dissenting.
Because appellate counsel provided ineffective assistance by failing to challenge Michael Miller’s 120-year prison sentence, I would conditionally grant the writ. The majority concludes that Miller fails to show that the state judiciary made any error of federal law in rejecting this claim. For the reasons that follow, I respectfully dissent.
I.
A jury convicted Miller of -three counts of child molestation, and the trial court sentenced him to 40 years in prison on each count running consecutively for a to-' tal of 120 years.1 On direct appeal, Mil*280ler’s counsel raised two nigh frivolous challenges — to the sufficiency of the evidence and the admission of other-acts evidence; he also noted a clerical error in the written judgment, which did not affect the length of Miller’s sentence. Counsel failed to raise a challenge to Miller’s virtual life sentence, .despite the fact that Miller asked him to and counsel in a letter to Miller said “it probably will be” one of the issues he, raised. The Indiana court of appeals remanded for correction of the clerical error but quickly rejected the evidentiary challenges.
Miller then petitioned for post-conviction relief in state court, arguing, inter alia, that his appellate counsel provided ineffective assistance. The state trial court held a hearing, at which appellate counsel testified he did not have any independent recollection of or knowledge- why he did not raise the sentencing issue on direct appeal. The trial court denied relief, and the Indiana court of appeals affirmed.
The state court of appeals found that counsel should have raised the sentencing issue on direct appeal and thus considered whether there was a reasonable probability that the outcome of the appeal would have been different had he done so. In analyzing this issue, the court distinguished two cases “available as precedent during Miller’s direct appeal,” opining that Miller’s case was more aggravated. The court noted that Miller acted .as a father figure to the victim and found the nature of the offense particularly contemptible because Miller had significant time to reflect upon the heinous nature of his actions between the dates on which the crimes occurred. The court also found that Miller’s prior criminal history, which included three Class D felony drug convictions and a Class A misdemeanor conviction, and the fact that he molested the victim over a six-year time span demonstrated his inability to lead a law abiding life and his depraved character. The court concluded:
that Miller has not established that his 120-year aggregate sentence is inappropriate in light of the nature of the offense and the character ofithe offender. Consequently, Miller cannot establish that there is a reasonable probability that his sentence would have been revised pursuant to Appellate Rule 7(B) if appellate counsel had raised the issue on direct appeal. We therefore-affirm the ■ trial court’s finding that Miller’s appellate counsel was not ineffective.
Miller v. State, 2013 Ind.App. Unpub. Lexis 377, at *19-20 (Ind.Ct.App. Mar. 26, 2013) (internal citations omitted).
Miller then sought habeas relief in federal court. The district court denied his petition, but we granted a ■ certificate of appealability on the issue of whether appellate counsel provided ineffective assistance and sua sponte appointed counsel to represent Miller on this appeal.
II.
Because his case is governed by the Antiterrorism and Effective Death Penalty Act (“AEDPA”), in order to obtain relief Miller must show that the decision of the last state, court to address his claim on the merits was contrary to, or involved -an unreasonable application of, clearly estab*281lished Federal law, as determined by the Supreme Court of the United States. See 28 U.S.C. § 2254(d). The standard is a demanding one, but I believe that Miller satisfies it.
In order to establish a claim of ineffective assistance, a defendant generally must show (1) that counsel’s performance was deficient, and (2) that the deficient performance prejudiced the defense. Strickland v. Washington, 466 U.S. 668, 687, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). The present case involves the performance of counsel on direct appeal. An appellate lawyer performs deficiently if he' abandons a non-frivolous claim that was both 'obvious and clearly stronger than the claims he actually presented. Shaw v. Wilson, 721 F.3d 908, 915 (7th Cir.2013). The court evaluates performance from the perspective of a'reasonable attorney at the time of the appeal, avoiding the distorting effects of hindsight. Id.
To determine prejudice in this context, the court asks whether there is a, reasonable probability that, but for counsel’s unprofessional errors, the result of the direct appeal would have been different. Id. at 918. As this court stated in a similar case involving appellate counsel’s failure to raise an issue of Indiana law on direct appeal:
In assessing prejudice, we must bear in mind once again that we are making a comparative inquiry about counsel’s choices; we are not resolving any issue of state law, and we are not telling the Indiana judiciary how it should approach this issue. Prejudice exists, .however, if counsel bypassed [a] non-frivolous argument that, if successful, would have resulted in the vacation of [the petitioner’s] conviction____ And when evaluating prejudice, • unlike when evaluating attorney performance,' hindsight is permissible. Lockhart v. Fretwell, 506 U.S. 364, 372, 113 S.Ct. 838, 122 L.Ed.2d 180 (1993); Hemstreet v. Greiner, 491 F.3d 84, 91 (2d Cir.2007); Eddmonds v. Peters, 93 F.3d 1307, 1326 n. 5 (7th Cir.1996). This means that the Indiana Supreme Court’s [later decisions are] relevant to whether the argument - [counsel] jettisoned was both nonfrlvolous-and stronger than the sufficiency argument he presented.
