must follow to attack their confinement in the habeas statute. To allow them to pursue release by other means would frustrate the intent of Congress. Id. at 489-90, 93 S.Ct. 1827.
We applied Preiser’s habeas-channeling rule fifteen years later in Chatman-Bey v. Thornburgh, 864 F.2d 804, 808-10 (D.C.Cir.1988) (en banc). That case was distinguishable from Preiser in two ways. First, it involved a federal, not state, prisoner. Id. at 808-09. Second, that prisoner would not necessarily secure an earlier release if he succeeded on the merits of his claim. Success on the merits would have won him nothing more than an earlier appointment with the parole board, which retained discretion to deny him parole. Id. at 809. We saw no legal significance in either distinction. With regard to the first distinction, we noted that channeling federal prisoners’ claims into habeas raised even fewer concerns than channeling state prisoners’ claims because federal prisoners have greater access to federal courts. Id. (citing Preiser, 411 U.S. at 501, 93 S.Ct. 1827 (Brennan, J., dissenting)). Regarding the second distinction, we explained that “Preiser cannot ... be limited to [claims that] would result in immediate release or a definite reduction in the actual amount of time to be spent in prison.” Id. Preiser taught us that “Congress’ provision of an express remedy for unlawful detentions means” that it intended prisoners to rely on that remedy exclusively. Id. This intention covered all claims relating to terms of detention, we reasoned, including Chatman-Bey’s claim “that he [was] being deprived of the chance to secure his release.” Id.
Both Heck v. Humphrey, 512 U.S. 477, 114 S.Ct. 2364, 129 L.Ed.2d 383 (1994), and Edwards v. Balisok, 520 U.S. 641, 117 S.Ct. 1584, 137 L.Ed.2d 906 (1997), the next two Supreme Court cases to apply Preiser’s habeas-channeling rule, involved state prisoners whose successful claims would, like those of the prisoners in Preiser, result in earlier or immediate release. Heck, 512 U.S. at 479-80, 114 S.Ct. 2364; Balisok, 520 U.S. at 648, 117 S.Ct. 1584. Each decision described the rule in terms that cast doubt on our view, expressed in Chatman-Bey, that its scope extended beyond claims for immediate release or a definite reduction in the length of imprisonment. In both cases, the Court stated that a prisoner must bring his claim in habeas if “a judgment in [his] favor ... would necessarily imply the invalidity of his conviction or sentence....” Heck, 512 U.S. at 487, 114 S.Ct. 2364 (emphasis added); Balisok, 520 U.S. at 643, 117 S.Ct. *971584 (quoting Heck, 512 U.S. at 487, 114 S.Ct. 2864).
In Anyanwutaku v. Moore, we relied on the fact that the plaintiffs in Heck and Balisok were state prisoners to hold that prisoners in the custody of the District of Columbia are required to bring their claims in habeas only when success on the merits would “necessarily imply, or automatically result in, a speedier release from prison.” 151 F.3d 1053, 1056 (D.C.Cir.1998) (internal quotation marks omitted). The Anyanwutaku habeas-channeling rule for District prisoners was narrower than the Chatman-Bey rule for federal prisoners. It channeled into habeas only claims that would guarantee a speedier release from prison. The Anyanwutaku panel distinguished Chatman-Bey on the basis that it “dealt expressly with federal prisoners.” Id. at 1057. In light of this distinction, the panel declined to “decide whether Chatman-Bey ha[d] any continuing vitality after Heck and Balisok.” Id. As a result, we were left with a narrow habeas-channeling rule for state and District prisoners and a broad rule for federal prisoners.
Two years later, we were required to confront the issue Anyanwutaku “left open” — Chatman-Bey’s “ ‘continuing vitality after Heck and Balisok.’” Razzoli v. Fed. Bureau of Prisons, 230 F.3d 371, 375 (D.C.Cir.2000) (quoting Anyanwutaku, 151 F.3d at 1057). Razzoli involved a federal prisoner who brought an action challenging the decision of the United States Parole Commission to delay his eligibility for parole by two years. Razzoli, 230 F.3d at 373. Because success on the merits would entitle Razzoli to earlier consideration for parole, but not necessarily earlier release from prison, the rule in Chatman-Bey required him to seek relief in habeas, but the rule in Anyanwutaku did not. Deciding that Heck and Balisok had not “flatly contradicted]” the Chatman-Bey holding, id. at 375, we held that a federal prisoner must still bring his claim in habeas even when success on the merits “would have a merely probabilistic impact on the duration of custody.” Id. at 373.
