**464The petitioner, Randy Breton, who currently is serving sentences for certain violent crimes he committed in 2011, brought this habeas action against the respondent, the Commissioner of Correction, claiming that a 2013 amendment to General Statutes (Rev. to 2013) § 54-125a; see Public Acts 2013, No. 13-3, § 59 (P.A. 13-3), codified at General Statutes (Supp. 2014) § 54-125a;1 which eliminated risk reduction credit **465awarded pursuant to General Statutes § 18-98e2 from the calculation *792of a violent offender's initial parole eligibility date, thereby requiring the offender to complete 85 percent of his definite sentence before becoming parole eligible, as applied retroactively to him, violates the ex post facto clause of the United States constitution3 because he was statutorily entitled to such earlier parole consideration when he committed the crimes for which he is now incarcerated. The respondent filed a motion to dismiss, and the habeas court dismissed the petition, concluding that the petitioner had failed **466to establish an ex post facto violation because, inter alia, the risk that the petitioner will suffer increased punishment, that is, a longer period of incarceration, as a result of the 2013 amendment is too remote in light of the discretionary nature of risk reduction credit and the fact that such credit, once earned, may be revoked at any time by the respondent for cause. On appeal,4 the petitioner renews his claim of an ex post facto violation. We agree with the petitioner that the ex post facto clause bars the respondent from applying the 2013 amendment to the petitioner, and, accordingly, we reverse the judgment of the habeas court.
The following undisputed facts and procedural history are relevant to our resolution of the petitioner's appeal. On October 27, 2011, the petitioner violated a restraining order, to which he later pleaded guilty. On December 29, 2011, the petitioner committed several additional offenses for which he was charged with two counts of assault in the first degree, one count of assault in the second degree, and one count of larceny in the third degree. On March 21, 2013, pursuant to a plea agreement, the petitioner entered a plea of nolo contendere to the assault and larceny charges, and, on August 22, 2013, he was sentenced to a total effective term of imprisonment of twenty years followed by five years of special parole. The petitioner also received a sentence of thirty months imprisonment for the restraining order violation, which term was imposed to run concurrently with the first sentence.
In 2011, before the petitioner committed his offenses, the legislature passed No. 11-51 of the 2011 Public Acts (P.A. 11-51), § 22, codified at General Statutes § 18-98e. Section 18-98e (a) provides that certain inmates who **467were convicted of crimes committed on or after October 1, 1994, "may be eligible to earn risk reduction credit toward a reduction of such person's sentence, in an amount not to exceed five days per month, at the discretion of the Commissioner of Correction ...." In addition, in 2011, General Statutes (Rev. to 2011) § 54-125a (b) provided that a person convicted of a violent crime was ineligible for parole until such person served at least 85 percent of the definite sentence imposed. The legislature amended that provision in 2011 to allow the application of "any risk reduction credit earned under the provisions of [ § 18-98e ]"; P.A. 11-51, § 25; to accelerate the date on which a violent offender would become eligible for parole. Accordingly, when the petitioner *793committed the offenses for which he is imprisoned, earned risk reduction credit was to be applied by the respondent both to reduce the length of a violent offender's sentence and to advance his or her initial parole eligibility date. See Perez v. Commissioner of Correction , 326 Conn. 357, 364, 163 A.3d 597 (2017) ("under the 2011 amendments, earned risk reduction credit was to be applied to an inmate's definite sentence to advance the inmate's end of sentence date, and the parole eligibility date calculated as a percentage of the sentence would advance in similar measure").
In 2013, after the petitioner was sentenced, the legislature again amended § 54-125a (b) (2), this time by removing the phrase "less any risk reduction credit earned under the provisions of [§] 18-98e." P.A. 13-3, § 59. Thus, under the 2013 amendment, violent offenders are still eligible to earn risk reduction credit to reduce their definite sentence, but that credit is no longer applied to advance their initial parole eligibility date. Consequently, when P.A. 13-3, § 59, became effective on July 1, 2013, inmates convicted of a violent offense thereafter were required to complete 85 percent of their definite sentence before they became eligible **468for parole. See Perez v. Commissioner of Correction , supra, 326 Conn. at 365, 163 A.3d 597.
