dissenting.
As I believe that King has sufficiently exhausted his state remedies, I must respectfully dissent. Accordingly, I also reach the merits of his habeas petition. First, though, I must address the majority’s contention that King’s arguments before the state and federal courts are irreconcilable.
As codified in 28 U.S.C. § 2254(b) and (c), the exhaustion doctrine requires that federal claims be first presented to the state court such that it has a full and fair opportunity to resolve them. Rose v. Lundy, 455 U.S. 509, 518, 102 S.Ct. 1198, 71 L.Ed.2d 379 (1982). In order to exhaust his or her state remedies, a petitioner must have fairly presented to the state court the same claim, or its substantial equivalent, that is the basis for the habeas petition. Picard v. Connor, 404 U.S. 270, 275, 92 S.Ct. 509, 30 L.Ed.2d 438 (1971). Generally, “exhaustion determinations of this genre are by their very nature case-specific.” Nadworny v. Fair, 872 F.2d 1093, 1095 (1st Cir.1989).
To “fairly” present a claim to a federal court, a habeas petitioner must first present the substance of his constitutional claim to the state court. Jells v. Mitchell, 538 F.3d 478, 504 (6th Cir.2008). We allow “variations in the legal theory or factual allegations urged in its support,” so long as the ultimate question between the state and federal courts is the same. Id. We have established four guideposts which can help navigate a determination as to whether a claim has been fairly presented: (1) reliance upon federal cases employing constitutional analysis; (2) reliance upon state cases employing federal constitutional analysis; (3) phrasing the claim in terms of constitutional law or in terms sufficiently particular to allege a denial of a specific constitutional right; or (4) alleging facts well within the mainstream of constitutional law. McMeans v. Brigano, 228 F.3d 674, 681 (6th Cir.2000). These guideposts are derived from the seminal case Daye v. Attorney General, 696 F.2d 186 (2d Cir.1982), where the Second Circuit explained that “[b]y the same legal ‘basis’ or ‘doctrine,’ we do not mean that there can be no substantial difference in the legal theory advanced to explain an alleged deviation from constitutional precepts.” Id. at 192 n. 4. Indeed, “all that is needed to alert the state courts to the constitutional nature of the claim is the exposition of the material *967facts” and the specific constitutional precept. Id.
Superficially, King’s claim to the Michigan courts was that the prosecutor relayed a promise that consecutive sentencing would not apply, which was left unfulfilled. He contended that this was a violation of due process. His argument continues that it was a due process violation because he was induced to accept a plea agreement on the basis of a grievous misrepresentation. He could not have known the consequences of his plea agreement if the contents of the agreement were misrepresented to him.
Whether King refers to it as a promise or a misrepresentation is inconsequential. The majority’s strained adherence to King’s usage of the term “unfulfilled promise of leniency” eclipses the substance of King’s claim. During the state proceedings, King’s assertion that “he pled guilty under the apprehension that his sentence would [] run concurrent with any term imposed for violation of his parole” precedes two paragraphs of federal and state case law which hold that a failure to fully understand the consequences of a plea generates an involuntary and unknowing plea in violation of a petitioner’s Due Process rights.1 The “substance” of the claim has remained the same throughout the proceedings under both state and federal proceedings — indeed, even a cursory review of the record shows that the substance of the claim was unearthed at the motion hearings to withdraw King’s plea.2 Accordingly, the majority’s concerns regarding federal-state comity are without merit.
The majority distinguishes Boykin v. Alabama’s requirement that a plea must be entered knowingly, intelligently, and voluntarily, 395 U.S. 238, 242-44, 89 S.Ct. 1709, 23 L.Ed.2d 274 (1967) and Santobello v. New York’s holding that, “[t]he plea must, of course, be voluntary and knowing and if it was induced by promises, the essence of those promises must in some way be made known.” 404 U.S. 257, 261, 92 S.Ct. 495, 30 L.Ed.2d 427 (1971). However, King’s claim is — and was — broader than just Santobello. I begin first by observing that, in his Brief in Support to Withdraw his Guilty Plea, King relies on multiple cases which stand for the broader proposition that an involuntary plea runs afoul of due process.3 Specifically, King cites Brady v. United States, 397 U.S. 742, 755, 90 S.Ct. 1463, 25 L.Ed.2d 747 (1970), for the proposition that a voluntary plea only results where the defendant is fully aware of the consequences of his plea, including the actual value of commitments made by the prosecutor. See Brief in Support of Motion to Withdraw Plea at 3-4. That King alleges that the prosecutor’s false assurances are the result of misunderstanding the consequences of his plea is an important, but not dispositive element *968of the exhaustion inquiry. In substance, King’s argument has been, and still is, the following: he was assured one thing and received another, entering into a guilty plea that precipitated consequences of which he was unaware.