Id.
There can be little doubt that counsel performed deficiently by failing to raise the sentencing issue on Miller’s direct appeal. The evidentiary challenges counsel raised were all but . doomed to fail, given the standards of review Indiana appellate courts apply, and the clerical correction did Miller no good.
On the other hand, Indiana appellate courts are authorized to independently “review and revise” sentences. Ind. Const. Art. 7, § 4; Pierce v. State, 949 N.E.2d 349, 352 (Ind.2011). This authority is implemented through Indiana Appellate Rule 7(B), which provides that the appellate court may revise a sentence if after due consideration of the trial court’s decision the appellate court finds the sentence is inappropriate in light,of the nature of the offense and the character of the offender. Pierce, 949 N.E.2d at 352. As Miller shows in his brief, Indiana appellate courts have not hesitated to use this authority; he cites no less -than 11 cases in which Indiana appellate courts shortened sentences in similar cases.2
*282The Indiana court of appeals found deficient performance in this case. The warden does not argue otherwise in this court. The issue is thus whether the state court’s prejudice finding was contrary to, or involved an unreasonable application of, federal law as determined by the Supreme Court.
The Indiana court correctly set forth the standards for evaluating ineffective assistance of appellate counsel claims. The court cited state rather than federal decisions, but there is nothing wrong with that (so long as the state cases do not contradict federal law as set forth by the Supreme Court). See, e.g., Early v. Packer, 537 U.S. 3, 8, 123 S.Ct. 362, 154 L.Ed.2d 263 (2002).
As Miller notes, however, the state court limited its analysis to cases pre-dating Miller’s direct appeal, ignoring later cases cited by the parties in which Indiana appellate courts revised sentences in similar cases. Miller argues that this is “contrary to” the holding of Lockhart v. Fretwell that, while the performance prong is evaluated without the benefit of hindsight, prejudice is measured based on the law as it exists at the time the court adjudicates the ineffective assistance claim. See 506 U.S. at 372, 113 S.Ct. 838; see also Shaw, 721 F.3d at 918 (“[W]hen evaluating prejudice, unlike when evaluating attorney performance, hindsight is permissible.”).
While the state court discussed only cases available at the time of Miller’s direct appeal, it did not explicitly state that later cases were irrelevant as a matter of law; as indicated, the court set forth the correct standards earlier in its decision. We will not grant a writ simply because the state court failed to cite all of the pertinent cases.3
*283However, I would hold that the Indiana court of appeals did unreasonably apply federal law. As indicated above,.the court stated:
that Miller has not established that his 120-year aggregate sentence is inappropriate in light of the nature of the offense and the character of the offender. Consequently, Miller cannot establish that there is a reasonable probability that his sentence would have been revised pursuant to.Appellate Rule 7(B) if appellate counsel had raised the issue on direct appeal.
The conclusion in the second sentence does not necessarily follow from- the premise in the first sentence. A court considering an ineffective assistance claim need not definitively resolve in the defendant’s favor the merits of the arguments counsel omitted, “for under Strickland [the defendant] need show only a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” United States v. Weathers, 493 F.3d 229, 238 (D.C.Cir.2007) (internal quote marks omitted, emphasis in original); see also Shaw, 721 F.3d at 918 (rejecting the respondent’s argument that the petitioner’s claim would have failed under state law because his “theory does not turn on the ultimate outcome in the state courts;, it depends only on the relative strength of this argument over the one counsel chose”).
Given the numerous cases in which Indiana appellate courts revised similar sentences, there is at least a reasonable chance Miller may have obtained such relief on direct appeal. That this panel of the Indiana court of appeals found the sentence appropriate on post-conviction review does not foreclose the possibility that the Indiana supreme court (or the different pánel of the court of appeals that heard Miller’s direct appeal)4 would have exercised independent authority differently.5 Indeed, had counsel raised the sentencing issue on direct appeal, there is every reason to believe that Miller’s case would have been on the list set forth in note 2, above.