In 2016, the petitioner filed an amended petition for a writ of habeas corpus, claiming that the 2013 amendment to § 54-125a (b) (2), as applied to him, violates the ex post facto clause because that amendment retroactively increased the amount of time he would be required to serve before becoming eligible for parole. At his habeas trial, the petitioner presented the testimony of Michelle Deveau, a records specialist with the Department of Correction (department), who testified that the petitioner at that time was eligible for risk reduction credit and that, as of the date of the habeas trial, had earned 158 such credits, the maximum number he could have earned at that point in his sentence. Deveau further testified that, although discretionary, risk reduction credit is awarded by the respondent routinely and that, each month, the department's computer system automatically posts it to the timesheets of eligible inmates.
The petitioner also adduced testimony from Heidi Palliardi, a supervisor with the department's Sentence Calculation and Interstate Management Unit, concerning the risk reduction credit program. She testified that risk reduction credit is governed by department administrative directive 4.2A and that, to remain eligible to receive such credit, inmates must follow all institutional rules, remain free of any disciplinary reports and comply with their individual "offender accountability plan," which is provided to every inmate after sentencing. Palliardi further explained that risk reduction credit is subject to forfeiture, after notice and a hearing, for failure to comply with any of the aforementioned program requirements. Finally, the petitioner presented the testimony of Richard Sparaco, the executive director of the Board of Pardons and Paroles (board), who explained that, under the 2013 amendment to § 54-125a **469(b) (2), "the [d]epartment ... no longer could apply risk reduction earned credits to [advance] the parole eligibility date for anyone [who] the board has designated ... a violent offender." Sparaco also stated that parole is granted at the initial parole hearing in approximately 55 percent of all cases.
Following the habeas trial, the court issued a memorandum of decision and dismissed the petition. The court concluded that the petitioner had failed to establish an ex post facto violation because, inter alia, the risk that the petitioner would suffer increased punishment as a result of *794the 2013 amendment to § 54-125a (b) (2) was entirely speculative due to the fact that the award of risk reduction credit is discretionary and the fact that such credit may be revoked by the respondent for cause at any time. We agree with the petitioner that, contrary to the determination of the habeas court, the prohibition of the ex post facto clause bars the retroactive application of the 2013 amendment to § 54-125a (b) (2) to him.5 **470Before addressing the merits of the petitioner's claim, we set forth the governing legal principles. The ex post facto clause of the United States constitution prohibits retroactive application of a law that "inflicts a greater punishment, than the law annexed to the crime, when committed." (Emphasis omitted.) Calder v. Bull , 3 U.S. (3 Dall.) 386, 390, 1 L.Ed. 648 (1798). In other words, the clause "forbids the application of any new punitive measure to a crime already consummated, to the detriment or material disadvantage of the wrongdoer." Lindsey v. Washington , 301 U.S. 397, 401, 57 S.Ct. 797, 81 L.Ed. 1182 (1937). Although a defendant claiming an ex post facto violation need not establish with certainty that retroactive application of a new law will result in greater or more onerous punishment, the United States Supreme Court has "made it clear that mere speculation or conjecture that a change in law will retrospectively increase the punishment for a crime will not suffice to establish a violation of the [e]x [p]ost [f]acto [c]lause. See California Dept. of Corrections v. Morales , 514 U.S. 499, 509 [115 S.Ct. 1597, 131 L.Ed.2d 588] (1995). [Rather, the] touchstone of [the] [c]ourt's inquiry is whether a given change in law presents a sufficient risk of increasing the measure of punishment attached to the covered crimes." (Internal quotation marks omitted.) Peugh v. United States , 569 U.S. 530, 539, 133 S.Ct. 2072, 186 L.Ed.2d 84 (2013). Put differently, the risk of an adverse impact on a prisoner's expected term *795of confinement must be "genuine"; **471Johnson v. Commissioner of Correction , 258 Conn. 804, 818, 786 A.2d 1091 (2002) ; and not merely "remote" or "conceivable ...." California Dept. of Corrections v. Morales , supra, at 508, 115 S.Ct. 1597. Generally speaking, however, "[t]he question when a change in law creates such a risk is a matter of degree; the test cannot be reduced to a single formula." (Internal quotation marks omitted.) Peugh v. United States , supra, at 539, 133 S.Ct. 2072.
