HINKLE, District Judge:

In this case a Florida prisoner challenges a statute widening the permissible gap between parole interviews. The district court dismissed the complaint for failure to state a claim on which relief can be granted. We affirm.

I

A Florida state court convicted the appellant Ben E. Jones on two counts of sexual battery. The crimes occurred in December 1978 and February 1979. The court sentenced Mr. Jones to life in prison. His offenses were parole eligible.

At that time Florida law required the Florida Parole Commission to conduct an initial interview and subsequent interviews at least every two years. Fla. Stat. §§ 947.172, 947.174(1) (1978); see also Tuff v. State, 732 So.2d 461, 462 n. 1 (Fla. 3d DCA 1999) (noting the every-two-year requirement).

In 2001, the Legislature adopted a statute allowing the Parole Commission to schedule interviews up to five years apart for offenders convicted of specific serious crimes, including sexual battery. Ch.2001-124, Laws of Fla. (2001). In 2010, the Legislature adopted a statute allowing the Commission to increase the interval to seven years. Ch.2010-95, Laws of Fla. (2010).

Mr. Jones's most recent interview was in 2012. The complaint alleges that Mr. Jones's next interview is scheduled for five years later. Mr. Jones filed this federal lawsuit in 2013 challenging the Commission's procedures on various grounds, including on the ground that the five-year interval between interviews violates the Ex Post Facto Clause.

The district court screened the complaint at the outset as required by the Prison Litigation Reform Act. See 28 U.S.C. § 1915A. The court dismissed the complaint for failure to state a claim on which relief can be granted. The court relied on Penoyer v. Briggs, 206 Fed.Appx. 962 (11th Cir. 2006) (unpublished), which affirmed the dismissal of a complaint challenging on Ex Post Facto grounds the Florida statute that increased the permissible interval between parole interviews to five years.

Mr. Jones moved for reconsideration, asserting, among other things, that it was error to dismiss the complaint without leave to amend. Mr. Jones did not tender an amended complaint or suggest what new allegations any amended complaint would include. The district court denied the motion for reconsideration.

Mr. Jones appeals. We address the Ex Post Facto claim and reject Mr. Jones's other claims without discussion.

II

To avoid dismissal for failure to state a claim on which relief can be granted, a complaint must include "factual content that allows the court to draw the reasonable inference that the defendant is liable for the misconduct alleged." Ashcroft v. Iqbal, 556 U.S. 662, 678, 129 S.Ct. 1937, 173 L.Ed.2d 868 (2009). The complaint's factual allegations, though not its legal conclusions, must be accepted as true. Id.; see also Bell Atl. Corp. v. Twombly, 550 U.S. 544, 555, 127 S.Ct. 1955, 167 L.Ed.2d 929 (2007).

Mr. Jones is proceeding pro se. A pro se pleading is held to a less stringent standard than a pleading drafted by an attorney; a pro se pleading is liberally construed. Hughes v. Lott, 350 F.3d 1157, 1160 (11th Cir. 2003). Even so, a pro se pleading must suggest (even if inartfully) that there is at least some factual support for a claim; it is not enough just to invoke a legal theory devoid of any factual basis.

We review de novo a district court's sua sponte dismissal for failure to state a claim for relief under 28 U.S.C. § 1915A(b). Harden v. Pataki, 320 F.3d 1289, 1292 (11th Cir. 2003).

III

In 1983, Florida abolished parole for most new crimes. Ch. 83-87, Laws of Fla. (1983). But offenders who, like Mr. Jones, committed crimes before that time remain eligible for parole. The Florida Parole Commission (now known as the "Florida Commission on Offender Review") is the agency that administers the parole system.

The general framework that applied when Mr. Jones committed his crimes remains in place today. A hearing examiner conducts an initial interview and determines the offender's presumptive parole release date ("PPRD") based on objective guidelines. Subsequent interviews are "limited to determining whether or not information has been gathered which might affect the presumptive parole release date." Fla. Stat. § 947.174(1) (1978). A PPRD can be modified only (1) based on new information, (2) based on institutional conduct, or (3) "for good cause in exceptional circumstances." Fla. Stat. §§ 947.16(4), 947.172(3), 947.173(3) (1978); McKahn v. Fla. Parole and Prob. Comm'n, 399 So.2d 476, 478 (Fla. 1st DCA 1981).