Accordingly, I find that, although the federal appellate and state briefs are not identical, the state brief incorporates sufficient indicia of the Boykin requirements such that, although there is a “variation in legal theory,” 4 the substance of his claim was presented to the state court. Jells, 538 F.3d at 504. The record clearly reflects that King alerted the Michigan courts to the constitutional nature of his claim, and that the ultimate issue — whether the prosecutor’s misrepresentations as to the consequences of his plea would violate his due process rights — was squarely before both courts. Id. Because King phrased him claim in clear terms of constitutional law, as well as alleged facts well within the mainstream of constitutional law, he fairly presented his claim. See Brigano, 228 F.3d at 681. Thus, whether the argument at the state level is labeled “an unfulfilled promise of leniency,” is irrelevant. See Jackson v. Edwards, 404 F.3d 612, 619 (2d Cir.2005) (“Although the federal principle may initially be attached to a different label than the one ultimately affixed in federal habeas proceeding,” as long as the substance of the argument is clearly raised in state court, the exhaustion doctrine is satisfied); Koontz v. Glossa, 731 F.2d 365, 369 (6th Cir.1983) (same). Having fulfilled his burden to show exhaustion, Caver v. Straub, 349 F.3d 340, 345 (6th Cir.2003), I will now address the merits of King’s case.
A. 28 U.S.C. § 2251(d)
Because King appeals from a state court proceeding, his petition is subject to AED-PA and the “Supreme Court opinions interpreting it.” Jones v. Bagley, 696 F.3d 475, 482 (6th Cir.2012). As amended by AEDPA, 28 U.S.C. § 2254, the governing statute, “sets several limits on the power of a federal court to grant an application for a writ of habeas corpus on behalf of a state prisoner.” Cullen v. Pinholster, — U.S. —, 131 S.Ct. 1388, 1398, 179 L.Ed.2d 557 (2011). Two provisions of § 2254 are particularly relevant to King’s petition: § 2254(a), which permits a federal court to entertain only those applications alleging that a person is in state custody “in violation of the Constitution or laws or treaties of the United States,” and § 2254(d), which governs the types of procedural errors made by a state court that will allow a petitioner to proceed.5
*969The § 2254(d) standard is “highly deferential.” Pinholster, 131 S.Ct. at 1398. Specifically, the statute states that a district court shall not grant a habeas petition with respect to any claim that was adjudicated on the merits in the state courts unless the adjudication resulted in a decision that: (1) was contrary to, or involved an unreasonable application of, clearly established federal law as determined by the Supreme Court; or (2) was based on an unreasonable determination of the facts in light of the evidence presented to the state courts. 28 U.S.C. § 2254(d). On appeal, King asserts that the decision of the reviewing court was both (a) contrary to established law, and, alternatively, (b) involved an unreasonable application of such law. I find his second argument more meritorious.6 Where, as here, the petitioner alleges that the state court arrived at a decision on the basis of a misapplication of a legal standard, that petitioner must search for his redress through the aperture of the “unreasonable application” prong. To satisfy this standard, the petitioner must show that the state court correctly identified the governing legal rule but applied it unreasonably to the facts of his case, or that it either unreasonably extended or unreasonably refused to extend a legal principle from Supreme Court precedent to a new context. 28 U.S.C. § 2254(d); Sanborn v. Parker, 629 F.3d 554, 564 (6th Cir.2010).