Furthermore, "[t]he United States Supreme Court has recognized that a law need not impair a vested right to violate the ex post facto prohibition. Evaluating whether a right has vested is important for claims under the [c]ontracts or [d]ue [p]rocess [c]lauses, which solely protect [preexisting] entitlements.... The presence or absence of an affirmative, enforceable right is not relevant, however, to the ex post facto prohibition, which forbids the imposition of punishment more severe than the punishment assigned by law when the act to be punished occurred. Critical to relief under the [e]x [p]ost [f]acto [c]lause is not an individual's right to less punishment, but the lack of fair notice and governmental restraint when the legislature increases punishment beyond what was prescribed when the crime was consummated. Thus, even if a statute merely alters penal provisions accorded by the grace of the legislature, it violates the [c]lause if it is both retrospective and more onerous than the law in effect on the date of the offense.... Weaver v. Graham , 450 U.S. 24, 29-31, 101 S.Ct. 960, 67 L.Ed.2d 17 (1981)....
"The United States Supreme Court also has recognized that [t]he presence of discretion does not displace the protections of the [e]x [p]ost [f]acto [c]lause. Garner v. Jones , 529 U.S. 244, 253, 120 S.Ct. 1362, 146 L.Ed.2d 236 (2000). Rather, [t]he controlling inquiry ... [is] whether retroactive application of the change in [the] law create[s] a sufficient risk of increasing the measure of punishment attached to the covered crimes.... Thus, unlike a due process claim, the ... focus **472[of which is] primarily on the degree of discretion enjoyed by the [governmental] authority, not on the estimated probability that the authority will act favorably in a particular case ... Giaimo v. New Haven , 257 Conn. 481, 508-509, 778 A.2d 33 (2001), quoting Kelley Property Development, Inc. v. Lebanon , 226 Conn. 314, 323, 627 A.2d 909 (1993) ; [an ex post facto claim's] primary focus ... is the probability of increased punishment." (Citations omitted; internal quotation marks omitted.) Johnson v. Commissioner of Correction , supra, 258 Conn. at 817-18, 786 A.2d 1091.
In addition, it is firmly established that statutes governing parole eligibility are part of the "law annexed to the crime" for ex post facto clause purposes. Calder v. Bull , supra, 3 U.S. (3 Dall.) at 390 ; see, e.g., Warden v. Marrero , 417 U.S. 653, 658, 94 S.Ct. 2532, 41 L.Ed.2d 383 (1974). As the United States Supreme Court explained in Marrero , "[a]lthough ... the precise time at which the offender becomes eligible for parole is not part of the sentence ... it is implicit in the terms of the sentence. And because it could not be seriously argued that sentencing decisions are made without regard to the period of time a defendant must spend in prison before becoming eligible for parole, or that such decisions would not be drastically affected by a substantial change in the proportion of the sentence required to be served before becoming eligible, parole eligibility can be properly viewed as being determined-and deliberately so-by the sentence of the [court]." Id., at 658, 94 S.Ct. 2532 ; see Weaver v. Graham , supra, 450 U.S. at 32, 101 S.Ct. 960 ("[w]e have *796previously recognized that a prisoner's eligibility for reduced imprisonment is a significant factor entering into both the defendant's decision to plea bargain and the judge's calculation of the sentence to be imposed").
Thus, in Weaver v. Graham , supra, 450 U.S. at 24, 101 S.Ct. 960, the Supreme Court struck down on ex post facto grounds **473a then newly enacted Florida statute that reduced the number of good time credits an inmate could earn because the law "effectively postponed the date when [the petitioner] would become eligible for early release." Lynce v. Mathis , 519 U.S. 433, 442, 117 S.Ct. 891, 137 L.Ed.2d 63 (1997) ; see Weaver v. Graham , supra, at 25-27, 101 S.Ct. 960. The court explained that the "retroactive alteration of parole or early release provisions, like the retroactive application of provisions that govern initial sentencing, implicates the [e]x [p]ost [f]acto [c]lause because such credits are one determinant of petitioner's prison term ... and ... [the petitioner's] effective sentence is altered once this determinant is changed." (Internal quotation marks omitted.) Lynce v. Mathis , supra, at 445, 117 S.Ct. 891, quoting Weaver v. Graham , supra, at 32, 101 S.Ct. 960. In Lynce , the court relied on Weaver in deciding that a statute that made certain offenders ineligible to continue earning "overcrowding" credits and retroactively revoked the overcrowding credits those inmates had already earned constituted increased punishment in violation of the ex post facto clause. See Lynce v. Mathis , supra, at 437-39, 442, 445-46, 117 S.Ct. 891.