The Florida Legislature has twice modified § 947.174 to allow the Parole Commission to extend the interval between subsequent interviews for offenders convicted of specific serious crimes, including sexual battery. Mr. Jones did not challenge the first increase, to five years, when it took effect in 2001, and the statute of limitations has run on any challenge to that increase. See Lovett v. Ray, 327 F.3d 1181, 1182-83 (11th Cir. 2003) (holding that the limitations period for a challenge to a change in the frequency of parole consideration runs from the date when the offender learns of the change, not the date of parole consideration, and that the continuing-violation doctrine does not apply). But Mr. Jones now challenges the scheduling of his next interview five years out, and, at least in this court, Mr. Jones challenges the statutory increase in the permissible interval between interviews to seven years, which took effect in 2010. His challenge to that increase is not time barred.

IV

The United States Constitution prohibits a state from passing an "ex post facto Law." U.S. Const. art. I, § 10. The prohibition extends to laws that "retroactively alter the definition of crimes or increase the punishment for criminal acts." Collins v. Youngblood, 497 U.S. 37, 43, 110 S.Ct. 2715, 111 L.Ed.2d 30 (1990). This includes an increase in punishment resulting from a change in a state's parole system.

A

The Supreme Court has twice considered—and twice rejected—an Ex Post Facto challenge to a state's increase in the interval between parole reviews.

First, in California Department of Corrections v. Morales, 514 U.S. 499, 115 S.Ct. 1597, 131 L.Ed.2d 588 (1995), an offender committed a crime while a statute was in effect that required the state parole board to conduct an initial hearing and subsequent hearings at least annually. An amendment allowed the parole board to increase the interval between hearings to two years, upon a finding (on a basis stated by the parole board) that the offender was unlikely to be paroled during the interim. Id. at 512-13, 115 S.Ct. 1597. The parole board could conduct a hearing sooner if warranted. Id. at 513, 115 S.Ct. 1597.

The Ninth Circuit held that the amendment violated the Ex Post Facto Clause. Morales v. Cal. Dep't of Corr., 16 F.3d 1001 (9th Cir. 1994). The Supreme Court reversed, concluding that a change in parole procedures violates the Ex Post Facto Clause only if it creates a sufficient risk that offenders will actually serve more time in prison. See id. at 509, 514. The Court explicitly declined to adopt a test for determining when a risk is sufficient, but the Court said not every conceivable risk is enough. Id. at 509. And the Court said the amendment at issue did not create a sufficient risk; it created "only the most speculative and attenuated possibility of producing the prohibited effect"—that is, of increasing offenders' punishment. Id. at 510.

Similarly, in Garner v. Jones, 529 U.S. 244, 120 S.Ct. 1362, 146 L.Ed.2d 236 (2000), an offender committed a crime in Georgia while the state parole board's policies called for consideration of parole every three years. The board changed the policies to increase the interval to eight years. Id. at 247, 120 S.Ct. 1362. The policies vested the board with discretion to set an earlier date and allowed an offender to seek expedited review upon a change in circumstances. Id. at 254, 120 S.Ct. 1362.

We held that the change in policies violated the Ex Post Facto Clause. Jones v. Garner, 164 F.3d 589 (11th Cir. 1999). But again the Supreme Court disagreed. The Court said the issue was whether the new policies created a significant risk of prolonging the offender's time in prison and that nothing inherent in the new policies created such a risk. 529 U.S. at 252, 120 S.Ct. 1362. The Court remanded for this court or the district court to consider whether the offender should be allowed discovery into the actual effect of the new policies in operation.