King argues that the state court’s “complete failure to consider [] King’s understanding of the consequences of his plea,” was an unreasonable application of Boykin and its progeny. King focuses on the fact that the judge failed to engage him in an inquiry as to whether he knew the true consequences of his plea. Rather, King asserts that the judge used an objective standard in concluding that he should have known, as it would have been “common knowledge” that the sentences would run concurrent to one another. As evidence that it was unreasonable to apply such a standard, King argues that the law requires a careful and subjective examination that takes into account, among other things, any promises on which a petitioner may rely. King asserts that the judge ignored the fact that he could not possibly have accounted for the extra thirty year addition to his maximum sentence, on the grounds that he was ensured that consecutive sentencing would not apply. Moreover, King alleges that the effects of the Michigan sentencing scheme which compels this result was ignored at his hearings, relying on the fact that the motion judge interpreted the issue to be a request for time served.
Whereas the touchstone analysis as to whether a plea is knowing and voluntary turns on a defendant’s understanding of the plea consequences, see Schneckloth v. Bustamonte, 412 U.S. 218, 238, 93 S.Ct. 2041, 36 L.Ed.2d 854 (1973), the state court in this case focused on what a person situated similarly would have understood. Therefore, the real issue in this case is whether it was unreasonable for the state court to frame the terms of the Boykin inquiry in the manner that it did. In other words, in a Boykin analysis, does a court assess the voluntariness and intelligence of a plea based on an objective or subjective standard? And, if the court applied the incorrect standard, did it do so unreasonably?
Before accepting a plea, the court must ensure that the defendant has “a full un*970derstanding of [its] ... possible consequences .... ” Brady, 397 U.S. at 748 n. 6, 90 S.Ct. 1463. A plea will be rendered involuntary if the defendant is not “fully aware of the direct consequences, including the actual value of any commitments made to him by the court, prosecutor, or his own counsel.” Id. at 756, 90 S.Ct. 1463 (quotation omitted). “Guilty pleas must be carefully scrutinized to determine whether the accused knew and understood all the rights to which he would be entitled at trial, and that he had intentionally chosen to forgo them.” Schneckloth, 412 U.S. at 238, 93 S.Ct. 2041. The very purpose of the knowing and voluntary inquiry is to determine whether the defendant “actually does understand the significance of a particular decision.... ” Godinez v. Moran, 509 U.S. 389, 401 n. 12, 113 S.Ct. 2680, 125 L.Ed.2d 321 (1993) (emphasis in original).
The case law indicates that the inquiry as to whether a guilty plea is knowing and voluntary is a “subjective, highly individualized test.” United States v. Powell, 269 F.3d 175, 185 (3d Cir.2001); see also Brown v. McKee, 340 Fed.Appx. 254, 257 (6th Cir.2009) (even subjective (misunderstanding as to consequences of guilty plea raises due process issues); Doe v. Woodford, 508 F.3d 563, 570 (9th Cir.2007) (subjective state of mind is relevant to determining voluntariness of plea). The inquiry must therefore be an extensive one, and must tailor to certain subjective factors.7
The record before the Court does not reflect that the court engaged King in a subjective or highly individualized inquiry. According to the court, it would have been “common knowledge” that the sentences “would have to be consecutive to the parole sentence.” Moreover, the court also framed the issue as whether the prosecutor’s statement could be fairly interpreted to mean that the robbery sentences would run consecutive to the underlying parole sentence. This fair interpretation was based on the face of the record and the plea sheet alone, bereft of scrutiny and investigation alike. It is incumbent on the court to carefully inquire as to what the defendant knows and expects about the consequences of his plea before the judge may accept it. No such careful scrutiny was executed here.
King contends that “the prosecutor misled him into believing that consecutive sentencing does not apply when in fact consecutive sentencing was mandatory under state law [and that] he entered his plea based in part on that false assurance.” King alleges that the promise of concurrent sentencing was an act of leniency— indeed to the amount of thirty years — and because it could not apply by law, it was a violation that rendered his plea involuntary. Specifically, King argues that his plea could not have been knowing or intelligent since his understanding was that all of the sentences were to run concurrently, *971not merely those derivative of the robbery. No discussion or inquiry as to the prosecutor’s promise or as to King’s understanding of his maximum release date occurred.