Since Weaver , federal courts uniformly have held that "it is unconstitutional to apply a statute that alters, to the defendant's disadvantage, the terms under which eligibility for [parole] is calculated, if that statute was enacted after the date of the underlying offense ...." United States v. Paskow , 11 F.3d 873, 878 (9th Cir. 1993). The court in Paskow elaborated on the rationale underlying its assertion: "The [United States] Supreme Court's affirmance of the ... decision in Greenfield v. Scafati , 277 F.Supp. 644 (D. Mass. 1967), aff'd mem., 390 U.S. 713, 88 S.Ct. 1409, 20 L.Ed.2d 250 (1968), makes that proposition indisputable ...." United States v. Paskow , supra, at 878. "Although the Supreme Court affirmed Greenfield in a memorandum [decision] ... its decision is controlling authority. Indeed, the **474[c]ourt has cited its memorandum [decision] with approval [in] Weaver v. Graham , [supra] 450 U.S. [at 34, 101 S.Ct. 960], and has described it as one of [t]he [c]ourt's precedents [ id., at 37, 101 S.Ct. 960 ] ...." (Citation omitted; internal quotation marks omitted.) United States v. Paskow , supra, at 878.
"In Greenfield , a defendant who was incarcerated following revocation of his parole challenged a statute that prohibited any state parole violator from receiving [good conduct] credits during his first six months in custody following revocation.... At the time [that] the defendant committed his underlying crime, all prisoners, including parole violators, could accumulate [good conduct] credits from the beginning of their incarceration. The new statute ... was adopted after the defendant committed his underlying crime, but before he committed the offense for which his parole was revoked. The [three judge panel in Greenfield ] held that application of the statute to the defendant violated the ex post facto clause, because the statute prevented him from being released as early as he might have been had he been permitted to amass [good conduct] credits under the statute in effect at the time he committed the underlying crime. Thus, according to the [three judge panel] and according to the Supreme Court, the statute operated retrospectively and to his *797detriment. As the [three judge panel] stated, the effect of the statute was to [extend] his sentence and [to] increas[e] his punishment beyond the amount he expected or had notice of when he committed his underlying crime. [ Greenfield v. Scafati , supra, 277 F. Supp. at] 645....
"[Federal] [c]ircuit courts that have considered the ex post facto issue have, without exception, followed Greenfield , holding that the ex post facto clause is violated when a defendant's eligibility for release is adversely affected under a statute that was not in effect at the time of the defendant's underlying crime but was adopted before the defendant committed the act for **475which his parole was revoked." (Citation omitted; emphasis omitted; internal quotation marks omitted.) United States v. Paskow , supra, 11 F.3d at 878-79 ; see also, e.g., Schwartz v. Muncy , 834 F.2d 396, 398 n.8 (4th Cir. 1987) ("parole eligibility is part of the law annexed to the crime at the time of a person's offense"); Burnside v. White , 760 F.2d 217, 220 (8th Cir.) ("[t]here is no question that a new parole statute may alter the consequences attached to a crime for which a prisoner already has been sentenced ... [and] to the degree that a statute does so, it has retrospective effect"), cert. denied, 474 U.S. 1022, 106 S.Ct. 576, 88 L.Ed.2d 559 (1985) ; Lerner v. Gill , 751 F.2d 450, 454 (1st Cir.) ("parole eligibility is part of the law annexed to the crime for ex post facto purposes" [internal quotation marks omitted] ), cert. denied, 472 U.S. 1010, 105 S.Ct. 2709, 86 L.Ed.2d 724 (1985) ; Shepard v. Taylor , 556 F.2d 648, 654 (2d Cir.1977) ("[s]ince parole eligibility is considered an integral part of any sentence ... official [postsentence] action that delays eligibility for supervised release runs afoul of the ex post facto proscription" [citation omitted] ).
