ON PETITION TO TRANSFER
BOEHM, Justice.
Jose Daniel Segura pleaded guilty to dealing in cocaine. He appeals the denial of his successive petition for postconviction relief, raising one issue: whether his trial counsel was ineffective for failing to inform him of the possibility of deportation if he pleaded guilty. In State v. Van Cleave, 674 N.E.2d 1293, 1306 (Ind.1996), we held that in order to upset a conviction based on a claim of ineffective assistance of coun[499]*499sel, a petitioner who pleads guilty must show a reasonable probability that he would not have been convicted if he had gone to trial. We hold today that the United States Supreme Court’s recent decision in Williams v. Taylor, 529 U.S. 362, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000), does not affect the Van Cleave standard for evaluating ineffective assistance of counsel claims as to errors or omissions of counsel that overlook or impair a defense. As to those claims, we remain of the view that in order to establish that the guilty plea would not have been entered if counsel had performed adequately, the petitioner must show that a defense was overlooked or impaired and that the defense would likely have changed the outcome of the proceeding. Similarly, if counsel’s shortcomings are claimed to have resulted in a lost opportunity to mitigate the penalty, in order to obtain a new sentencing hearing, the petitioner must show a reasonable probability that the oversight would have affected the sentence.
This case presents a claim that counsel’s incorrect advice as to the penal consequences led the petitioner to plead guilty when he otherwise would not have done so. However, this is not a claim that, through erroneous advice, a sentence less than the potential maximum was promised or predicted to induce a plea. Rather, the claim is that the maximum was misdescribed by trial counsel. This error in advice would have weighed equally in the calculation of the consequences of conviction after trial and conviction after a plea. As to such a claim, we conclude that a finding of prejudice requires evidence demonstrating a reasonable probability that the erroneous or omitted advice materially affected the decision to plead guilty.
Factual and Procedural Background
In 1995, Segura pleaded guilty to dealing in cocaine.1 He was sentenced to a term of ten years imprisonment. In 1996, Segura filed an unsuccessful petition for postconviction relief. In September 1998, the Court of Appeals, pursuant to Post-Conviction Rule 1, section 12, granted Seg-ura permission to file a successive petition for postconviction relief. Segura’s second petition contended that his trial counsel was ineffective because he failed to inform Segura that he could be deported as a result of his guilty plea. At the hearing, Segura’s trial counsel testified that the two had never discussed deportation as a possible consequence of the guilty plea.
The postconviction court denied the second petition and the Court of Appeals affirmed in a not-for-publication opinion. Segura v. State, No. 10A01-9906-PC-218 (Ind. Ct.App. April 4, 2000). In so doing, the Court of Appeals relied on the standard set forth by this Court in State v. Van Cleave, 674 N.E.2d 1293 (Ind.1996), for evaluating a claim of ineffective assistance of counsel by a petitioner who had pleaded guilty. Two weeks later, on April 18, 2000, the United States Supreme Court issued Williams v. Taylor, 529 U.S. 362, 120 S.Ct. 1495, 146 L.Ed.2d 389. On May 1, Segura filed a petition for rehearing in light of the Williams decision. The Court of Appeals granted the petition and again affirmed the denial of relief. Segura v. State, 729 N.E.2d 594, 597 (Ind.Ct.App.2000). The Court of Appeals acknowledged that this Court’s opinion in Van Cleave had relied in part on an interpretation of Lockhart v. Fretwell, 506 U.S. 364, 113 S.Ct. 838, 122 L.Ed.2d 180 (1993), that was inconsistent with Williams. Segura, 729 N.E.2d at 596-97. The Court of Appeals noted the possible effect of Williams, [500]*500but held that any change in the doctrine announced in Van Cleave must come from this Court. Id. On September 1, 2000, this Court granted transfer.