The lack of a searching inquiry may well be attributed to confusion at the motion hearings to withdraw King’s plea. Although King’s counsel informed the judge that his maximum release date was “6/18/2062,” the fact that the alteration of the maximum discharge date is a direct consequence of the consecutive running of the parole sentence was not expressly examined by the court. Again, because we combine all of the maximum sentences as required by Mich. Comp. Laws § 768.7a, including the sentence imposed for his parole conviction, this result is compelled. However, the motion judge interpreted the issue as one for credit for time served— indeed observing that they were “down to about a 90 day issue,” and that he did not want King to be “mad at everybody ... over 90 days.” Indeed, before ruling that “it would be common knowledge” that the parole sentence would run consecutive to the robbery sentences, the discussion focused on what the judge deemed to be a “request for credit.” However, the consequences to the sentence are much more dire than a 90 day issue. That the judge ignored the fact that the imposition of consecutive sentencing elongates King’s maximum release date by thirty years assures me that the court unreasonably erred.
That King was not asked to testify at either plea withdrawal hearing further assures me that the court was less concerned with King’s actual knowledge of his plea, and on what inducements he relied, and more focused on what the motion judge felt that a person in King’s shoes should have known. What this person should have known, however, is not the standard. The court did not supply any law or factual basis for the assertion that it “would be common knowledge that sentence would have to be consecutive to the parole sentence,” nor do I infer that its effect, namely to increase the possible maximum sentence by thirty years, would have been common knowledge.
Nor does a “fair interpretation” of “the record and plea sheet,” in this case, satisfy the requirement that King have “a full understanding of the ... possible consequences of his plea....” Brady, 397 U.S. at 748 n. 6, 90 S.Ct. 1463. In this case, the evidence suggests that King did not know that his sentence was to run consecutive to his parole sentence, and thus he could not possibly have known that his maximum sentence would end in 2062. The face of the record indicates that the recondite sentencing effects went unbeknownst to King. He did not — and could not — know because the prosecutor, twice, informed King that consecutive sentencing would not apply.8 Assessing the intelligence and voluntariness of King’s plea on the basis of what a reasonable person should have known, or what was an acceptable interpretation of the record, without inquiring as to what King actually knew, especially in light of the conflicting representations made, was an unreasonable application of federal law. See id.; Boykin, 395 U.S. at 242, 89 S.Ct. 1709.
B. 28 U.S.C. § 225í(a)
In finding that the state court’s decision was “an unreasonable application of ... *972clearly established federal law,” I would typically next conduct an independent review of the record under 28 U.S.C. § 2254(a). This inquiry would assess whether the defendant is “in custody in violation of the Constitution or laws or treaties of the United States.” West v. Bell, 550 F.3d 542, 553 (6th Cir.2008) (quoting 28 U.S.C. § 2254(a)). In my view, however, such a review would be improper on the limited record provided. Especially in light of Boykin’s requirement that the court is compelled to make its conclusion on the basis of the accused’s understanding of the consequences of his plea, I find that the most fitting solution would be to remand the case to the federal district court to hold an evidentiary hearing.
Because the unusual posture of this case presents issues as to whether an evidentia-ry hearing is even allowed on remand, I am compelled to comment on the ramifications of Cullen v. Pinholster, — U.S. —, 131 S.Ct. 1388, 179 L.Ed.2d 557 (2011). Pinholster bars a reviewing federal court from considering any new evidence when determining whether § 2254(d)(1) is satisfied. Id. at 1400 & n. 7. This result accords with the fact that “AEDPA’s statutory scheme is designed to strongly discourage” state prisoners from submitting new evidence in federal court. Id. at 1401.