*284As Miller shows in his brief, the state supreme court has on multiple occasions reduced sentences in cases arguably more aggravated than his. In Serino v. State, for instance, a jury found the defendant guilty of 26 counts of child molesting and sexual misconduct with a minor.6 The defendant in that case became a father figure to the victim, his girlfriend’s son, but then repeatedly sexually abused the boy over a three year period, beginning when the victim was 11, including fondling, oral sex, and ,anal sex. 798 N.E.2d at 853. In imposing a 385-year sentence, the trial court noted that Serino was in a position of trust with the victim and exploited that trust, that Serino was not charged with all crimes committed against the child, and that Serino had other pending charges for sex crimes against another child, in co‘n7 eluding that a lesser sentence would depreciate the seriousness of the crimes. Id. at 857. Based primarily on the testimony from numerous witnesses attesting to Seri-no’s positive character traits, the supreme court reduced the sentence to 90 years. Id. at 858.
Miller also occupied a position of trust with the victim, and his crimes also occurred over a several year period. However, Miller was convicted of just three counts, he had no other charges pending, and the record discloses no uncharged conduct related to the same victim. As discussed in note 1, supra, Miller also produced numerous positive character witnesses at his sentencing. Yet he is serving a sentence 30 years longer than Seri-no.
In Smith v. State, a jury convicted the defendant of four counts of child molesting, based on his step-daughter’s report that' he had sexual intercourse with-her when she was 10 years old and ón three other occasions before she turned 14. 889 N.E.2d at 262. The trial court imposed four consecutive sentences of 30-years, 120 years total, based on Smith’s extensive criminal history, including two sex offenses; the extended period of time over which Smith molested the victim; the heinous violation of trust represented by Smith’s abuse of his step-daughter; and the additiohal psychological abuse he inflicted on the girl. Id. at 263-64. Despite these aggravated circumstan'ces, the Indiana ■ supreme court reduced the.'Sentence to a- total of 60 years. Id. at 264. Miller sits in prison for twice as long, despite the absence of prior sex offenses on his record or any evidence of additional psychological abuse of the victim.
And in Sanchez v. State, the defendant sexually abused his step-daughters, ages six and nine, resulting in three child molestation convictions. 938 N.E.2d at 721. The trial court imposed, concurrent sentences of 40 years for the two counts pertaining to the first child, and a consecutive sentence of 40 years for the count pertaining to the second child. Id. The state supreme court, noting that Sanchez did not use significant force on the girls or cause injury, and that he lacked an extensive prior record, made the sentences concurrent, reducing the total term to 40 years. Id. at 723. Miller was convicted of three counts involving the same victim, who testified that Miller riever threatened him or inflicted any injury (other than temporary pain from the intercourse). Yet Miller’s sentence is three times as- long as Sanchez’s.7
*285Finally, while perhaps not an independent basis for granting relief, the Indiana court’s rejection of Miller’s sentencing claim is weakened by its failure to consider post-2004 cases. As discussed above, in January 2003 the Indiana supreme court made it easier for appellate-courts to revise sentences.8 It thus stands to reason that there would be more case-law favorable to Miller’s position after 2003 than before. The warden argues that, because the text of the rule did not change, the court’s analysis was the same in 2013 as it would have been in 2004. But given-the vagueness of the rule surely the Indiana supreme court’s application of it in similar cases, is important. Those cases offer insight into what might have happened had counsel challenged Miller’s sentencé on direct appeal. It is true, as the warden notes, that the Indiana court of appeals used the present tense — “Miller has npt established that his 120-year aggregate sentence is inappropriate,” Miller, 2013 Ind.App. Unpub. Lexis 377, at *19 (emphasis added.) But this shows only that this panel found the sentence appropriate; it does not show that another panel of the court or the Indiana supreme court would not have revised Miller’s sentence on direct appeal. Nor does the tense the court used change the fact that it ignored post-2003 case-law.
III.
The majority says that the state court’s decision was not based on federal law at all; rather, it rests on the conclusion that, as a matter of state law, an appeal challenging the sentence would have been futile. A federal court cannot disagree with a state court’s resolution of an issue of state law.