I. Deportation as a Penal Consequence
Because Segura alleges prejudice from advice as to deportation, we must decide as a threshold issue whether a failure to counsel about the possibility of deportation constitutes deficient performance as required under Hill. There is a split of authority on this point. The majority of federal circuit courts hold that, as a matter of law, failure to advise of the prospect of deportation as a result of conviction is not deficient performance by counsel in connection with a guilty plea. United States v. George, 869 F.2d 333, 337 (7th Cir.1989); United States v. Yearwood, 863 F.2d 6, 7-8 (4th Cir.1988); United States v. Campbell, 778 F.2d 764, 768-69 (11th Cir.1985). State courts are also split on the issue. Compare, e.g., Alanis v. State, 583 N.W.2d 573, 579 (Minn.1998), with State v. Figueroa, 639 A.2d 495, 499-500 (R.I.1994).
The question has never been addressed by this Court, but the Indiana Court of Appeals has held that “the consequence of deportation, whether labeled collateral or not, is of sufficient seriousness that it constitutes ineffective assistance for an attorney to fail to advise a noncitizen defendant of the deportation consequences of a guilty plea.” Williams v. State, 641 N.E.2d 44, 49 (Ind.Ct.App.1994). We agree with the Court of Appeals that the failure to advise of the consequence of deportation can, under some circumstances, constitute deficient performance. Otherwise stated, we cannot say that this failure as a matter of law never constitutes deficient performance. Whether it is deficient in a given case is fact sensitive and turns on a number of factors. These presumably include the knowledge of the lawyer of the client’s status as an alien, the client’s familiarity with the consequences of conviction, the severity of criminal penal consequences, and the likely subsequent effects of deportation. Other factors undoubtedly will be relevant in given circumstances. The postconviction court found no deficient performance on the part of Segura’s counsel. It is not clear, however, whether this was a holding that, as a matter of law, the failure to advise Segura of the risk of deportation was merely a collateral matter, or whether this was a finding of adequate performance on the facts of this case.2 Because we conclude that Segura failed to establish the prejudice prong, we need not resolve this issue. Strickland v. Washington, 466 U.S. 668, 697, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984).
II. Prejudice in a Guilty Plea Setting
A. Precedent to Date
To prevail on a claim of ineffective assistance of counsel, a petitioner [501]*501must show two things: (1) the lawyer’s performance fell below an “objective standard of reasonableness,” Strickland v. Washington, 466 U.S. 668, 687-88, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984); and (2) “there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Id. at 694, 104 S.Ct. 2052. Effectiveness of counsel is a mixed question of law and fact. Id. at 698, 104 S.Ct. 2052. The specific problem presented today is the relationship of Williams v. Taylor, 529 U.S. 362, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000), to our decision in State v. Van Cleave, 674 N.E.2d 1293 (Ind.1996), as it relates to the “prejudice” prong of the analysis. Van Cleave held that to set aside a conviction, a petitioner who has pleaded guilty must establish that there is a reasonable probability that he would not have been convicted had he gone to trial. 674 N.E.2d at 1306. Until Van Cleave, most Indiana courts had cited the passage from Hill v. Lockhart, 474 U.S. 52, 59, 106 S.Ct. 366, 88 L.Ed.2d 203 (1985), described below, to the effect that it was sufficient to set aside a conviction if the postconviction court concluded that there was a reasonable probability the petitioner would not have pleaded guilty and would have gone to trial. Most, if not all, of these statements were in the course of denying relief for failure to meet even that standard, and do not address how this showing could be made. Nonetheless, this mantra was repeated a number of times without challenge. See, e.g., Burse v. State, 515 N.E.2d 1383, 1385-86 (Ind.1987).
