Judge Jolly, now a Senior Judge of this court, participated in the consideration of this en banc case. Judge Graves is recused and did not participate in this decision. Judges Willett and Ho also did not participate in this decision.

It is worth noting also that at no point did the district court address the Mississippi Supreme Court's comparative juror analysis conducted in Chamberlin's postconviction proceeding. The district court's only reference to the Mississippi Supreme Court's postconviction decision was in passing at the very beginning of its opinion. See Chamberlin I , 2015 WL 1485901 , at *1.

The Supreme Court in Miller-El II found that the prosecution had invidiously discriminated in striking ten out of eleven prospective black jurors. Miller-El II , 545 U.S. at 265-66 , 125 S.Ct. 2317 . As one factor in considering the totality of the pretrial record, the Court employed a comparative juror analysis. The district court, however, took this approach to set up as "clearly established law" that Miller-El II "require[s]" a comparative juror analysis. Chamberlin I , 2015 WL 1485901 , at *17. Consequently, the district court held that the Mississippi Supreme Court's decision not to conduct a comparative juror analysis violated this "clearly established law." Id. This holding is erroneous on two grounds.

First, Miller-El II did not clearly establish any requirement that a state court conduct a comparative juror analysis at all, let alone sua sponte . Judge Ikuta of the Ninth Circuit recently examined this issue in depth; we find her analysis compelling. See McDaniels v. Kirkland , 813 F.3d 770 , 782-85 (9th Cir. 2015) (Ikuta, J., concurring). Judge Ikuta explained:

Because Miller-El II considered only whether the state court made an unreasonable factual determination, the Supreme Court did not discuss, let alone squarely establish, a new procedural rule that state courts must conduct comparative juror analysis when evaluating a Batson claim. At no point did Miller-El II suggest that the state court in that case violated the petitioner's constitutional rights by failing to adhere to such a procedural rule. Accordingly, because Miller-El II does not provide a clear answer to the question whether a state court must conduct comparative juror analysis as part of its Batson inquiry, we cannot hold that a state court which fails to conduct comparative juror analysis violates clearly established Federal law, as determined by Miller-El II .

Id. at 783 (internal quotation marks and citation omitted). This is especially true where, as here, the defendant never sought a comparative juror analysis. Nowhere in Miller-El II did the Supreme Court imply-let alone clearly establish-that a state court must conduct a comparative juror analysis sua sponte . Cf. United States v. Atkins , 843 F.3d 625 , 634 (6th Cir. 2016) ("To begin with, the government is correct that the district court's failure to conduct its own comparative juror analysis is not sufficient to require reversal.").

Second, regardless of whether it was required to do so, the Mississippi Supreme Court did conduct a comparative juror analysis in Chamberlin's case, albeit in a postconviction proceeding instead of on direct appeal. Chamberlin's Batson claim was inextricably intertwined with the ineffective assistance of counsel argument she raised at the postconviction proceeding. She argued in relevant part that her trial counsel was ineffective because he should have sought a comparative juror analysis in the trial court. In response to this contention, the Mississippi Supreme Court stated that it had conducted a "thorough review of the record ... including the jury questionnaires provided by Chamberlin." Chamberlin , 55 So.3d at 1051-52 . It found no evidence of "disparate treatment of the struck jurors," and concluded that the identical Batson claim that eventually came before the district court was "without merit." Id. at 1052 .

The district court thus erred twice as it pertains to the "clearly established law" ground for habeas relief under AEDPA. First, it erred in concluding that clearly established federal law required the Mississippi Supreme Court to conduct a comparative juror analysis sua sponte . Second, it erred in failing to address the comparative juror analysis the Mississippi Supreme Court did conduct, albeit in the postconviction context.

B. Unreasonable Determination of the Facts

The district court further concluded that the state court's factual finding that the prosecution did not invidiously discriminate during jury selection was an unreasonable determination of the facts in light of the evidence presented. The district court rested its holding on a comparative juror analysis between Sturgis/Minor and Cooper alone. 3

Before reaching those arguments, however, it is important to stress that the district court did not grant proper deference to the Mississippi Supreme Court's factual findings at the postconviction proceeding. As noted above, the district court concluded it did not need to defer to the state court's factual findings under AEDPA because those findings did not include the requisite comparative juror analysis. We have already explained that conclusion was error because there is no requirement to conduct such a comparison, particularly sua sponte . But even if such a requirement did exist, the Mississippi Supreme Court's factual findings during the postconviction proceeding-findings made pursuant to a comparative juror analysis-would be entitled to AEDPA deference. We federal courts are required to defer to the Mississippi Supreme Court's factual finding that a comparative juror analysis in Chamberlin's case produced no evidence of disparate treatment of black prospective jurors.

Even if we were not required to defer to the state court's factual findings, however, we would still hold that the district court erred in concluding that Chamberlin established that the prosecution's proffered race-neutral reasons were pretextual. To show why this is so, we turn to the comparative juror analysis.

The Supreme Court has instructed that, when analyzing Batson challenges, "bare statistics" are not the be-all and end-all. Miller-El II , 545 U.S. at 241 , 125 S.Ct. 2317 . "Side-by-side comparisons of some black venire panelists who were struck and white panelists allowed to serve" can be "[m]ore powerful." Id. The crux of the district court's ruling is its erroneous comparison of black prospective jurors Sturgis and Minor, who were struck by the prosecution, to white juror Cooper, who was kept.

The district court's determination on this front can be boiled down in this way: (1) the prosecution said questions 30, 34, and 35 were the reasons Sturgis and Minor were struck; (2) Cooper answered those questions identically; therefore (3) questions 30, 34, and 35 could not have been the real reasons Sturgis and Minor were struck, else Cooper would have been struck as well. Accordingly, the prosecution's proffered race-neutral reasons for striking Sturgis and Minor must have been pretextual.

But questions 30, 34, and 35 were not the only questions Sturgis, Minor, and Cooper had to answer. They were rather three questions out of dozens on a pages-long jury questionnaire. And if Cooper in particular gave other responses that materially differentiated him from Sturgis and Minor and made him a more favorable juror for the prosecution, then the district court's ruling does not follow.