Although federal courts must “turn a blind eye to new evidence in deciding whether a petitioner has satisfied § 2254(d)(l)’s threshold obstacle to federal habeas relief,” Id. at 1413 (Sotomayor, J., dissenting), once the threshold is cleared, the limitations almost disappear; the relevant provision shifts from § 2254(d) to § 2254(a). Stated differently, Pinholster will not allow a petitioner to introduce new evidence to show an unreasonable application of Supreme Court law under § 2254(d), but it will not preclude a habeas court from considering new evidence as to whether a petitioner is being held in custody in violation of his constitutional rights under § 2254(a). See Stokley v. Ryan, 659 F.3d 802, 808 (9th Cir.2011) cert. denied, — U.S. —, 133 S.Ct. 134, 184 L.Ed.2d 65 (2012) (holding that bar on new evidence is coterminous with the scope of § 2254(d)); Ballinger v. Prelesnik, 2:09-CV-13886, 2011 WL 5216277, at *1 (E.D.Mich. Nov. 2, 2011) (denying a motion for reconsideration of the court’s decision to grant an evidentiary hearing because it had “already determined that 2254(d)’s hurdle has been cleared” and a hearing was necessary to determine whether Petitioner’s constitutional rights were violated under § 2254(a)); Love v. Cate, 449 Fed.Appx. 570, 571 n. 1 (9th Cir.2011) (where a court has already found, under § 2254, an unreasonable application of federal law, a federal court may hold an evidentiary hearing to determine whether a prosecutor’s peremptory strike was racially motivated).
I acknowledge the proposition that “[m]ost times a finding in favor of the petitioner under 2254(d) means that a for-tiori petitioner is held in violation of his constitutional rights under 2254(a).” Bal-linger, 2011 WL 5216277, at *1 n. 1. However, this is not the case here. Justice Breyer, writing separately in Pinholster, described at which times a petitioner may have eluded § 2254(d), but, on those same facts, may not have established a constitutional violation under § 2254(a).
For example, if the state-court rejection assumed the habeas petitioner’s facts (deciding that, even if those facts were true, federal law was not violated), then (after finding the state court wrong on a(d) ground) an (e) hearing might be needed to determine whether the facts alleged were indeed true. Or if the *973state-court rejection rested on a state ground, which a federal habeas court found inadequate, then an (e) hearing might be needed to consider the petitioner’s (now unblocked) substantive federal claim. Or if the state-court rejection rested on only one of several related federal grounds (e.g., that counsel’s assistance was not “inadequate”), then, if the federal court found that the state court’s decision in respect to the ground it decided violated (d), an (e) hearing might be needed to consider other related parts of the whole constitutional claim (e.g., whether the counsel’s “inadequate” assistance was also prejudicial). There may be other situations in which an (e) hearing is needed as well.
Pinholster, 131 S.Ct. at 1412 (Breyer, J., concurring in part).
The case before us is one that requires such a hearing. Our “decision [with] respect to the ground [that the state court] violated (d),” is, in this case, insufficient— accordingly, “an (e) hearing might be needed to consider other related parts of the whole constitutional claim.” Id. While I am satisfied that the record is sufficient to make an assessment as to whether the state court judge unreasonably applied Boykin, I am not satisfied that it is sufficient to reach a conclusion with respect to the merits of King’s due process claim.
In deciding whether to grant an eviden-tiary hearing, a federal court must consider whether such a hearing could enable an applicant to prove the petition’s factual allegations, which, if true, would entitle the applicant to federal habeas relief. Schriro v. Landrigan, 550 U.S. 465, 474, 127 S.Ct. 1938, 167 L.Ed.2d 836 (2007). This Court has held that a habeas applicant is generally entitled to an evidentiary hearing if he “alleges sufficient grounds for release, relevant facts are in dispute, and the state courts did not hold a full and fair eviden-tiary hearing.” Sawyer v. Hofbauer, 299 F.3d 605, 610 (6th Cir.2002) (internal quotation marks and citation omitted).
King has made such a showing. Legitimate issues exist as to whether, with respect to his plea, King possessed a “full understanding of the ... possible consequences _” Brady, 397 U.S. at 748 n. 6, 90 S.Ct. 1463. King’s attestation that he relied on the prosecutor’s misrepresentation that consecutive sentencing would not apply allows a strong inference that he was unaware that his actual release date would be thirty years longer than he expected. This is not an immaterial oversight. Further investigation is needed to address King’s due process claim. If King is correct that the prosecutor induced him into accepting the plea on this basis, he may be entitled to habeas relief. Landrigan, 550 U.S. at 474, 127 S.Ct. 1933.
With certain questions left unanswered, an evidentiary hearing at which King testifies would be the most beneficial resolution of this habeas petition. The record does not reflect what, exactly, King was informed by his defense attorney, or the extent to which the prosecutor stressed concurrent sentencing in inducing King to accept the plea. With that in mind, I would REVERSE the order of the district court and REMAND the case to the district court to conduct an evidentiary hearing to determine whether King may establish that his plea was not “knowing, intelligent, or voluntary.”