Van Cleave, 674 N.E.2d at 1297-98, rejected the Burse formulation, and in doing so relied in part on Lockhart v. Fretwell, 506 U.S. 364, 369, 113 S.Ct. 838, 122 L.Ed.2d 180 (1993). In Fretwell, the United States Supreme Court elaborated on the prejudice prong of Strickland: “[A]n analysis focusing solely on mere outcome determination, without attention to whether the result of the proceeding was fundamentally unfair or unreliable, is defective.” Fretwell, 506 U.S. at 369, 113 S.Ct. 838. This Court took the view that Fretwell amplified Strickland’s prejudice prong by requiring the petitioner to show that the result of a proceeding was “fundamentally unfair or unreliable,” in addition to showing that the outcome would have been different but for counsel’s mistakes. Williams, however, made clear that Fret-well did not alter the preexisting Strickland showing. In Williams, the United States Supreme Court held that Fretwell did not require a showing that the conviction or sentence was “unfair or unreliable” to establish the prejudice prong of a claim of ineffective assistance of counsel in all cases. 529 U.S. at 391-93, 120 S.Ct. 1495. Rather, Fretwell applies in the rare instance where “the likelihood of a different outcome attributable to an incorrect interpretation of the law should be regarded as a potential ‘windfall’ to the defendant rather than the legitimate ‘prejudice’ contemplated by ... Strickland.” Id. at 392, 120 S.Ct. 1495. Williams made clear that the prejudice is to be measured by the oft-quoted “reasonable probability” of a different “result” set forth in Strickland. The issue is therefore whether the conclusion reached in Van Cleave was correct without the support we found in Fretwell for that result.
B. Prejudice from Counsel’s Shortcomings Affecting a Defense or Sentencing
Segura and the Court of Appeals correctly noted that Van Cleave relied in part on Fretwell in elaborating the Strickland standard in the context of a guilty plea. However, Van Cleave relied not only on Fretwell, but also on Strickland and Hill. 674 N.E.2d at 1296-97. Although Fretwell [502]*502bolstered our confidence in the conclusion reached in Van Cleave, we conclude that the Van Cleave interpretation of the prejudice prong remains valid under Strickland and Hill, and we reaffirm that holding as to claims of counsel’s errors that, if corrected, would either raise a defense or affect the penalty.
Strickland dealt with the results of a trial. One year later, in Hill, 474 U.S. at 57-60, 106 S.Ct. 366, the United States Supreme Court addressed the application of Strickland in a guilty plea setting. The petitioner’s claim in Hill was that he had been incorrectly advised as to his eligibility for parole if he pleaded guilty. Id. at 60, 106 S.Ct. 366. In the frequently quoted passage on which Segura relies, Hill stated that, “in order to satisfy the ‘prejudice’ requirement, the defendant must show that there is a reasonable probability that, but for counsel’s errors, he would not have pleaded guilty and would have insisted on going to trial.” Id. at 59, 106 S.Ct. 366. However, later in that same opinion, the United States Supreme Court observed that in the case of many commonly alleged errors of counsel, “the resolution of the ‘prejudice’ inquiry will depend largely on whether the affirmative defense likely would have succeeded at trial.” Id. Hill further observed that predicting the outcome of a possible trial is to be done on objective evidence, without regard to the “idiosyncrasies of the potential decision-maker.” Id. at 60, 106 S.Ct. 366 (quoting Strickland, 466 U.S. at 695, 104 S.Ct. 2052). The Court also quoted with approval a passage from Evans v. Meyer, 742 F.2d 371, 375 (7th Cir.1984), which rejected a claim because it was “inconceivable” that the petitioner “would have gone to trial ..., or that if he had done so he ... would have been acquitted.”3 474 U.S. at 59, 106 S.Ct. 366. These comments were in the context of a discussion of errors that affect a defense, i.e., those that overlook a defense, fail to develop available evidence for a defense, or impair a defense by inadequate legal analysis.