B

Together Morales and Jones compel rejection of the claim that the Florida statutory change at issue here is facially unconstitutional. Here, as in Morales and Jones, the interval between parole assessments has increased. But here, as in Morales and Jones, the substantive policies that govern parole have remained the same. And here the risk that the change will increase an offender's actual punishment is no greater than—indeed, probably less than—the risk in Morales or Jones.

This is so because here, as in Morales, the Commission can delay a periodic review only upon finding, on grounds set out in writing, that an offender is unlikely to be paroled in the interim. Here, as in Morales and Jones, an earlier periodic review can be scheduled, as a matter of discretion, if circumstances change. And here, much more than in Morales or Jones, the scope of a periodic review is limited. In Florida each offender has an objectively determined PPRD that can be changed only on specified grounds. The subject of subsequent interviews is not a freewheeling consideration of factors that might impact a discretionary parole decision; the subject of subsequent interviews is limited to whether there are grounds to change the PPRD. At least on its face, this suggests that the determination to delay an interview can be made more reliably, and the chance that the delay will impact the actual parole date is less, than in California or Georgia.

To be sure, Mr. Jones points to other differences in the California, Georgia, and Florida provisions that he says cut the other way. The California change applied to those convicted of two or more murders—a category one hopes is quite small. The Georgia change applied to offenders serving life sentences. The Florida change applies only to specific offenses, see Fla. Stat. § 947.174(1)(b), but the category is broader than in California or Georgia. Still, Florida abolished parole long ago, so the number of affected offenders may not be large. More importantly, the Ex Post Facto Clause protects all comers; if the California or Georgia change had violated the Clause, the fact that the change impacted only a limited category of offenders would not have saved it. The critical issue is whether a change creates a significant risk of producing the prohibited effect, not whether it will have the prohibited effect on all of a state's offenders or only a limited category.

Mr. Jones also complains that Florida law allows the Commission to schedule an earlier interview based on changed circumstances but does not allow an offender to initiate the process. Still, nothing in Florida law prohibits an offender from bringing circumstances to the Commission's attention and asking for an earlier interview. That the offender cannot compel an earlier interview is not surprising and does not meaningfully distinguish the case from Morales or Jones. In California and Georgia, an offender can request, but cannot compel, earlier consideration.

Here, as in Morales and Jones, nothing inherent in the state's decision to enlarge the maximum period between parole considerations poses a significant risk that offenders will serve more time in prison. The Florida change is not facially unconstitutional.

C

That leaves for consideration Mr. Jones's claim that the Florida change is unconstitutional as applied to him. The easiest answer may be that according to Mr. Jones's own complaint, he is scheduled for an interview just five years after his last interview in 2012. The statutory increase in the maximum authorized interval to seven years apparently has not affected Mr. Jones at all; if this is correct, the change cannot yet have been unconstitutional as applied to him.

More importantly, Mr. Jones has alleged no facts suggesting that an earlier interview would make any difference in his case at all. In 2012, when the Commission set Mr. Jones's next interview five years out, the Commission was required to find that he was unlikely to be paroled prior to that time. The Commission was required to state the grounds for that determination in writing. An earlier interview could make a difference only if Mr. Jones could qualify for a change in his PPRD based on new information, based on institutional conduct, or "for good cause in exceptional circumstances." See Fla. Stat. §§ 947.16(4), 947.172(3), 947.173(3) (1978). Mr. Jones has not alleged facts showing any of these prerequisites to a change in his PPRD. And although Mr. Jones says he should have been given leave to amend his complaint, he did not proffer in the district court, and he has not proffered even now, any facts showing any of the prerequisites to a change in his PPRD.

In short, Mr. Jones has alleged no facts suggesting his parole prospects will be different in 2017, or even in 2019, than they were in 2012. We have recognized that in the absence of a showing that an earlier interview would have made a difference, a delayed interview does not violate the Ex Post Facto Clause. See Jones v. Ray, 279 F.3d 944, 946 (11th Cir. 2001).