Consider, for example, question 53, which asked prospective jurors to circle the response that best matched their opinion on the death penalty. Sturgis and Minor circled "Generally Favor" and "No Opinion," respectively. Cooper, by contrast, circled "Strongly Favor," and then wrote in "for rape, murder, child abuse, [and] spousal abuse" by hand in the margin. Cooper clearly answered a key question in a way that materially distinguished him from Sturgis and Minor. Thus, the most logical explanation for the prosecution's not striking Cooper was not because he was white while Sturgis and Minor were black, but because Cooper was a more favorable juror based on his answers to other questions.

This conclusion is further confirmed by the existence of an additional black juror, Carter, who was accepted by the prosecution. Carter gave worse (from a prosecutor's perspective) answers to question 30 and 34 than did Sturgis and Minor, and gave the same answer as they did to question 35. But she answered question 53 in the same manner as Cooper: circling "Strongly Favor" and then writing in by hand additional crimes for which she felt the death penalty was appropriate. And again, Carter-a black prospective juror-was accepted by the prosecution.

Indeed, the district court conceded that the prosecution could reasonably have viewed Cooper as a more favorable juror than Sturgis and Minor in light of his answer to question 53. But it decisively concluded that it could not consider Cooper's answer to question 53, because question 53 was not one of the race-neutral reasons given by the prosecution for striking Sturgis and Minor. See Chamberlin I , 2015 WL 1485901 , at *6 ("While [his response to question 53] might have made Cooper a slightly more desirable juror, it was not a rationale offered by the prosecutor."). The district court concluded, in other words, that to look at Cooper's other answers would be to allow the State to construct an impermissible post hoc explanation for its strikes of black jurors. This conclusion was erroneous for a number of reasons.

First, the district court took out of context the Miller-El II admonition that "a prosecutor simply has got to state his reasons as best he can and stand or fall on the plausibility of the reasons he gives." Miller-El II , 545 U.S. at 252 , 125 S.Ct. 2317 . The Court was careful to limit its warning only to the prosecutor's "reason[s] for striking [a] juror " at the second prong of the Batson test. Id. at 251 , 125 S.Ct. 2317 (emphasis added). This narrow focus is essential to maintaining the integrity of the Batson framework, which requires a focus on the actual, contemporary reasons articulated for the prosecutor's decision to strike a prospective juror. The timely expressed neutral reasons, after all, are what must be tested for veracity by the trial court and later reviewing courts. And this is what the Supreme Court meant in stating the "stand or fall" proposition: it criticized both the prosecutor and later reviewing courts for accepting either entirely different substituted reasons or post hoc reasons for strikes. The Court's rationale, however, does not extend to preventing the prosecution from later supporting its originally proffered reasons with additional record evidence, especially if a defendant is allowed to raise objections to juror selection years after a conviction and to allege newly discovered comparisons to other prospective jurors. Nothing in the "stand or fall" statement means that the prosecutor would forfeit the opportunity to respond to such contentions.

In addition, the Court specifically noted that when a prosecutor gives a facially race-neutral reason for striking a black juror, a reviewing court must "assess the plausibility of that reason in light of all evidence with a bearing on it ." Id. at 251-52 , 125 S.Ct. 2317 (emphasis added); see also Snyder v. Louisiana , 552 U.S. 472 , 483, 128 S.Ct. 1203 , 170 L.Ed.2d 175 (2008) ("We recognize that retrospective comparison of jurors based on a cold appellate record may be very misleading when alleged similarities were not raised at trial. In that situation, an appellate court must be mindful that an exploration of the alleged similarities at the time of trial might have shown that the jurors in question were not really comparable."). The Court thus drew a distinction between: (1) inventing a new reason for a strike after the fact (not allowed); and (2) reviewing the record to test the veracity of the prosecution's reasons already given in their proper time (required). Cooper's answer to question 53 is an example of the latter, because it goes directly to the key issue of whether Sturgis' and Minor's responses to questions 30, 34, and 35 were the real reasons they were struck.

There is, accordingly, a crucial difference between asserting a new reason for striking one juror and an explanation for keeping another. They are not two sides of the same coin, as the dissent asserts. In the former scenario, the prosecutor effectively concedes that his initial (race-neutral) reasons were insufficient bases for striking the juror. Miller-El 's "stand or fall" requirement applies to this situation, blocking such post hoc rationalizations. See Miller-ElII , 545 U.S. at 250-52 , 125 S.Ct. 2317 . In the latter, the prosecutor's bases for the strike remain in full effect, so Miller-El 's requirement is not implicated. See United States v. Wilkerson , 556 Fed.Appx. 360 , 365 (5th Cir. 2014) (unpublished) (noting that the prosecution should be afforded the opportunity to demonstrate "meaningful distinctions" between asserted comparators). Instead, the prosecutor is highlighting a crucial difference between the black and non-black jurors that prevented the non-black juror from being struck despite sharing strike-worthy characteristics with a black counterpart that was struck. 4

Second, to hold that the prosecution is not allowed to point to Cooper's other jury questionnaire responses is to engage in a bait-and-switch that vitiates the probative value of the jury comparison in the first place. At jury selection, the prosecution was asked to explain why it struck black jurors Sturgis and Minor, as Batson requires. It did so. Then, years later on federal habeas, the defense altered its approach, and the prosecution was now asked to explain why it kept white juror Cooper. And yet, despite the change in inquiry, the prosecution was not allowed to respond, even by pointing to record evidence it undeniably would have been able to identify had a timely objection been made-it was stuck with the answer it had given to an entirely different question during jury selection. Not only is this state of affairs manifestly unfair, it is inconsistent with the Supreme Court's directive regarding juror analysis in Snyder . If a court does not consider the entire context in which a white juror was accepted, then he/she cannot serve as a useful comparator.