Hill reasoned that prejudice from an error or omission of counsel that has the effect of overlooking or impairing a defense is to be evaluated by measuring the likelihood of success of that defense. In those cases, prejudice will “closely resemble” the prejudice inquiry for an error at trial. Id. This analysis assumes that the uninvestigated evidence or the overlooked legal point will be competently evaluated, and the decision to go to trial or plead guilty will turn on the likelihood of success. If so, the prejudice in the context of a guilty plea is equivalent to an evaluation of the merits of the defense. As we observed in Van Cleave, 674 N.E.2d at 1299, that is the path followed by the Seventh Circuit4 and the Connecticut Supreme Court5 in reaching results similar to Van Cleave based solely on Strickland and Hill. Similarly, the Tenth Circuit held it necessary to “determine whether it is likely that a jury would have acquitted” in evaluating prejudice from failure to advise that a “depraved mind” was required to be established to prove the crime. Miller v. [503]*503Champion, 161 F.3d 1249, 1256-57 (10th Cir.1998). We also observe that many cases stating the general proposition that the test of prejudice is whether the petitioner would not have pleaded guilty and would have gone to trial, in fact addressed allegations that, if proved, would have altered the calculus of probability of conviction.6 As such, they reach the same result as Van Cleave by proceeding to measure the effect on a decision to plead by evaluating the probability of success of the omitted defense or evidence.
In Van Cleave, we were less certain that this reasoning applied to all claims of ineffective assistance. “[I]n our view, Hill’s reference to the outcome of a possible trial was not put so strongly that we can confidently extract from that case the ‘reasonable probability’ we hold is required under Fretwell.” Van Cleave, 674 N.E.2d at 1299. Now that Williams has made clear that Fretwell did not alter Strickland, we must resolve the point we left open in Van Cleave, namely, whether Hill alone leads to the same result we reached in Van Cleave. In Van Cleave, the postconviction court granted relief and ordered a new trial. 674 N.E.2d at 1295. The State appealed the setting aside of the guilty plea, but not the ruling that there was ineffective assistance at the penalty phase. Id. at 1294. As a result, the only contested shortcomings of counsel — specifically, the failure to raise an intoxication defense— bore on the likelihood of conviction or acquittal.
For the reasons given above, we conclude that Hill standing alone requires a showing of a reasonable probability of success at trial if the alleged error is one that would have affected a defense. This result seems preferable for several reasons. In Van Cleave, we identified sound reasons for requiring that a petitioner who pleads guilty show a reasonable probability of acquittal in order to prevail in a postcon-viction attack on the conviction based on a claim of ineffective assistance of counsel. Id. at 1300-02. As Hill emphasized, the State has an interest in the finality of guilty pleas. 474 U.S. at 58, 106 S.Ct. 366. This is in part grounded in the cost of a new trial, and the demands on judicial resources that are imposed by revisiting the guilty plea, see United States v. Timreck, 441 U.S. 780, 784-85, 99 S.Ct. 2085, 60 L.Ed.2d 634 (1979), but also in concerns about the toll a retrial exacts from victims and witnesses who are required to revisit the crime years later.
A new trial is of course necessary if an unreliable plea has been accepted. But its costs should not be imposed needlessly, and that would be the result if the petitioner cannot show a reasonable probability that the ultimate result — -conviction— would not have occurred despite counsel’s eiror as to a defense. A requirement of a showing of a reasonable probability of success on the merits is consistent with the literal language of Strickland. It is also not unfair. It permits raising any defense that has a reasonable probability of success, but prevents a petitioner who had no valid defense from causing and benefiting from the many difficulties of a retrial years after the events, including missing or deceased witnesses, stale evidence, and fading memories. To the extent that the decision to enter a guilty plea is largely a petitioner’s decision, it is different from the tactical or investigatory steps that are the bases of most claims of ineffective [504]*504assistance of counsel. As this Court observed in Van Cleave, 674 N.E.2d at 1301:
Demonstrating prejudice seems particularly appropriate in the context of a claim of ineffective assistance by a defendant who has pleaded guilty. The guilty plea, virtually uniquely among all procedural steps, involves the judgment of the defendant as well as his attorney. ... [T]he decision to plead is often strongly if not overwhelmingly influenced by the attorney’s advice. But it is equally true that the defendant appreciates the significance of the plea and is uniquely able to evaluate its factual accuracy. The requirement that the court satisfy itself as to the factual basis for the plea is designed to ensure that only guilty defendants plead guilty, and also that the defendant’s decision to waive a jury trial is an informed and reflective one. Many decisions at trial — calling a given witness, asserting a defense, or the extent of cross-examination — are difficult if not impossible for the defendant to make, and reliance on counsel is unavoidable. In contrast, the decision whether to plead guilty is ultimately the prerogative of the defendant, and the defendant alone. More than conjecture or hope for a lucky break at trial should be required to upset that action years later. The Supreme Court has often reminded us that “in judging prejudice and the likelihood of a different outcome, [a] defendant has no entitlement to the luck of a lawless decisionmaker.” Nix v. Whiteside, 475 U.S. 157, 175, 106 S.Ct. 988, 89 L.Ed.2d 123 (1986) (citing Strickland) (internal quotations omitted).