Davis does not dispute that, under our decision in Razzoli, he must bring his equal protection challenge by means of a habeas petition even though his claim has only a “probabilistic impact on the duration of custody.” If his equal protection challenge succeeds, Davis is at best one step closer to an earlier release from prison. A victory would not secure his immediate release or even a reduction in his time served because the district court would retain discretion to deny him any sentence reduction under § 3582(c)(2) even if he were to prevail on his equal protection challenge. See Dillon v. United States, 560 U.S. 817, 130 S.Ct. 2683, 2691-92, 177 L.Ed.2d 271 (2010). Under Razzoli, that makes no difference for federal prisoners. Razzoli channels their claims into habeas based on the possibility of an earlier release, not on its certainty. See Razzoli, 230 F.3d at 373.
Seeking to avoid the force of Razzoli, Davis argues instead that two Supreme Court decisions — Wilkinson and Skinner — have undermined its reasoning. As Davis points out, we are not bound by circuit precedent that has been “eviscerated by subsequent Supreme Court cases.” Dellums v. U.S. Nuclear Regulatory Comm’n, 863 F.2d 968, 978 n. 11 (D.C.Cir.1988).
The Razzoli court offered four reasons for its decision to retain Chatman-Bey’s broader habeas-channeling rule for federal prisoners instead of taking direction from Heck and Balisok. 230 F.3d at 375-76. First, both Heck and Balisok involved claims that would have a definite impact on the duration of custody, and the Supreme Court had yet to decide whether a “proba*98bilistic” claim like Razzoli’s “need be brought in habeas.” Id. at 375. Second, the Razzoli panel reasoned that habeas cannot be exclusive if it is not available, and habeas is at least available for “probabilistic” claims. Id. Third, the Seventh Circuit had interpreted Preiser, Heck, and Balisok to call for habeas-channeling even of “probabilistic” claims. Id. at 376. Fourth, the Supreme Court had yet to decide a habeas-channeling case that involved federal prisoners, leaving the panel without guidance on whether the distinction between federal and state prisoners was significant. The panel decided it was significant because extending Anyanwutaku’s narrower rule to federal prisoners might lead to congestion in the D.C. Circuit. The habeas statutes require prisoners to bring their petitions in the places they are imprisoned or were sentenced. This works to distribute prisoners’ cases across the nation. By contrast, many other federal prisoner causes of action could be brought in the D.C. Circuit, where many of the federal agencies that might be defendants, like the Commission, are found. Id.
Because three of these four reasons are no longer sound, we overturn Razzoli. Like Razzoli, the state prisoners in Wilkinson v. Dotson challenged their parole eligibility dates, and success on the merits of their claims would not necessarily have resulted in earlier release. Wilkinson, 544 U.S. at 76-77, 125 S.Ct. 1242. Confronting such “probabilistic” claims for the first time, the Supreme Court held that they need not be brought in habeas. Claims that “will not necessarily imply the invalidity of confinement or shorten its duration” are not at the “core” of habeas and therefore may be pursued through other causes of action. Id. at 82, 125 S.Ct. 1242 (emphasis added). After Wilkinson, the Seventh Circuit no longer channels “probabilistic” claims into habeas. See, e.g., Grennier v. Frank, 453 F.3d 442, 444 (7th Cir.2006).
In Skinner v. Switzer, the Supreme Court suggested that habeas might not even be available for “probabilistic” claims, undercutting another reason for the Razzoli rule. — U.S.-, 131 S.Ct. 1289, 1299, 179 L.Ed.2d 233 (2011) (“Switzer has found no case ... in which the Court has recognized habeas as the sole remedy, or even an available one, where the relief sought would neither terminat[e] custody, accelerate] the future date of release from custody, nor reduc[e] the level of custody.” (quoting Wilkinson, 544 U.S. at 86, 125 S.Ct. 1242 (Scalia, J., concurring)) (alterations in original)). In other words, “probabilistic” claims may not even lie within the bounds of habeas, much less at its core. If habeas is not even “proper” for claims with only a probabilistic impact on custody, see Razzoli, 230 F.3d at 375, it could not be the case that Congress intended that prisoners asserting such claims should be limited to habeas.