A hypothetical will help to illustrate the point:

1. Prosecutor decides, as a default position, to strike all jurors who express concerns about the legal burden of proof.

2. Prosecutor reviews juror questionnaires and notes that Jurors A, B (both black) and C (white) have expressed concerns about the legal burden of proof. Consequently, Prosecutor intends to strike all three by default.

3. Upon further review, Prosecutor notes that Juror C alone strongly favors the death penalty. Because this is a capital case, Prosecutor decides to make an exception to the default rule and retain Juror C because of his favorable death penalty views.

4. Prosecutor strikes Jurors A and B as planned. Responding to a Batson challenge, Prosecutor explains that A and B both expressed concerns about the legal burden of proof.

5. Prosecutor never mentions white Juror C because the law does not require Prosecutor to explain why he decided to keep any specific juror.

In this scenario, when Prosecutor strikes Jurors A and B for their position on the legal burden of proof, Prosecutor has concluded that their position on the legal burden of proof is a sufficiently strong basis to strike them. This implies that there are no other overriding reasons to accept A and B as jurors. Conversely, Juror C is not a comparator because his position on the legal burden of proof is not sufficiently strong to strike him; instead, his position on the legal burden of proof is redeemed by his stance on the death penalty, making him desirable as a juror.

If the defendant in such a case later raises a comparative jury analysis between Jurors A, B, and C as part of a Batson challenge, an accurate and honest assessment of the prosecutor's motives must allow the prosecution to point to white Juror C's view of the death penalty as the reason he was kept. Otherwise the Batson analysis risks capturing too many false positives, precisely because Juror C is no longer an accurate comparator to Jurors A and B.

Third, consider the related issue of the burden that would be placed on the prosecution at jury selection going forward if the district court's reasoning stands. In order to protect against future comparative juror analysis, the prosecution will not only have to explain why it struck black jurors-as Batson requires-but also why it kept white jurors. Indeed, the prosecution will have to explain why it kept every white juror, because it does not know which white jurors will be selected as comparators at some later date. In Chamberlin's case, for example, the only way the prosecution could have avoided the outcome dictated by the district court was by explaining why it kept Cooper. But the prosecution could not have known that Cooper would be the eventual comparator chosen and not some other juror, so it would have had to explain why it kept every white juror. 5 Such a requirement would make the jury selection process impractical, whereas considering the totality of the circumstances conforms with the Court's instruction in Batson , Miller-El II , and Snyder . 6

Fourth, the procedure for conducting a comparative juror analysis described by the district court creates perverse incentives for both the defense and the prosecution. For the defense it is better not to raise comparative juror analysis in the trial court and to wait until much later in the game to point to a white comparator, because the prosecution will be stuck with whatever reasons it gave for striking black jurors in the trial court and allowed no other explanation-no matter how compelling and/or how certain it is that it would have been raised had a timely objection been made. And for the prosecution, it might be deemed strategically advantageous to be less detailed when giving race-neutral reasons in the trial court, because the more general the answers, the harder it will be to conduct a formal side-by-side comparison down the line.

Our holding today does not eviscerate Batson protections: We simply allow a prosecutor the chance to respond whenever the court engages in a comparative juror analysis. Important limitations on that response remain. For one, the prosecutor is constrained by the voir dire record, which helps guard against the fabrication of new distinctions that did not motivate the initial decision. Moreover, even if the prosecutor provides a supported basis for keeping a non-black juror, the court must still determine whether that basis provides an adequately redeeming reason to override the strike-worthy characteristics the non-black juror shares with the black jurors who were struck. Perhaps most importantly, allowing this response does not permit the prosecutor to change his original reasons for striking black jurors. Such protections will guard against the rare cases in which a Batson violation is followed by an ongoing, planned concealment of that violation by the various prosecutors involved in each case.

Conversely, to hold that a reviewing court cannot look at the totality of the circumstances in order to determine an accurate comparator when conducting a comparative jury analysis sua sponte and belatedly on federal habeas is to invert the Batson framework, rendering it unjust, impractical, and contrary to its original purpose.

* * *

The prosecution in Chamberlin's case did what it was supposed to do: it rejected some black prospective jurors and accepted others, accepted some white prospective jurors and rejected others. When asked why it struck individual black prospective jurors, it gave specific race-neutral reasons for the strikes. The Mississippi Supreme Court found on multiple occasions that the prosecution did not invidiously discriminate against black prospective jurors. Then, on federal habeas-where AEDPA deference is the rule-the prosecution was asked to explain years later why it kept a white juror. Yet, when it tried to answer that question with reference to record evidence it would have identified had the defense timely objected, the district court concluded it could not do so. No case-not Batson , Miller-El II , or any other-has ever suggested, let alone mandated, this distortion of the Batson regime.

IV

We find that neither statutory ground for granting federal habeas relief under AEDPA applies to Chamberlin's case. Accordingly, we REVERSE the district court's order granting Chamberlin's petition for a writ of habeas corpus, and VACATE the district court's order setting aside Chamberlin's conviction and sentence.

GREGG COSTA, Circuit Judge, dissenting, joined by STEWART, Chief Judge, DAVIS, DENNIS, and PRADO, Circuit Judges:

The jury-the voice of "We the People" in our justice system-was of such importance to the Founding generation that it is one of only two rights included in both the Constitution and Bill of Rights. 1 U.S. CONST. art. III; amends. VI, VII ; see also THE FEDERALIST No. 83 (Alexander Hamilton) ("The friends and adversaries of the plan of the [constitutional] convention, if they agree on nothing else, concur at least in the value they set upon the trial by jury"). In the latter, it is the only right that is a focus of two amendments. As is the case for many of our finest institutions, the greatest obstacle to the jury system's achieving its full promise has been racial discrimination. From the earliest applications of the Equal Protection Clause to the present, that guarantee's most prominent role in the criminal justice system has been ferreting out such discrimination in the composition of juries. See, e.g., Strauder v. West Virginia, 100 U.S. 303 , 25 L.Ed. 664 (1879) ; Neal v. Delaware , 103 U.S. 370 , 26 L.Ed. 567 (1880) ; Norris v. Alabama , 294 U.S. 587 , 55 S.Ct. 579 , 79 L.Ed. 1074 (1935) ; Patton v. Mississippi , 332 U.S. 463 , 68 S.Ct. 184 , 92 L.Ed. 76 (1947) ; Hernandez v. Texas , 347 U.S. 475 , 74 S.Ct. 667 , 98 L.Ed. 866 (1954) ; Batson v. Kentucky , 476 U.S. 79 , 106 S.Ct. 1712 , 90 L.Ed.2d 69 (1986) ; Miller-El v. Dretke , 545 U.S. 231 , 125 S.Ct. 2317 , 162 L.Ed.2d 196 (2005) ( Miller-El II ); Snyder v. Louisiana , 552 U.S. 472 , 128 S.Ct. 1203 , 170 L.Ed.2d 175 (2008) ; Foster v. Chatman , --- U.S. ----, 136 S.Ct. 1737 , 195 L.Ed.2d 1 (2016). Unlike other trial rights, the one requiring a jury protects not just those charged with crimes, but all citizens whose service on the jury is essential to ensuring that a cross section of the community is making the important, in this case life-or-death, decisions our justice system confronts. J.E.B. v. Alabama ex rel. T.B. , 511 U.S. 127 , 140, 114 S.Ct. 1419 , 128 L.Ed.2d 89 (1994) ("Discrimination in jury selection ... causes harms to the litigants, the community, and the individual jurors who are wrongfully excluded from participation in the judicial process."); Powers v. Ohio , 499 U.S. 400 , 406, 111 S.Ct. 1364 , 113 L.Ed.2d 411 (1991) (" Batson recognized that a prosecutor's discriminatory use of peremptory challenges harms the excluded jurors and the community at large.").

The problem of racial discrimination in the makeup of juries is now largely one about the exercise of peremptory strikes. It has been three decades since the Supreme Court recognized that discriminatory use of a strike violates the Constitution. Batson , 476 U.S. at 84 , 106 S.Ct. 1712 . Even though a high proportion of the recent cases in which the Supreme Court has found a Batson violation come from states in our circuit, 2 you can count on one hand the number of cases from this court finding the discriminatory use of a preemptory strike. It appears that only two of the hundreds of Batson decisions in our circuit have ever found that a strike was discriminatory (a few others vacated convictions based on procedural error in application of the Batson framework). See Hayes v. Thaler , 361 Fed.Appx. 563 (5th Cir. 2010) ; Reed v. Quarterman , 555 F.3d 364 (5th Cir. 2009). Yet the concern today is that a decision affirming the district court's finding of discriminatory strikes, which would put us at a once-a-decade rate of finding Batson violations, would impose too much of a "burden ... on the prosecution." Maj. Op. at 843.

The two cases in which we have found discrimination both relied in large part on comparative juror analysis. Today's opinion saps most of the force out of this one tool that has ever resulted in us finding a Batson violation. Despite the only reasons cited at trial for striking two black jurors applying equally to an accepted white juror, the majority rejects the direct conclusion to be drawn from this inconsistency that the proffered reasons could not have been the real reasons for the strikes. If this case in which the compared jurors are identical with respect to the reasons stated at trial is not enough (the standard only requires that they be similarly situated), it is difficult to see how comparative analysis will ever support a finding of discrimination.

What is even more troubling is that we have been down this road before. The way the majority opinion gets around the identical comparison is to differentiate the jurors based on reasons not cited during the Batson inquiry at trial. In Miller-El II , the Supreme Court found error in our application of comparative juror analysis that did the same thing: "substitution of a reason" for the strike that was not offered at trial. Miller-El II , 545 U.S. at 252 , 125 S.Ct. 2317 . The Court set forth the following rule in unmistakable terms:

It is true that peremptories are often the subject of instinct and it can sometimes be hard to say what the reason is. But when illegitimate grounds like race are in issue, a prosecutor simply has got to state his reasons as best he can and stand or fall on the plausibility of the reasons he gives. A Batson challenge does not call for a mere exercise in thinking up any rational basis. If the stated reason does not hold up, its pretextual significance does not fade because a trial judge, or an appeals court, can imagine a reason that might not have been shown up as false.

Id . (citation omitted); see also Reed , 555 F.3d at 376 ("We must consider only the State's asserted reasons for striking the black jurors and compare those reasons with its treatment of the nonblack jurors" (citing Miller-El II , 545 U.S. at 252 , 125 S.Ct. 2317 )). Yet that is exactly what the majority opinion does: it uses the answers to questions not identified at trial as the basis for overturning the district court's finding that clear and convincing evidence of discrimination exists. As will be explored further, this approach used to avoid the clear import of a direct comparison of the reasons stated at trial is the same rejected analysis of our Miller-El II opinion and the Supreme Court dissent. It is one thing to make a mistake; it is quite another to not learn from it. 3

I.

Before getting to those critical errors in the majority opinion's comparative juror analysis, it is important to note the revealing pattern of discriminatory strikes. The majority opinion does not even mention the highly disproportionate strikes of black prospective jurors. Instead it breezily says the prosecution "rejected some black prospective jurors and accepted others, accepted some white prospective jurors and rejected others." Maj. Op. at 844. It is no wonder the majority opinion does not details those "some"s and "others"; they are nothing alike.

The prosecution struck seven of the first eight black venire members it considered, which included the challenged strikes of Sturgis and Minor. The inverse was true for the first eight white jurors the prosecution considered: it accepted seven of eight (and ended up accepting nine of the first ten whites). Only after defense counsel started raising Batson objections and the prosecution was running out of strikes did it accept the two black jurors who ended up on the jury, and the second was only accepted in a moment of confusion when the prosecutor believed the juror had already been struck. Cf. Miller-El II , 545 U.S. at 249-50 , 125 S.Ct. 2317 (finding unpersuasive that, towards the end of jury selection, the prosecution accepted a black juror, noting that most of the prosecution's challenges were gone and the prosecutor "had to exercise prudent restraint" at that point).