With the foregoing principles in mind, we must determine whether the 2013 amendment to § 54-125a (b) (2) creates a sufficient risk that the petitioner will be incarcerated longer than he would have been under the 2011 amendment to the statute. As we previously discussed, the habeas court concluded that that risk was too speculative and attenuated to warrant relief under the ex post facto clause because the granting of risk reduction credit is discretionary and the respondent can revoke those credits at any time for cause. In considering whether the habeas court was correct in its analysis, it is instructive to examine cases in which changes to parole eligibility rules were determined not to have violated the ex post facto clause on the ground that the risk of increased punishment was deemed too **476speculative and attenuated. The seminal case, from which the terms "speculative" and "attenuated" derive for the purpose of evaluating the scope of the ex post facto clause, is California Dept. of Corrections v. Morales , supra, 514 U.S. at 508-509, 115 S.Ct. 1597. In Morales , the respondent, Jose Ramon Morales, claimed that a California statute that had changed the frequency-from every year to every three years-within which prisoners convicted of more than one homicide must be reconsidered for parole violated the ex post facto clause. Id., at 503-504, 115 S.Ct. 1597.
As in every ex post facto case, the controlling inquiry for the court was whether the statute's retroactive application created a sufficient risk of increasing the measure of punishment attached to the petitioner's crime. Id., at 509, 115 S.Ct. 1597. In concluding that it did not, the court emphasized three points. First, the court noted that the statute applied only to prisoners who had committed multiple murders, "a class of prisoners for whom the likelihood of release on parole [was extremely] remote" to begin with. Id., at 510, 115 S.Ct. 1597. To demonstrate just how remote, *798the court took judicial notice of evidence presented in another California case that 85 percent of all California prisoners-not just those convicted of multiple homicides-are denied parole at subsequent parole hearings. Id., at 510-11, 115 S.Ct. 1597. Second, it was extremely important to the court that "[t]he amendment has no effect on the date of any prisoner's initial parole suitability hearing; it affects the timing only of subsequent hearings." (Emphasis in original.) Id., at 511, 115 S.Ct. 1597 ; see also id., at 507, 115 S.Ct. 1597 (noting that "[t]he amendment ... left unchanged the substantive formula for securing any reductions to [the petitioner's] sentencing range ... [and] had no effect on the standards for fixing a prisoner's initial date of eligibility for parole" [citations omitted; internal quotation marks omitted] ); id., at 512, 115 S.Ct. 1597 ("there is no reason to conclude that the amendment will have any effect on **477any prisoner's actual term of confinement"). Finally, the court emphasized that the statute did not require the parole board to delay subsequent hearings by up to two years; rather, it merely gave it the discretion to do so "to avoid the futility of going through the motions of reannouncing its denial of parole suitability on a yearly basis." Id.
Recently, in Perez v. Commissioner of Correction , supra, 326 Conn. at 357, 163 A.3d 597, we relied on the court's reasoning in Morales in concluding that retroactive application of another 2013 amendment to General Statutes (Rev. to 2013) § 54-125a (e), which allowed the board to decline to hold a prisoner's initial parole eligibility hearing if certain conditions were met; see Public Acts 2013, No. 13-247, § 376; did not run afoul of the ex post facto clause. See Perez v. Commissioner of Correction , supra, at 377-78, 163 A.3d 597. As we explained in Perez , "the statute in effect when the petitioner [Dominic Perez] committed his offense [provided] that the board shall conduct a hearing when a person has completed 85 percent of his total effective sentence. General Statutes (Rev. to 2009) § 54-125a (e). The 2013 amendment provide[d] that the board may conduct a hearing at that time ... but require[d] that, in the event that the board declines to hold a hearing, it must document the specific reasons for not doing so and provide such reasons to the offender." (Emphasis in original.) Perez v. Commissioner of Correction , supra, at 375-76, 163 A.3d 597. In reaching our decision in Perez , we likened the 2013 amendment to General Statutes (Rev. to 2013) § 54-125a (e) to the California statute at issue in Morales , observing that, as in Morales , the amendment did not affect a violent offender's initial parole eligibility date but, instead, merely permitted the parole board to delay the initial hearing, as long as the board documented its reasons for doing so. Id., at 377-78, 163 A.3d 597. We concluded that, "[b]ecause the parole hearing provision [did] not alter the **478calculation of when an inmate is eligible for parole, and because the board ... still [was required to] consider the inmate's parole suitability at that time, the elimination of a mandatory hearing in the 2013 parole hearing provision [did] not increase the punishment imposed for the petitioner's offense." Id.