Because they involve state prisoners, Wilkinson and Skinner leave untouched the Razzoli panel’s final reason for adopting a broad habeas-channeling rule tor federal prisoners: the concentration in the D.C. Circuit of the agencies commonly named in federal prisoner actions. See Razzoli, 230 F.3d at 376. We hold that this reason is not strong enough, standing alone, to support the continued use of the Razzoli rule. Statutes and rules governing venue are adequate to protect the interests' of justice in other cases. See Starnes v. McGuire, 512 F.2d 918, 929-33 (D.C.Cir.1974) (en banc). We see no reason they cannot work just as well here. Anxiety over case congestion cannot foreclose a remedy created by Congress, as with the Declaratory Judgment Act, or by the Constitution, as with Bivens.
*99Acknowledging the weakness of the venue consideration, Appellee’s Br. 30, the Commission insists that there is another distinction between state and federal prisoners that justifies Razzoli’s broader habeas-channeling rule for federal prisoners: They have readier access to federal courts. Appellee’s Br. 19-30. The Commission argues that Razzoli’s broader rule would not be appropriate for state prisoners because the high procedural barriers of the habeas statute would undermine § 1983’s policy of providing access to federal courts for victims of state abuses of civil rights. This argument does not amount to much. It might be true that a broad habeas-channeling rule would be inappropriate for state prisoners, but this argument does not explain why a narrow habeas-channeling rule is not also appropriate for federal prisoners. Our decision in this case will neither advance nor inhibit § 1983’s policy. Federal prisoners are unlikely to bring claims arising from their imprisonment by means of this statute, which creates a cause of action against individuals acting under color of state or District of Columbia law. The Commission’s reasoning does not justify perpetuating Razzoli’s distinction between state and federal prisoners.
Because the Supreme Court has knocked out three of the pillars on which Razzoli rests, we now allow that holding to fall.2 Adopting Wilkinson’s habeas-ehanneling rule, we hold that a federal prisoner need bring his claim in habeas only if success on the merits will “necessarily imply the invalidity of confinement or shorten its duration.” Wilkinson, 544 U.S. at 82, 125 S.Ct. 1242. Otherwise, he may bring his claim through a variety of causes of action.3 And so it is with Davis. Success with his equal protection challenges to Amendment 706 or Amendment 750 will not “necessarily imply the invalidity of [his] confinement or shorten its duration.” Id. Success would do no more than allow him to seek a sentence reduction, which the district court retains the discretion to deny. 18 U.S.C. § 3582(c)(2). His claim for declaratory relief avoids the habeaschanneling rule we announce today, and its dismissal was improper. In so holding, we take no position on whether dismissal for failure to state an equal protection claim might otherwise be proper.
Ill
The district court also dismissed Davis’s Bivens action for lack of subject matter jurisdiction because it found the claim “patently insubstantial.” Davis, 812 F.Supp.2d at 2. We disagree with that finding. We have cautioned that “patently insubstantial” presents an especially high bar for dismissing a claim for lack of subject matter jurisdiction and is no substitute for a dismissal on the merits under Rule 12(b)(6). Best v. Kelly, 39 F.3d 328, *100330-31 (D.C.Cir.1994). Patently insubstantial claims must be “essentially fictitious,” consisting of such things as “bizarre conspiracy theories, ... fantastic government manipulations of [the claimant’s] will or mind, [or] supernatural intervention.” Id. at 330.
Davis’s complaint is admittedly flawed under Bivens, and possibly fatally so. He does not identify an individual federal officer, and he does not request damages. See Correctional Servs. Corp. v. Malesko, 534 U.S. 61, 66, 122 S.Ct. 515, 151 L.Ed.2d 456 (2001) (stating that Bivens recognizes a cause of action for damages against federal officers in their personal capacity). But Davis’s claim, flawed though it may be, is not based on plainly fictitious allegations, and his pleading errors may be corrected through the liberal construction or amendment we are accustomed to providing a pro se prisoner. Of course, his claim might also be dismissed for failure to state a claim. We take no view on that matter. What we do determine, however, is that the district court had jurisdiction to take up the merits of his inartfully pled Bivens claim.
IV
For the foregoing reasons, we reverse the district court’s dismissal of Davis’s complaint and remand for further proceedings consistent with this opinion.
So ordered.