Even including those late, post-objection decisions, the overall numbers evince discrimination. The prosecution struck nearly two times as many black jurors as it accepted (eight strikes compared to five accepted, including one alternate), while accepting more than four times as many white jurors as it struck (five strikes compared to twenty-three accepted, including three alternates). It exercised 62% of its strikes on black jurors, despite black jurors making up only 31% of qualified prospective jurors.

This racial breakdown of the strikes is even more telling when compared with the results random strikes would predict. Given the demographics of the venire, the probability that random, race-neutral strikes would result in 8 of the 13 struck jurors being black was about 1 in a 100. See generally Joseph L. Gastwirth, Statistical Testing of Peremptory Challenge Data for Possible Discrimination , 69 VAND. L. REV. EN BANC 51, 59-62 (2016) ; Joseph L. Gastwirth , Case Comment: Statistical Tests for the Analysis of Data on Peremptory Challenges , 4 L. PROBABILITY & RISK 179, 182 (2005) (both showing how to complete this analysis). That probability of roughly 0.01 is "smaller than 0.05, the most frequently used probability level for determining statistical significance, which is equivalent to the two-standard deviation criterion" that the Supreme Court found to be the point at which the possibility of a race-neutral explanation was "suspect" in a case challenging exclusion of Hispanic grand jurors in south Texas. Gastwirth , Statistical Testing of Peremptory Challenge Data , at 60; Castaneda v. Partida , 430 U.S. 482 , 496 n.17, 97 S.Ct. 1272 , 51 L.Ed.2d 498 (1977). Looking at the strikes in terms of odds rather than probability reveals another stark statistic: a black juror had more than seven times the odds of being struck that a white juror did. Id . at 66. 4

When the Supreme Court has considered statistical evidence of discrimination in jury selection, it has focused on this demographic breakdown of strikes. Miller-El I , 537 U.S. at 331, 123 S.Ct. 1029 ("African-Americans were excluded from petitioner's jury in a ratio significantly higher than Caucasians were"). As to those strikes, the majority opinion cannot dispute that the prosecutor was far more likely to strike black potential jurors than white venire members. This assessment of a lawyer's overall strikes is not just part of the prima facie case that makes up the first stage of Batson. As common sense would dictate, disproportionate strike rates involving the entire venire are also relevant in considering the ultimate question whether a particular strike was discriminatory. Miller-El II , 545 U.S. at 265 , 125 S.Ct. 2317 (considering statistics of overall strikes in concluding that discrimination existed); Miller-El I , 537 U.S. at 342, 123 S.Ct. 1029 ("[T]he statistical evidence alone raises some debate as to whether the prosecution acted with a race-based reason when striking prospective jurors."); Hayes , 361 Fed.Appx. at 570 (recognizing that the prosecutor's using 8 of 11 strikes against black jurors is "indicative of discriminatory intent"); Fields v. Thaler , 588 F.3d 270 , 274 (5th Cir. 2009) (explaining that at the final Batson step "the defendant may rely on all relevant circumstances to raise an inference of purposeful discrimination"); cf. Reeves v. Sanderson Plumbing Prods., Inc. , 530 U.S. 133 , 148-49, 120 S.Ct. 2097 , 147 L.Ed.2d 105 (2000) (recognizing in the analogous burden shifting framework for employment discrimination cases that the "strength of the plaintiff's prima facie case" remains relevant to the ultimate question).

That two blacks ended up on the jury-one only because the prosecutor mistakenly though the juror had already been struck-does not overcome the strong inference to be drawn from the disproportionate strikes. Other courts of appeals have explained why this is the case:

The final composition of the jury ... offers no reliable indication of whether the prosecutor intentionally discriminated in excluding a member of the defendant's race. ... "[A] Batson inquiry focuses on whether or not racial discrimination exists in the striking of a black person from the jury, not on the fact that other blacks may remain on the jury panel."

Holloway v. Horn , 355 F.3d 707 , 728-29 (3d Cir. 2004) (quoting United Statesv. Johnson , 873 F.2d 1137 , 1139 n.1 (8th Cir. 1989) ). Commentators have also criticized looking at the final makeup of the jury rather than strikes, with one stating this is "not so much a method as an excuse" in that it "fails to address the primary purpose of the Batson rule-the protection of individual jurors." Kenneth J. Melilli, Batson in Practice: What We Have Learned About Batson and Peremptory Challenges , 71 NOTRE DAME L. REV. 447, 475 (1996) ; see also Gastwirth, supra , at 56 (finding the approach unreasonable because it ignores the prosecutor's peremptory strikes). Indeed, academic authority-not just that of law professors, but also statisticians-recognizes that the rate at which the prosecutor struck black jurors as compared to nonblack jurors is the most probative metric in Batson cases. See David C. Baldus, et al., Statistical Proof of Racial Discrimination in the Use of Peremptory Challenges , 97 IOWA L. REV. 1425, 1455 (2012) ("[T]he most probative measures are based on contrasts between the prosecutorial strike rates of black and nonblack venire members."); Gastwirth, Statistical Testing of Peremptory Challenge Data, at 60 (considering the comparative strike rates for black and nonblack jurors and the difference between the number of minorities struck by the prosecutor to the number expected if those challenged were a random sample).

In terms of those strikes, it is worth repeating that a black juror was more than seven times as likely to be struck as a white one and the random chance that so many blacks would be struck is a remote 1 in a 100. "Happenstance is unlikely to produce this disparity." 5 Miller-El II , 545 U.S. at 241 , 125 S.Ct. 2317 (quoting Miller-El I , 537 U.S. at 342, 123 S.Ct. 1029 ).

II.

A.

Comparative juror analysis is a tool that helps determine whether this disproportionate exclusion of black jurors was the extraordinary coincidence it would take to defy these odds. An understanding of probability is not needed to see the mistake the majority opinion's approach makes with this inquiry; the most routine judicial task of reading precedent reveals it.