Similarly, if the error or omission has the result of overlooking evidence or circumstances that affect the sentence imposed, prejudice is evaluated by the reasonable probability that it had that effect.
C. Prejudice from Counsel’s Legal Advice as to Penal Consequences
An attorney’s incorrect advice as to penal consequences generally falls into two basic groups: (1) claims of promised leniency and (2) claims of incorrect advice as to the law. Those in the second category do not claim a promised benefit from a plea as compared to the result of a trial. Rather, they claim the range of penal consequences was undervalued. These cases, like Hill, present situations where the advice is equally erroneous whether the defendant pleads or goes to trial. Segura’s claim is of that sort.
1. Claims of Promised Leniency
Some petitions allege in substance a promise of leniency in sentencing. In other words, the claim is that a different result was predicted or guaranteed to result from a plea. In some courts, this claim has been supported by independent evidence substantiating the contention that counsel promised or predicted that a plea agreement would produce a lesser sentence. See, e.g., State v. Bowers, 192 Ariz. 419, 966 P.2d 1023, 1029 (Ct.App.1998). We agree that, if a petition cites independent evidence controverting the record of the plea proceedings and supporting a claim of intimidation by an exaggerated penalty or enticement by an understated maximum exposure, it may state a claim. Some petitions have been supported by objective facts — typically affidavits of counsel — that support the claim and do not controvert the record. Under these circumstances, in an era predating a developed body of ineffective assistance law, this error has been held to create a factual issue as to whether the plea was voluntary. Dube v. State, 257 Ind. 398, 275 N.E.2d 7 (1971). Whether viewed as ineffective assistance of counsel or an involuntary plea, the postconviction court must resolve the [505]*505factual issue of the materiality of the bad advice in the decision to plead, and post-conviction relief may be granted if the plea can be shown to have been influenced by counsel’s error. However, if the postcon-viction court finds that the petitioner would have pleaded guilty even if competently advised as to the penal consequences, the error in advice is immaterial to the decision to plead and there is no prejudice.
2.Segura’s Claim of Incorrect Advice as to the Law
It is less clear what the general formulation of prejudice taken from Hill means where the ineffective assistance is omitted advice as to potential penal consequences that is claimed to have been material to the petitioner’s decision, but would have been equally erroneous whether the defendant pleaded guilty or was convicted at trial. We agree with Chief Judge Aspen that it is “far from obvious” how a petitioner is to make a showing of prejudice from an error by counsel of this type. United States v. Ryan, 986 F.Supp. 509, 513 (N.D.Ill.1997) (claim that erroneous advice as to length of sentence, career offender status, etc., coerced plea). Segu-ra’s claim of ineffective assistance is not based on an allegation of ineffective representation that resulted in overlooked evidence or an unasserted legal defense. Nor does he claim that his counsel failed to develop evidence that would have mitigated his penalty. Rather, he argues that the failure to explain the full penal consequences — in his case, the risk of deportation — caused him to accept a plea agreement that he would have rejected if he had been properly advised. This is for these purposes the same as the claim in Hill that the petitioner was incorrectly advised as to eligibility for parole. Both bear only on the petitioner’s claim that he inadequately understood the penal consequences of a guilty plea. Neither suggests that if counsel had performed flawlessly a conviction would not have resulted from a trial or a different penalty would have been imposed.