In the present case, it cannot reasonably be argued that the 2013 amendment to General Statutes (Rev. to 2013) § 54-125a (b) (2) does not "alter the calculation of when [the petitioner] is eligible for parole ...." Id., at 377, 163 A.3d 597. It clearly does so by eliminating risk reduction credit from that calculation. Indeed, the petitioner has consistently earned the maximum number of risk reduction credits that were available to him, and the respondent has provided no reason to believe either that the petitioner will be denied risk reduction *799credit in the future or that any credit that he earns or already has earned is likely to be revoked. In such circumstances, it strikes us as quite speculative to conclude that the petitioner's release date will not be adversely affected by retroactively applying the 2013 amendment to him. See, e.g., Lynce v. Mathis , supra, 519 U.S. at 444, 117 S.Ct. 891 ( Morales "rested squarely on the conclusion that a prisoner's ultimate date of release would be entirely unaffected by the change in the timing of [parole] suitability hearings" [internal quotation marks omitted] ); see also Weaver v. Graham , supra, 450 U.S. at 35-36, 101 S.Ct. 960 (finding ex post facto violation when "the new provision constrict[ed] the inmate's opportunity to earn early release, and thereby [made] more onerous the punishment for crimes committed before its enactment"); United States v. Paskow , supra, 11 F.3d at 877 (in considering whether new law violates ex post facto clause, "a court must focus on the change in the defendant's eligibility to receive a lesser sentence than a new law may permit, regardless of whether the defendant would actually have received the lesser sentence" [emphasis in original] ). Under the **479reasoning of Morales and Perez , therefore, applying the amendment retroactively to the petitioner does not pass muster under the ex post facto clause.
In support of his contrary contention, the respondent relies on several cases that have no material bearing on the present case because they were all brought under the due process clause. For example, the respondent cites Baker v. Commissioner of Correction , 281 Conn. 241, 914 A.2d 1034 (2007), for the proposition that "parole eligibility ... is not within the terms of the sentence imposed"; (internal quotation marks omitted) id., at 260, 914 A.2d 1034 ; and that "parole eligibility under § 54-125a does not constitute a cognizable liberty interest sufficient to invoke habeas jurisdiction." Id., at 261-62, 914 A.2d 1034. On the basis of this language, the respondent argues that "parole in Connecticut ... is of no constitutional significance" and that "there cannot be an ex post facto violation because parole is not part of the sentence."
In Baker , however, we took pains to distinguish the due process claim at issue in that case from an ex post facto claim-a claim that was not made in that case-stating in relevant part: "In Johnson v. Commissioner of Correction , supra, 258 Conn. at 808, 818-19, 786 A.2d 1091, this court concluded that the habeas court had jurisdiction to consider the petitioner's ex post facto challenge to the board's parole eligibility calculation for his sentence based on its retroactive application of § 54-125a (b) (2) and (c). We distinguished the petition in Johnson from that in Vincenzo v. Warden , 26 Conn. App. 132, 142, 599 A.2d 31 (1991), [in which] the Appellate Court had [upheld the decision of the habeas court rejecting] a due process challenge to [the retroactive application of] the predecessor to § 54-125a ... because the broad discretion vested in the board to determine whether to consider an inmate's parole suitability under the statute did not give rise to a cognizable liberty interest. Johnson v. Commissioner of Correction , supra, at 816-17, 786 A.2d 1091. This **480court explained that, [u]nlike [the petitioner in Vincenzo ], however, the petitioner in [ Johnson was] claiming a violation of his rights under the ex post facto clause as opposed to the due process clause. The United States Supreme Court has recognized that a law need not impair a vested right to violate the ex post facto prohibition. Evaluating whether a right has vested is important for claims under the ... [d]ue [p]rocess [clause], which solely protect[s] [preexisting] entitlements." (Internal *800quotation marks omitted.) Baker v. Commissioner of Correction , supra, 281 Conn. at 260-61, 914 A.2d 1034. Furthermore, the respondent's assertions cannot be squared with the controlling case law previously discussed, which makes eminently clear that the rules governing parole eligibility are part of the law annexed to the crime for purposes of the ex post facto clause. See, e.g., Warden v. Marrero , supra, 417 U.S. at 658, 94 S.Ct. 2532.