The rationale for comparative juror analysis is this: "If a prosecutor's proffered reason for striking a black panelist applies just as well to an otherwise-similar nonblack who is permitted to serve, that is evidence tending to prove purposeful discrimination to be considered at Batson 's third step." Miller-El II , 545 U.S. at 241 , 125 S.Ct. 2317 . Such "side-by-side comparisons" can be "powerful" evidence of discrimination because they reveal that a lawyer's race-neutral reasons are pretext. Miller-El II , 545 U.S. at 241 , 125 S.Ct. 2317 . To put it in the practical context of jury selection, the idea is that if a lawyer is so troubled by a juror's views on certain issues that the concern leads to striking that juror, then one would expect to see the same concern with another juror expressing the same views. In a jury selection like this one involving written questionnaires, if the lawyer highlighted answers on one juror's form as troubling, the same answers on another juror's form should also be highlighted. If those highlighted answers are later cited by the lawyer in response to a Batson challenge for which the trial court has found prima facie support and thus requested an explanation, it should not be hard for the lawyer to see and consider all the questionnaires with that answer highlighted.

Miller-El II shows how this analysis can reveal pretext. The state struck a potential black juror purportedly because he "said that he could only give death if he thought a person could not be rehabilitated." 545 U.S. at 243 , 125 S.Ct. 2317 . If that were the real reason, the Court noted, the prosecutor "should have worried about a number of white panel members he accepted" who expressed similar views. Id. at 244-45 , 125 S.Ct. 2317 . Likewise, although the prosecutor's purported reason for striking another prospective juror (that he considered death "an easy way out") was reasonable on its face, "its plausibility [wa]s severely undercut by the prosecution's failure to object to other panel members who expressed views much like [his.]" Id. at 247-48 , 125 S.Ct. 2317 ; see also Foster , 136 S.Ct. at 1751 (finding "otherwise legitimate reason[s]" for striking prospective black jurors "difficult to credit in light of the State's acceptance of" white jurors to whom those reasons also applied); Snyder , 552 U. S. at 483 , 128 S.Ct. 1203 (same); Reed , 555 F.3d at 372-73 (same).

The Miller-El II comparison revealed the prosecutor's reasons to be pretextual and thus powerful evidence of discrimination even though other reasons the prosecutor gave for striking black jurors did not also apply to accepted white jurors. 545 U.S. at 247 , 125 S.Ct. 2317 . For example, the prosecutor gave an additional reason for striking two black jurors-that they had relatives who had been convicted of a crime-which did not apply to the white jurors to whom the Court compared them. Miller-El II , 545 U.S. at 246-47 , 125 S.Ct. 2317 ; id. at 290-91 , 125 S.Ct. 2317 (Thomas, J., dissenting). The Court nonetheless rejected the argument that pretext can be found only when an accepted white juror "match[es] all" of the reasons the prosecutor gave for striking a black juror. Id. at 247 n.6, 125 S.Ct. 2317 (quoting Thomas, J., dissenting). A rule that "no comparison is probative unless the situation of the individuals compared is identical in all respects" identified by the prosecutor would, it explained, "leave Batson inoperable; potential jurors are not products of a set of cookie cutters." Id .

The jurors "identical in all respects" that Miller-El II thought unlikely exist here. Every reason the prosecutor identified for excluding Sturgis and Minor applied to Cooper, the white juror who was not struck. 6 All three said they were "not sure" if they were emotionally capable of announcing a verdict of death; were "not sure" if they would hold the State to a higher burden of proof than the law requires given that it was a death penalty case; and "yes," they would want to be one hundred percent certain of the defendant's guilt before finding her guilty. Comparative juror analysis thus shows that the prosecutor's reason for striking Sturgis and Minor could not have been their answers to questions 30, 34, and 35. Otherwise, he would not have accepted Cooper who had the same answers the prosecution did not like. The perfect match among the answers of these jurors means that even more than in the other cases that have found pretext based on a comparative juror analysis, "[t]he prosecutors' chosen race-neutral reasons for the strikes do not hold up and are so far at odds with the evidence that pretext is the fair conclusion, indicating the very discrimination the explanations were meant to deny." Miller-El II , 545 U.S. at 265 , 125 S.Ct. 2317 ; Snyder , 552 U.S. at 485 , 128 S.Ct. 1203 ("The prosecution's proffer of this pretextual explanation naturally gives rise to an inference of discriminatory intent."); Reed , 555 F.3d at 380-81 ("[T]he comparative analysis demonstrates what was really going on: the prosecution used its peremptory challenges to ensure that African-Americans would not serve on Reed's jury.").

B.

How does the majority opinion try to avoid the implication of pretext that is stronger in this case than those in which the Supreme Court and our court have used comparative analysis to find Batson violations? It first does so by invoking AEDPA deference. That deference is substantial in allowing a federal court to grant postconviction relief only if the state court's rejection of the claim "was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding." 28 U.S.C. § 2254 (d)(2). And a state court's factual findings are presumed to be sound unless the petitioner rebuts the "presumption of correctness by clear and convincing evidence." 28 U.S.C. § 2254 (e)(1).