.King’s state brief cited to McCarthy v. United States, 394 U.S. 459, 89 S.Ct. 1166, 22 L.Ed.2d 418 (1969) (an involuntary guilty plea violates due process of law); Machibroda v. United States, 368 U.S. 487, 82 S.Ct. 510, 7 L.Ed.2d 473 (1962) (a guilty plea is not constitutionally valid unless entered in a voluntary, knowing, and intelligent manner); and In re Valle, 364 Mich. 471, 477, 110 N.W.2d 673 (1961) (stressing the importance of a knowing and voluntary plea).
.The relevant inquiry was reduced and officially formalized at the motion to withdraw hearing, where the court stated: "The substance of the motion is that the gentleman believed that when he entered into the plea agreement that the bargain for aspect of the agreement of no consecutive sentence applied not only to the two counts or two charges he pled to, but rather it also applied to the parole sentence because he was on parole at the time of the commission of the offenses.”
.See note 1, supra.
. I do not conclude that Santobello and Boy-kin navigate parallel or identical paths on the due process track. Rather, Boykin, Santobel-lo’s progenitor, laid the groundwork for assessing the constitutional validity of guilty pleas. The specificity of King’s argument at the state level is necessarily just as much a “Boykin ” claim, insofar as it attacks the plea as unknowing and involuntary, as it is a "Santobello ” claim. The difference is Santobello provides further edification when the prosecutor’s misconduct is the source of the constitutional infirmity. This syllogism readily explains why the Report and Recommendation begins its discussion section with Boykin. This is so, in light of the fact that King's § 2254 petition was almost identical to his Brief in Support of his Motion to Withdraw his Guilty Plea, because King’s assertions at the state level are necessarily rooted in Boy-kin.
. I am satisfied that King's claim was "adjudicated on the merits” by the Michigan courts. Danner v. Motley, 448 F.3d 372, 376 (6th Cir.2006) (quoting 28 U.S.C. § 2254(d)); see also Werth v. Bell, 692 F.3d 486, 493 (6th Cir.2012) ("absent some indication or Michigan procedural principle to the contrary, we must presume that an unexplained summary order is an adjudication on the merits for AEDPA purposes”).
. Much of King’s appellate brief supports his argument that the state court’s decision was "contrary to” established Supreme Court precedent. His argument proceeds that, even if we did not find that it "directly contravened” precedent, the same facts and analysis support a ruling on "unreasonable application” grounds.
. The conclusion is well-supported by the case law. In evaluating a plea, the court’s inquiry as to whether the defendant truly understands the consequences of his plea must account for all of the relevant factors and idiosyncrasies. For example, the court must inquire into the "background, experience, and conduct of the accused." Johnson v. Zerbst, 304 U.S. 458, 464, 58 S.Ct. 1019, 82 L.Ed. 1461 (1938). The mental capacity of the defendant must also be examined. See, e.g., Parke v. Raley, 506 U.S. 20, 30-31, 113 S.Ct. 517, 121 L.Ed.2d 391 (1992); Henderson v. Morgan, 426 U.S. 637, 647, 96 S.Ct. 2253, 49 L.Ed.2d 108 (1976). In the same vein, whether the defendant is ill, U.S. v. Bonilla, 579 F.3d 1233, 1239 (11th Cir.2009), or on drags, U.S. v. King, 506 F.3d 532, 534 (7th Cir.2007) are relevant inquiries. Also considered in a voluntariness inquiry include the time afforded to the defendant to consider the plea, U.S. v. Robinson, 587 F.3d 1122, 1127 (D.C.Cir.2009), or the alternatives left to the defendant should he choose to go to trial. U.S. v. Morris, 633 F.3d 885, 888-89 (9th Cir.2011).
. That the parole sentence was to automatically run consecutive to the robbery sentences as required by law does not alter our decision. The legal invariability of the sentence does not, in this case, supply us with any insight as to King’s state of knowledge in accepting the plea offer. If it were to be adduced at an evidentiary hearing that the prosecutor informed him of this effect, however, our holding would become moot.