The respondent next claims, consistent with the determination of the habeas court, that any risk of increased punishment arising out the retroactive application of the 2013 amendment to the petitioner is too speculative and attenuated to constitute an ex post facto violation because the award of risk reduction credit is discretionary and any such credit that may be awarded is subject to revocation by the respondent for cause. As we previously explained, however, the undisputed testimony adduced at the petitioner's habeas trial belies this contention. That evidence established that, although discretionary and subject to revocation for cause, risk reduction credits are awarded by the respondent routinely and are revoked only for acts of institutional misconduct; moreover, the petitioner has earned all such credits for which he was eligible, and, to date, he has not forfeited any of those credits.
Notably, the respondent has failed to identify a single case in which a court has concluded that eliminating good time credits from the calculation of an offender's **481initial parole eligibility dates did not violate the ex post facto clause merely because the credits at issue were discretionary rather than mandatory, and our independent research has not revealed any such case. This is not surprising for the reasons set forth by the Maryland Court of Appeals in Secretary, Dept. of Public Safety & Correctional Services v. Demby , 390 Md. 580, 890 A.2d 310 (2006). In that case, the court stated in relevant part: "The nature of the special project credits ... is irrelevant for the purposes of [the] ex post facto analysis. The focus of [the] analysis is not on whether these credits were mandatory or discretionary but, rather, whether the amendments to the regulations [that provide] these credits [have] the effect of lengthening [the] respondents' sentences and [are] more 'onerous' than the prior law.... [T]he sentences of those individuals ... whose qualifying crimes have been changed to disqualifying crimes by the amendments, have clearly been lengthened.
"[The court does] not find the increased punishment caused by the amendments ... to be 'speculative and attenuated.' ... [T]he case from which that language originates, Morales ... is factually distinguishable from the present case. In Morales , the statutory change affected the frequency of parole eligibility hearings for inmates by giving parole officials the ability, after meeting several procedural safeguards, to postpone an inmate's yearly evaluation by up to three years when potential safety issues, among other things, [are] a concern, and parole officials [believe] the inmate would not be eligible for parole during the extended period regardless. [Morales'] ex post facto claims were rejected as the chances of an increased punishment were 'speculative and attenuated.' [In this case, the] respondents will clearly serve a longer period of time as a result of the amendments, and the determination of that increase is far easier than in Morales .
**482"Moreover ... the language included in the regulation providing that, 'this section may not be interpreted ... to mean that an inmate who is eligible to receive the credits described in this section has a right to these credits or ... will continue *801to receive these credits in the future,' does not provide sufficient notice to inmates for the purposes of the ex post facto prohibition.... That disclaimer alone does not exempt the regulation from ex post facto scrutiny." (Citation omitted; footnote omitted.) Id., at 617-18, 890 A.2d 310.
The only case cited by the respondent that even arguably may be read to provide a measure of support for his position is Abed v. Armstrong , 209 F.3d 63 (2d Cir.), cert. denied, 531 U.S. 897, 121 S.Ct. 229, 148 L.Ed.2d 164 (2000), in which the Second Circuit Court of Appeals, applying the law of Connecticut, rejected a claim that a then newly adopted directive of the department disqualifying prison gang members from earning good time credits violated the ex post facto clause. Id., at 66. The court gave two reasons for its decision, one of which was the discretionary nature of the credits. Specifically, the court stated: The petitioner's "argument that the [d]irective increased his punishment by restricting his eligibility to earn good time credit assumes that before the [d]irective [ General Statutes §] 18-7a (c) automatically entitled all inmates to be eligible to earn good time credit. That assumption is erroneous. Unlike the statutes at issue in both Weaver and Lynce , [§] 18-7a (c) does not automatically confer the right to earn good time credit on all inmates. Rather, the statute [provides] only that inmates 'may' earn good time credit ... thereby rendering good time credit a discretionary matter....