We addressed a similar issue in Shaw, a case which also involved counsel’s failure to raise an issue of Indiana state law on direct appeal. In Shaw, the state amended the information after the deadline set by the then-applicable procedural rule (Section 35-34-1-5). 721 F.3d at 911. Although no Indiana appellate court ever had invalidated an amendment under that rule, the Indiana supreme court had stated that tardy amendments, if substantive in nature, were impermissible. Id. at 912 (citing Haak v. Indiana, 695 N.E.2d 944, 951 (Ind.1998)). Trial counsel challenged the amendment, but appellate counsel failed to renew the challenge on direct appeal, instead raising a near frivolous challenge to the sufficiency of the evidence. Id. Four years after Shaw lost his direct appeal, the Indiana supreme court reversed a conviction based on an untimely substantive amendment, confirming what it had said in Haak. Id. (citing Fajardo v. Indiana, 859 N.E.2d 1201 (Ind.2007)).9 Shaw then sought post-conviction relief, arguing that his appellate lawyer provided ineffective assistance by not raising the amendment issue. The Indiana court of appeals denied relief, noting that while some decisions of the state supreme court included dicta supporting Shaw, prior to Fajardo no amendment had actually been invalidated under the rule. Given the state of the law, the court found that counsels performarice was not deficient. Id. at 913. The court *286also found no prejudice because Shaw had been granted a continuance to prepare for trial. Id.
When Shaw sought habeas relief, the state argued that the federal courts were entirely prohibited from evaluating the Indiana court of appeals’ assessment of Shaw’s claim because the claim involved a question of state law. Id. at 914. We rejected that contention:
The state’s argument, however, misses the point that Shaw is making. Shaw is not asking (and has no reason to ask) that we second-guess an Indiana court on the meaning of Section 35-34-1-5. Shaw is making a simpler point: a competent lawyer in Indiana should have recognized that there was a state statute under which relief for his client was possible and would have pursued that theory on appeal. An argument about the validity of the state’s effort to amend the indictment would have been materially stronger than the frivolous sufficiency-of-the-evidence point that [counsel] raised. With that much accepted, there is no further role for the federal judiciary: whether the Indiana appellate court would have been persuaded, or if not, whether the Indiana Supreme Court would have granted transfer, is immaterial. The state’s argument that even this kind of comparative assessment is out of bounds, if accepted, would foreclose federal review of almost any ineffectiveness claim that rests on an attorney’s mishandling of a state-law issue, no matter how egregiously deficient the attorney’s performance. It is well established that a defense attorney’s failure to raise a state-law issue can constitute ineffectiveness.
Id. at 914-15,
The majority indicates that Shaw permits an ineffective assistance claim where counsel’s error prevented the state judiciary from recognizing the force of a potential state law argument. Here, according to the majority, appellate counsel’s mistake did not prevent the Indiana judiciary from looking at Miller’s sentence; the Indiana court of appeals did so on post-conviction review, concluding that a possible review of the sentence in 2004 would have done him no good — as a matter of state law. For two reasons, this misses the mark.
First, because habeas petitioners must exhaust their state court remedies before turning to the federal courts, it will often be the case that the state post-conviction court addresses the merits of the argument counsel omitted. Indeed, the Indiana court of appeals did so in Shaw, 721 F.3d at 913.10 We granted habeas relief, despite the fact that the omitted claim may have failed as a matter of state law; we declined to be drawn into the content of state law, noting that “it is necessary only to conclude that the amendment issue was clearly stronger than the sufficiency argument.” Id. at 916. We said the same thing assessing prejudice:
[W]e are making a comparative inquiry about counsel’s choices; we are not resolving any issue of state law, and we are not telling the Indiana judiciary how it should approach this issue. Prejudice exists, however, if counsel bypassed an nonfrivolous argument that, if successful, would have resulted in the vacation *287of Shaw’s conviction (just as the conviction in Fajardo later was).
Id. at 918.
Second, the majority makes the same mistake as the Indiana court of appeals, equating one panel’s discretionary rejection of Miller’s sentencing claim (based solely on pre-2004 case-law), with a finding that there is no reasonable probability that the state supreme court or another panel of the court of appeals would have modified the sentence on direct appeal. As indicated above, review under Rule 7(B) involves “an exercise of judgment that is unlike the usual appellate process.” Cardwell, 895 N.E.2d at 1223. The Indiana supreme court has explained that:
whether we regard a sentence as appropriate at the end of the day turns on our sense of the culpability of .the defendant, the severity of the crime, the damage done to others, and myriad other factors that come to light in a given case. Individual judgments as- to the proper balance to be struck among these considerations will necessarily vary from person to person, and judges, whether they sit on trial or appellate benches, are no exception. There is thus no right answer as to the proper sentence in any given case. As a result, the-role of an appellate court in reviewing a sentence is unlike its role in reviewing an appeal for legal error or sufficiency of evidence.