3. Hill’s Discussion of the Issue
The petitioner in Hill did not allege that he would not have pleaded if properly advised. For that reason, the United States Supreme Court found it “unnecessary to determine whether there may be circumstances under which erroneous advice by counsel as to parole eligibility may be deemed constitutionally ineffective assistance of counsel.” Hill, 474 U.S at 60, 106 S.Ct. 366. Hill thus declined to rule on the prejudice showing required for a claim of ineffective assistance based on defective advice as to the penal consequences.
Hill did, however, go on to observe that petitioner alleged “no special circumstances that might support the conclusion that he placed particular emphasis on his parole eligibility in deciding whether or not to plead guilty.” Id. The Court then observed that the petitioner’s assessment of his parole eligibility would have equally affected his view of the expected penal consequences of both a plea and a conviction after trial. Id. Unlike Hill, Segura has alleged that he would not have pleaded, but, like Hill, Segura has alleged no “special circumstances” as to why his decision was affected by the alleged omission of counsel. And, it is clear that here, as in Hill, the omitted or misdescribed penal consequences flow equally from either a plea or a conviction after trial.
4. Post-Hill Decisions
Hill himself ultimately prevailed in the Eighth Circuit on his claim that bad advice as to parole eligibility caused him to plead when he would not have done so if proper[506]*506ly advised. After losing in the Supreme Court, Hill filed a second petition that cured the pleading defect identified in Hill by alleging that his guilty plea was caused by the bad advice. Ultimately, the Eighth Circuit affirmed the grant of habeas corpus. Hill v. Lockhart, 877 F.2d 698 (8th Cir.1989). In the Eighth Circuit’s view:
To succeed under Strickland, Hill need not show prejudice in the sense that he probably would have been acquitted or given a shorter sentence at trial, but for his attorney’s error. All we must find here is a reasonable probability that the result of the plea process would have been different — that Hill “would not have pleaded guilty and would have insisted on going to trial.”
Id. at 704 (quoting Hill, 474 U.S. at 59, 106 S.Ct. 366). This was affirmed en banc by a five-to-four decision. Hill v. Lockhart, 894 F.2d 1009 (8th Cir.1990). The Eighth Circuit later explained its holding in Hill in Hale v. Lockhart, 903 F.2d 545, 549 (8th Cir.1990): “The holding in Hill, however, was narrow, and rested primarily on the district court’s finding that petitioner pleaded guilty as a direct consequence of his counsel’s erroneous advice and that, but for this advice, the outcome of the plea process would have been different.” Similarly, the Eleventh Circuit has held that prejudice is shown when a guilty plea is induced by a failure to advise that a guilty plea in state court would not preclude federal authorities from imposing sanctions for parole violations based on the same conduct. Finch v. Vaughn, 67 F.3d 909, 916-17 (11th Cir.1995). Some courts have found no deficient performance in the failure to advise as to sentencing or penal consequences. United States v. Gordon, 4 F.3d 1567, 1570-71 (10th Cir.1993); Ford v. Lockhart, 904 F.2d 458, 462-63 (8th Cir.1990).
Faced with this uncertainty as to what needs to be alleged and how it may be proven, a number of other courts have rejected claims of prejudice after a guilty plea, holding it is insufficient for the petitioner, without more specific facts, merely to allege in postconviction proceedings that he would not have pleaded if he had been properly represented. Others reject claims that a plea would not have been entered when the record of the plea proceeding establishes that the sentencing parameters were known and the factual basis for the plea was established. United States v. Standiford, 148 F.3d 864, 869 (7th Cir.1998); Arango-Alvarez v. United States, 134 F.3d 888, 892-93 (7th Cir.1998); Jones v. Page, 76 F.3d 831, 844-45 (7th Cir.1996). Some have formulated the test as whether a correct understanding of the law would have affected counsel’s recommendation to plead.7 All of these approaches, though phrased differently from the way we expressed it in Van Cleave, lead to the same ultimate conclusion as to the required showing of prejudice. If a change in counsel’s recommendation is the test, because a plea agreement is virtually assured to produce no worse penal consequences than a conviction after trial, a change in counsel’s recommendation would necessarily turn on an evaluation of whether an adequate legal performance would produce a reasonable chance of a better result from a trial. This formulation thus amounts to the same conclusion announced in Van Cleave : a showing of prejudice to upset a guilty plea requires a showing of a reasonable probability of a result of not guilty. Similarly, these cases in one way or another suggest that to show prejudice [507]*507the petitioner’s allegation must controvert the record that he was told of the maximum penal consequence.