Mr. Jones cites statistics showing a precipitous decline in the number of offenders paroled in Florida. But the primary explanation for that decline is clear and unrelated to the change in the frequency of parole interviews. Florida abolished parole in 1983. With every passing year—with every release of a parole-eligible offender who has served an entire sentence or been paroled—the number of parole-eligible inmates decreases. Florida would have had a precipitous decline in the number of offenders released on parole with or without a change in the interview schedule.

Mr. Jones also says he must participate in a transitional program to be approved for parole but cannot enroll in the program until he nears release—a catch-22 that denies him (and others) an opportunity for earlier-than-projected release. The argument seems to confuse the cause with the effect. Completion of the program does not cause parole eligibility. Instead, the program is provided near a projected release date because that is when an offender will transition out of prison and thus needs to participate in the program. In any event, there is no apparent connection between any of this and the change in the interview schedule. If, as seems unlikely, there is really a catch-22, then it exists whether interviews are conducted every two or five or seven years.

Finally, Mr. Jones notes that he has already served more than 30 years on his sentence and is 77 years old. The suggestion apparently is that he has served long enough and should be paroled. But Florida law affords an offender no right to parole, even after decades in prison or at age 77. Florida instead establishes a PPRD based on objective criteria and limits the circumstances under which the PPRD can be changed. This was the substantive law when Mr. Jones committed his crimes, and it is the substantive law now. The Ex Post Facto Clause affords an offender no relief from the law that was in effect when a crime was committed.

Florida's statutory change in the maximum permissible interval between parole interviews is not unconstitutional as applied to Mr. Jones.

D

In reaching these conclusions, we have not overlooked our decisions allowing offenders who challenged the Georgia policy change at issue in Jones to take discovery or present evidence on the actual effects of the policy change. See Jones v. Garner, 211 F.3d 1225, 1225 (11th Cir. 2000) (on remand from the Supreme Court, remanding in turn to the district court "to determine, after permitting sufficient discovery, whether the [new Georgia policy] in its operation created a significant risk of increased punishment for" the plaintiff); Harris v. Hammonds, 217 F.3d 1346, 1350 (11th Cir. 2000) (concluding that challenges to the Georgia policy change must be addressed on a case-by-case basis, noting that the case was decided in the district court before the Supreme Court decided Jones, and remanding to provide the offender "the opportunity to make the showing required by the Supreme Court in Jones.").

Those cases are different from this one in three important respects.

First, Jones and Harris arose in Georgia, where the substantive standards that govern parole are much different than in Florida. When each subsequent consideration of an offender's possible parole is wide open, the chance that the interval will make a difference is greater than when the issues at the subsequent consideration are as circumscribed as in Florida. And in Florida, unlike in Georgia, the interview can be delayed only on an explicit written finding that the offender is unlikely to be paroled in the interim. The California statute at issue in Morales had the same requirement for an explicit finding, and the Supreme Court held the statute constitutional; the Court did not remand for further proceedings as it did in Jones.

Second, the offenders in Jones and Harris had no advance notice, when they originally presented their cases in the district court, of the standard the Supreme Court would later adopt in Jones. This made it appropriate to give the offenders another chance to plead and make a case in the district court.

And third, Mr. Jones, unlike the offenders in those cases, still has been unable to allege facts that would make a difference—facts that, in Mr. Jones's case, would support a change in his PPRD and an earlier release. It is not enough that the Florida change in the interview schedule conceivably could affect another offender, or even many other offenders. Mr. Jones has alleged no facts plausibly suggesting the change has adversely affected him. Even in Georgia, the absence of such a showing is fatal to an offender's Ex Post Facto claim. See Jones v. Ray, 279 F.3d 944, 946 (11th Cir. 2001).

That other offenders under other circumstances were afforded an opportunity to take discovery or present evidence does not mean that Mr. Jones is entitled to the same opportunity. Mr. Jones has not alleged facts plausibly suggesting that he is at a significant risk of increased time in prison as a result of the Florida statute allowing longer intervals between parole interviews. The district court properly screened Mr. Jones's complaint as required by the Prison Litigation Reform Act and properly dismissed the complaint for failure to state a claim on which relief can be granted.

V

For these reasons, the judgment of the district court is affirmed.