"The standard is demanding but not insatiable; ... '[d]eference does not by definition preclude relief.' " Miller-El II , 545 U.S. at 240 , 125 S.Ct. 2317 (quoting Miller-El I , 537 U.S. at 340, 123 S.Ct. 1029 ). And in Miller-El II and Reed , comparative juror analysis less compelling than the identical comparison in this case helped meet that standard. See id . at 240, 266, 125 S.Ct. 2317 (granting relief under section 2254(d)(2) ); Reed , 555 F.3d at 372-73 (same). Indeed, courts have issued writs under AEDPA relying solely on comparative juror analysis to find a Batson violation. Hayes , 361 Fed.Appx. at 573 ; see also Drain v. Woods , 595 Fed.Appx. 558 , 571-81 (6th Cir. 2014) (granting writ by finding a Batson violation relying only on a comparative juror analysis); cf. United States v. Taylor , 636 F.3d 901 , 906 (7th Cir. 2011) (Sykes, J.) (relying on striking comparative juror analysis to find discrimination under the also deferential "clearly erroneous" standard that governs review of federal trial court rulings). Here the damning comparative juror analysis does not stand alone. It is reinforced by the pattern of overall strikes, which makes it highly unlikely as a statistical matter that the disproportionate striking of black jurors was "mere happenstance." Miller-El II , 545 U.S. at 241 , 125 S.Ct. 2317 . There is also the absence of follow-up questions about the responses that supposedly motivated the prosecutor's strikes of Sturgis and Minor that one would expect if those were real concerns. See, e.g., Miller-El II , 545 U.S. at 246 , 125 S.Ct. 2317 (citing Ex parte Travis , 776 So.2d 874 , 881 (Ala. 2000) ); Davis v. Fisk Elec. Co. , 268 S.W.3d 508 , 519 (Tex. 2008) ; Jackson v. Stroud , 539 S.W.3d 502 , 510-11, 2017 WL 6519913 , at *5 (Tex. App.-Houston [1st Dist.], Dec. 21, 2017) (all explaining that a "failure to engage in any meaningful voir dire examination on the issues that he claims concerned him suggests that his explanation on appeal is pretextual").

The majority opinion defers to findings of the state court in rejecting an ineffective assistance of counsel claim, rather than its findings on direct appeal rejecting the Batson claim. This is curious. The Mississippi Supreme Court's rejection of the Batson claim on direct appeal did not address comparative juror analysis (as will be discussed below, that makes it no different than most of the cases in which the Supreme Court or our court have conducted comparative juror analysis during federal habeas). Chamberlin v. State , 989 So.2d 320 336-39 (Miss. 2008). At the state postconviction stage, Chamberlin raised the ineffective assistance claim challenging her trial counsel's failure to conduct a comparative analysis during jury selection. Chamberlin v. State , 55 So.3d 1046 , 1051-52 (Miss. 2010). If a claim of ineffective assistance of counsel that relates to an underlying Batson issue is the relevant state court ruling for AEDPA purposes on the direct Batson claim, then does a defendant who fails to bring a direct Batson claim in state court nonetheless exhaust that claim for purposes of federal review by bringing it on state habeas in the context of a Strickland claim? Can a habeas petitioner resuscitate a procedurally defaulted Batson claim by raising a Strickland claim in state court challenging the attorney's conduct during jury selection? That would seem to be the implication given the majority opinion's treatment of the ruling on the Sixth Amendment claim as the direct Batson "claim" for purposes of AEDPA review. 28 U.S.C. § 2254 (d) (granting deference to state court "adjudication of the claim").

In any event, reliance on the state habeas ruling only puts the error of the state court's and majority opinion's comparative juror analysis front and center:

• This is what the Supreme Court of the United States has said cannot be done in comparing the jurors: "If the stated reason does not hold up, its pretextual significance does not fade because a trial judge, or an appeals court, can imagine a reason that might not have been shown up as false." Miller-El II , 545 U.S. at 252 , 125 S.Ct. 2317 .

• This is what the Supreme Court of Mississippi said in concluding that the comparison would not show discrimination: "[A] thorough review of the record in this case, including the jury questionnaires provided by Chamberlin , discloses that each of the African-American jurors struck had at least one response in his or her questionnaire that differentiated him or her from the white jurors who were accepted by the State." Chamberlin , 55 So.3d at 1051-52 .

Conducting a "thorough review" of the entire record to identify as reasons for the strikes distinctions among the comparators that were not contemporaneously cited violates Miller-El II . As the stand-or-fall principle recognizes, such differences will just about always exist when every possible characteristic is fair game. The state habeas court's use of comparative juror analysis is thus "contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States" in Miller-El II. 28 U.S.C. § 2254 (d)(1). If the state habeas ruling on ineffective assistance is indeed the Batson decision we are reviewing, then its plain legal error makes this is an even easier case than I thought.

But regardless of whether the Mississippi court committed AEDPA legal error under section 2254(d)(1) and whether the state court must conduct a comparative juror analysis as the district court concluded, the Supreme Court has made clear that a federal habeas court can consider comparative juror analysis in its section 2254(d)(2) review of whether a state court Batson ruling was based on an "unreasonable determination of the facts." 7 Miller-El II , 545 U.S. at 240 , 125 S.Ct. 2317 . As will be discussed further, it can do so even when the defense made no comparison at any level in state court. See id . at 241 & n.2, 125 S.Ct. 2317 ; Reed , 555 F.3d at 372-73 ; Woodward v. Epps , 580 F.3d 318 , 338 (5th Cir. 2009) (all conducting comparative analysis as part of the section 2254(d)(2) analysis even though no comparison was identified in state court).

C.

At least as a general matter, the majority opinion recognizes that a federal habeas court reviewing a Batson claim can consider comparative juror analysis not raised at trial because it engages in that inquiry. But in doing so, the majority opinion makes the same mistake as the Mississippi habeas court in relying on juror differences not identified at trial. It cannot contest the obvious: that on the questions the prosecutor cited during jury selection as his reasons for excluding black jurors Sturgis and Minor, accepted white juror Cooper gave the same responses. Instead, it argues that it should now be able to identify differences among those prospective jurors on their responses to other questions. The example is the three prospective jurors' differing answers to a separate question about the death penalty (question 53): Cooper was strongly in favor of the death penalty whereas Sturgis "generally favored" it and Minor had "no opinion." Maj. Op. at 840.

So how does the majority opinion get around Miller-El II 's command that a prosecutor has to "stand or fall on the plausibility of the reasons he gives"? 545 U.S. at 252 , 125 S.Ct. 2317 ; see also Reed , 555 F.3d at 376 ("We must consider only the State's asserted reasons for striking the black jurors and compare those reasons with its treatment of the nonblack jurors" (citing Miller-El II , 545 U.S. at 252 , 125 S.Ct. 2317 )). It thinks the Supreme Court's emphatic prohibition on post-trial justifications can be overcome by repackaging the argument made by the State about the different answers to question 53. What the State candidly recognizes in its briefing is a new reason for striking the black jurors is now a new reason for keeping the white juror.