Id. at 1224. Because there is no “right” or “wrong” answer under Appellate Rule 7(B), it is incorrect to say, as the majority does, that the Indiana court of appeals definitively resolved the issue as a matter of state law in finding the sentence appropriate. See id. at 1225 (explaining that the goal of appellate review is “not to achieve a perceived ‘correct’ result in each case”).11
The majority notes that Miller cites no Supreme, Court decision holding that a state, must give a petitioner the benefit of state law precedent that comes after his conviction became final. This, too, misses the mark. _ I agree that, while it would have been permissible for the Indiana court of appeals to cite post-2004 cases, no rule of federal law required it to do so. But Miller’s point is that those cases are available' to us in evaluating the state court’s prejudice determination. Considering later cases does not, as the majority suggests, effectively require that all state law decisions apply retroactively. The operative rule in Indiana was the same in 2004 as in 2013; Miller does not seek benefit of a new, more lenient standard, he simply directs our attention to cases applying the rule which support his position that there is a reasonable probability his sentence would have been modified had his lawyer raised the issue.-
The majority contends that Lockhart does not entitle Miller to the benefit of hindsight. While it is true that Lockhart, a pre-AEDPA case, applied case-law postdating the petitioner’s direct appeal to deny habeas relief, we have not understood the case to be limited to those circumstances. See Shaw, 721 F.3d at 918. In Shaw, for .instance, we considered the Indiana supreme court’s post-direct-appeal decision in Fajardo in evaluating prejudice. Id. The majority correctly notes that Lockhart did not change Strickland’s prejudice component, see note 3, swpra, but the Supreme Court’s later decisions do not,'as the majority suggests, understand Lockhart as an cmii-hindsight decision, Maryland v. Kulbicki, — U.S. -, 136 S.Ct. 2, 4, — L.Ed.2d - (2015), cited *288by the majority, quoted Lockhart in assessing performance, not prejudice. The Court explicitly declined to address prejudice in that case. Id. at 5.
IV.
For these reasons, I would vacate the district court’s decision and remand with instructions to grant the writ unless Indiana affords Miller a new appeal.
. The majority notes that the sexual abuse, including anal intercourse, began when the victim was nine and continued for six years, and that the state court in imposing an effective life term considered not only the nature of Miller’s conduct but also his four prior convictions, his failure to reform after stretches of imprisonment, and the absence of mitigating factors. While it isn't necessary to address the facts at length, I do note that the victim in this case came forward years after the alleged abuse, first disclosing it during a psychiatric hospitalization. During Miller’s trial, the victim testified that he was drunk and high on marijuana during the alleged assaults. He did not know how old he was when the assaults occurred (although he did. know he was under the age of 14, an element of the charges). He was not sure of the dates of, or the. addresses where, -the assaults took place. The “only thing that [he was] a hundred percent sure of is that man [Miller] had *280sex With [him].'' At the' sentencing hearing, Miller presented testimony from nuiherous witnesses as to his good character. Several of the witnesses had children of their own; none expressed concern about Miller acting inappropriately. One witness indicated that Miller had taken her 15-year-old son under his wing, turning his life around. "'I can't thank Mike enough for what he’s done for my son.” The victim, apparently incarcerated at thé time of Miller’s sentencing, did not appear to make a statement. Finally, while Miller had prior felony convictions, they were for drug offenses, not sexual misconduct.