5. Prejudice from Advice that Omits or Misdescribes Penal Consequences
We have no clear guidance from the United States Supreme Court as to how to resolve these seemingly inconsistent theoretical approaches to evaluate prejudice from incorrect legal advice in the guilty plea setting. We see no reason to require revisiting a guilty plea if, at the end of the day, the inevitable result is conviction and the same sentence. Yet, we agree that in extreme cases, a credible scenario can be posited that results in a truly innocent defendant pleading guilty because of incorrect advice as to the consequences. The cases where a showing of prejudice from incorrect advice as to the inevitable consequences of conviction will be able to be made may be few. If such a circumstance is shown, however, the defendant should not be stripped of the presumption of innocence, the requirement of proof beyond a reasonable doubt, and the other procedural rights that are not available in postconviction proceedings. To require a showing of innocence to obtain a new trial would have that effect. Accordingly, we conclude that in order to state a claim for postconviction relief a petitioner may not simply allege that a plea would not have been entered. Nor is the petitioner’s conclusory testimony to that effect sufficient to prove prejudice. To state a claim of prejudice from counsel’s omission or misdescription of penal consequences that attaches to both a plea and a conviction at trial, the petitioner must allege, in Hill’s terms, “special circumstances,”8 or, as others have put it, “objective facts”9 supporting the conclusion that the decision to plead was driven by the erroneous advice.
We believe a showing of prejudice from incorrect advice as to the penal consequences is to be judged by an objective standard, i.e., there must be a showing of facts that support a reasonable probability that the hypothetical reasonable defendant would have elected to go to trial if properly advised. Nevertheless, as we understand Strickland and Hill as informed by Williams, a petitioner may be entitled to relief if there is an objectively credible factual and legal basis from which it may be concluded that “there is a reasonable probability that, but for counsel’s errors, he would not have pleaded guilty and would have insisted on going to trial.” Hill, 474 U.S. at 59, 106 S.Ct. 366.
In sum, for the reasons given in Part II, to prove this in the case of claims related to a defense or failure to mitigate a penalty, it must be shown that there is a reasonable probability that a more favorable result would have obtained in a competently run trial. However, for claims relating to penal consequences, a petitioner must establish, by objective facts, circumstances that support the conclusion that counsel’s errors in advice as to penal consequences were material to the decision to plead. Merely alleging that the petitioner would not have pleaded is insufficient. Rather, specific facts, in addition to the petitioner’s conclusory allegation, must establish an objective reasonable probability that competent representation would have caused the petitioner not to enter a plea. See Toro v. Fairman, 940 F.2d 1065, 1068 (7th Cir.1991) (requiring objective evidence that a plea would have [508]*508been accepted); see also United States v. Gordon, 156 F.3d 376, 380-81 (2d Cir.1998) (affirming finding that the disparity between the sentence exposure represented by the attorney and the actual maximum sentence was objective evidence of prejudice, i.e., that defendant had rejected a beneficial plea agreement based on the erroneous advice). This case does not meet that standard. Segura offers nothing more than the naked allegation that his decision to plead would have been affected by counsel’s advice.
Conclusion
The judgment of the postconviction court is affirmed.
DICKSON, J., and RUCKER, J., concur.
SULLIVAN, J., concurs in result with separate opinion in which SHEPARD, C.J., concurs.