"In addition, and again unlike Lynce and Weaver , the [d]irective was not applied retroactively to [the petitioner]. No good time credit earned by [the petitioner] prior to the [d]irective was forfeited, and [the petitioner]
**483was not classified as a [gang member] until after the [d]irective was in effect. By its terms, the [d]irective concerned ongoing and future-not past-conduct."6 (Citation omitted.) Id.
The present case is distinguishable from Abed in two crucial respects. First, it is undisputed that the 2013 amendment is being applied to the petitioner retroactively, not punishing him for acts committed while he was in prison, as was the case in Abed . Second, in contrast to the petitioner in Abed , the petitioner in the present case is not claiming a right to earn risk reduction credits. Indeed, he acknowledges that the award of such credits is discretionary with the respondent and, further, that, ultimately, his opportunity to continue to earn them is a matter of legislative grace. He claims only that the risk reduction credits that he does accumulate over the years-however few or many that may be-must be applied to reduce his definite sentence and to advance his initial parole eligibility date in accordance with the law in existence at the time of his *802offense, which law did not give the respondent discretion to refuse to include those credits in calculating his initial parole eligibility date. In other words, the **484petitioner is claiming a right only to the more beneficial formula for calculating his parole eligibility date, not to the credits on which that formula is predicated.7
Our conclusion that the 2013 amendment to § 54-125a (b) (2), as applied to the petitioner, violates the ex post facto clause is reinforced by the legislative history surrounding the enactment. As the habeas court noted, many legislators who supported the 2013 amendment did so out of concern that the prior version of § 54-125a was too lenient, at least with respect to violent offenders. Those legislators wanted to eliminate risk reduction credit from the calculation of the initial parole eligibility date for violent offenders to ensure that they could not be paroled prior to completing 85 percent of their definite sentences. This legislative purpose bears consideration in the present case for the same reason that the court in Lynce v. Mathis , supra, 519 U.S. at 433, 117 S.Ct. 891, reviewed similar legislative history in discussing its decision in Morales . In Lynce , the court considered **485it relevant that the retroactive change to the parole eligibility rules at issue in Morales was not intended to prevent the early release of prisoners. Specifically, the court in Lynce stated: "[The court] concluded [in Morales that] the change at issue had neither the purpose nor the effect of increasing the quantum of punishment. Whether such a purpose alone would be a sufficient basis for concluding that a law violated the [e]x [p]ost [f]acto [c]lause when it actually had no such effect is a question the [c]ourt has never addressed. Moreover, in Morales , [the court's] statements regarding purpose did not refer to the purpose behind the creation of the original sentencing scheme; they referred instead to the question whether, in changing that sentencing scheme, the legislature intended to lengthen the inmate's sentence. To the extent that any purpose might be relevant in this case, it would only be the purpose behind the [new statute ]. Here, unlike in Morales, there is no evidence that the legislature's change in the sentencing scheme was merely to save time or money. Rather, it is quite obvious that the retrospective change was intended to prevent the early release of prisoners convicted of [murder related ] offenses who had accumulated overcrowding credits ." (Emphasis added.) Id., at 444-45, 117 S.Ct. 891. It is similarly *803clear that the 2013 amendment to § 54-125a (b) (2) was intended to prevent the early release of certain offenders and not to conserve time or resources.
It is true, of course, that only a relatively small percentage of inmates-namely, those inmates who, like the petitioner, are incarcerated for committing a violent crime between 2011 and 2013-will be affected by our holding today. Moreover, the only relief to which those inmates are entitled is parole consideration prior to completion of 85 percent of their sentence; whether to grant parole at that time is a decision that remains solely within the broad discretion of the board. But **486the ex post facto clause safeguards the right of those inmates to such consideration regardless of whether they are granted parole at that initial hearing. Accordingly, the petitioner is entitled to the relief he seeks.
The judgment is reversed and the case is remanded with direction to render judgment for the petitioner.
In this opinion the other justices concurred.