. Pierce v. State, 949 N.E.2d 349 (Ind.2011) (revising 124 year sentence on four counts of child molestation to 80 years); Sanchez v. State, 938 N.E.2d 720 (Ind.2010) (revising total sentence of 80 years on three counts of child molestation to 40 years); Harris v. State, 897 N.E.2d 927 (Ind.2008) (revising consecutive sentences of 50 years on two counts of *282child molesting to concurrent); Smith v. State, 889 N.E.2d 261 (Ind.2008) (revising four consecutive sentences of 30 years each, a total of 120 years, to a total of 60 years); Monroe v. State, 886 N.E,2d 578 (Ind.2008) (reducing sentence of 100 years to 50 years); Estes v. State, 827 N.E.2d 27 (Ind.2005) (revising sentence of 267 years on 14 counts of child molesting and sexual misconduct with a minor to 120 years); Serino v. State, 798 N.E.2d 852 (Ind.2003) (revising sentence of 385 years on 26 counts of child molestation to 90 years); Kien v. State, 782 N.E.2d 398 (Ind.Ct.App.2003) (revising consecutive sentences of 40 years on three counts, a total of 120 years, to 80 years total); Ortiz v. State, 766 N.E.2d 370 (Ind.2002) (revising 30 year consecutive sentences on child molesting counts to run concurrently); Haycraft v. State, 760 N.E.2d 203 (Ind.Ct.App.2001) (revising 190 year sentence for child molesting and related offenses to 150 years); Walker v. State, 747 N.E.2d 536 (Ind.2001) (revising consecutive sentences of 40 years on two counts of child molesting to be concurrent). As discussed below, one of the issues in this case is whether the Indiana court of appeals erred by failing to consider cases post-dating Miller's direct appeal. Five of the cases cited above pre-date the decision on Miller’s direct appeal; six post-date the direct appeal. It is also important to note that in January 2003 the Indiana supreme court modified the standard for revising a sentence; formerly, appellate courts could revise a sentence only if it was "manifestly unreasonable.” Serino, 798 N.E.2d at 856. Finding this barrier to review too high, the court modified the rule to allow revision “if, after due consideration of the trial court's decision, the Court finds that the sentence is inappropriate in light of the nature of the offense and the character of the offender." Id. (quoting Ind. Appellate Rule 7(B)). This more lenient standard applied at the time of Miller's direct appeal, but there were a limited number of cases applying it at that time.
. The warden argues that Lockhart v. Fretwell has no applicability here because that case is limited to situations in which the defendant would receive a windfall if the court did not consider subsequent legal developments. In Lockhart, counsel failed to raise an objection supported by case-law at the time of sentencing; however, by the time the defendant sought habeas relief, the favorable case-law had been overruled. Under these circumstances, the Supreme Court refused to focus *283"on mere outcome determination, without attention to whether the result of the proceeding was fundamentally unfair or unreliable.” 506 U.S. at 369, 113 S.Ct. 838. Lockhart did not supplant the general prejudice rule set forth in Strickland, e.g., a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different. Williams v. Taylor, 529 U.S. 362, 391-93, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000). Lockhart’s rule that the court also consider • unfairness only comes into play in the unusual circumstance where the defendant attempts to demonstrate prejudice based on considerations that, as a matter of law, ought not inform the inquiry, id. at 393 n. 18, 120 S.Ct. 1495, such as later- rejected case-law, Shaw, 721 F.3d at 919. The warden contends that the present case does not involve a change in the law. However, after Lockhart, courts have held that prejudice is determined with the benefit of hindsight in all cases, whether focused on the outcome or unfairness. See, e.g., Lynch v. Dolce, 789 F.3d 303, 311 (2d Cir.2015); Eddmonds, 93 F.3d at 1326 n. 5. The warden cites no authority to the. contrary. Thus, this aspect of Lockhart applies in the present case.
. The Indiana court of appeals has 15 members. The court’s three-judge panels sit together for terms of four months, after which the judges rotate, http://www.in.gov/judiciary/ appeals/2Í347.htm (last visited Apr. 21, 2016). None of the judges who heard Miller's direct appeal served on the panel that decided his post-conviction appeal.
. It is important to note that appellate review under Rule 7(B) involves "an exercise of judgment that is unlike the usual appellate process, and is very similar to the trial court’s [exercise of discretion].” Cardwell v. State, 895 N.E.2d 1219, 1223 (Ind.2008). The Indiana supreme court has described Rule 7(B) as “a standard leaving much to the unconstrained judgment of the appellate court.” Id. at 1224.
. Serino was decided on November 19, 2003, nearly seven months before the court of appeals decided Miller’s direct appeal.
. I do not cite these cases to show that the Indiana court of appeals erroneously exercised its discretion under Appellate Rule 7(B) in Miller’s case., The only issue we need to *285decide is whether Miller has shown, a reasonable probability, a better than negligible chance, see Harris v. Thompson, 698 F.3d 609, 644 (7th Cir.2012), of a different outcome. Because the Indiana supreme court has reduced sentences in cases arguably more aggravated than Miller’s, I find that Miller has made the required showing.
. The Indiana court of appeals acknowledged the changed standard, and that it applied to Miller’s sentence.
. The Indiana legislature later changed the rule in response to Fajardo Id. at 912-13.
. Specifically, the Indiana court of appeals noted the absence of authority supporting a challenge to the amendment at the time of the direct appeal. Because state law did not support the claim at the time, the court found that counsel’s performance was not deficient. Id.
. It would be different if Miller had argued, say, that Indiana law prohibited consecutive sentences in cases like his, and the court of appeals rejected that contention under state law