Of course, this is just the other side of the same coin. If the difference between the three was question 53, that would mean Sturgis and Minor were struck not only because of their answers to questions 30, 34, and 35, but also because of their more lukewarm support of the death penalty conveyed in response to question 53. As its name demonstrates, the inquiry is a comparative one that requires differentiating the answers of struck and accepted jurors. That means citing different answers to the same question as a reason for keeping one juror is the same as saying the difference was a reason for striking the other juror. See, e.g. , United States v. Taylor , 636 F.3d 901 , 906 (7th Cir. 2011) (properly framing the key question as the reason "for striking [the black juror] but not [the white juror]," which recognizes that the reasons for striking a black juror and keeping a white one are inseparable). To use a simple example, assume a prosecutor struck Jurors A and B on the ground that they wore hats in the courtroom (this is sometimes cited as a reason for strikes on the ground that it shows a lack of respect for the process). But Juror C was also wearing a hat. When this is later pointed out, the reviewing court speculates that Juror C must have been kept in the panel despite the hat because she expressed greater support for the death penalty on a questionnaire than did Jurors A and B. If the court were able to read the prosecutor's mind and this were in fact the real reason for the disparate treatment, then that would mean the hat was not the deal breaker; it alone was not enough for a strike as shown by the acceptance of Juror C. Jurors A and B thus would have been struck, per the court's conjecture, because they wore a hat and were less supportive of the death penalty. And if that were in fact the case, Miller-El II says the prosecutor had to cite both of those reasons.

The view that courts may credit new reasons jurors were kept despite sharing the trait the prosecution claimed justified striking black jurors-a novel position as the en banc court cites no other example of a court doing this-would make meaningless Miller-El II 's bar on considering new reasons for strikes. Whether labelled as reasons for striking the black juror or ones for keeping the comparators, allowing new explanations years after trial turns the Batson inquiry into a "mere exercise in thinking up any rational basis" as there is no way to ensure the post-trial justification is what actually motivated the decisions made during jury selection. Miller-El II , 545 U.S. at 252 , 125 S.Ct. 2317 . Today's decision demonstrates this as it does not even try to test the genuineness of the new explanation the state offers in its brief by, for example, requiring a hearing at which the prosecutor who selected the jury would testify. See infra Section IV.

Miller-El II shows why the distinction between reasons for striking and keeping comparators is empty. The new reason for striking the black juror our court offered that the Supreme Court rejected-his ambivalence about the death penalty-could just as easily have been treated as a reason for keeping the white jurors: their firmer support of the death penalty. 545 U.S. at 250-52 , 125 S.Ct. 2317 ; Miller-El v. Dretke , 361 F.3d 849 , 856-57 (5th Cir. 2004). Indeed, that is how the Miller-El II dissent characterized the difference: a white juror was likely kept because the juror "was adamant about the value of the death penalty for callous crimes." 545 U.S. at 294 , 125 S.Ct. 2317 (Thomas, J., dissenting). As the dissent explained more fully:

In explaining why veniremen Hearn, Witt, and Gutierrez were more favorable to the State than Fields, the majority faults me for 'focus[ing] on reasons the prosecution itself did not offer.' The majority's complaint is hard to understand. The State accepted Hearn, Witt, and Gutierrez. Although it is apparent from the voir dire transcript why the State wanted to seat these veniremen on the jury, it was never required to 'offer' its reasons for doing so.

Id . at 306 n.4, 125 S.Ct. 2317 (Thomas, J., dissenting) (emphasis in original). The Supreme Court majority rejected this attempt to offer never-before-cited reasons for keeping white jurors, viewing it as a violation of the stand-or-fall principle: "The dissent offers other reasons why these nonblack panel members who expressed views on rehabilitation similar to Field's were otherwise more acceptable to the prosecution than he was. In doing so, the dissent focuses on reasons the prosecution itself did not offer." Id . at 245 n.4, 125 S.Ct. 2317 . Today's opinion is thus directly at odds with how Miller-El II treated new reasons, even those for "keeping white jurors": it did not consider them. The Supreme Court's refusal to consider new justifications, whether framed as a reason for excluding the black juror or in mirror-image terms as a reason for accepting nonblack jurors, binds us.

The majority also says we can look at answers to questions other than the three cited at trial because Miller-El II instructed courts to evaluate whether a prosecutors' stated reason is plausible "in light of all evidence with a bearing on it." 545 U.S. at 251-52 , 125 S.Ct. 2317 . But that should not be read to provide an end run around the same opinion's emphatic prohibition on considering new reasons. And what matters most is what Miller-El II actually did: refuse to consider reasons for differential treatment not mentioned in the trial court. Miller-El II shows the way to reconcile these two principles. There is a difference between evidence bearing on the plausibility of the prosecutor's stated reason, which reviewing courts should consider, and new reasons, which they may not. In evaluating whether proffered reasons were plausible, Miller-El II looked to evidence of the prosecutor's veracity other than just the juror comparisons: did he rely on misrepresentations about stricken jurors' answers, probe jurors about the areas of concern, or give inconsistent explanations for strikes? Id. at 244-51 , 125 S.Ct. 2317 . All of these inquiries kept the focus on the reasons for the strikes asserted at trial.

In contrast, Miller-El II refused to consider a new reason this court had identified on appeal. Id. at 252 , 125 S.Ct. 2317 . The prosecutor initially had explained a strike by saying the potential juror thought the death penalty was "too easy on some defendants." Id. at 250 , 125 S.Ct. 2317 . When the defendant pointed out during federal habeas that the same reason applied to white jurors the state accepted, our court found the real reason for the strike must have been the struck black juror's "general ambivalence about the [death] penalty and his ability to impose it." Id. at 248-51 , 125 S.Ct. 2317 ; see Miller-El , 361 F.3d at 856-57 . Miller-El II rejected this approach, similar to that of today's opinion, because the "Court of Appeals's ... substitution of a reason ... does nothing to satisfy the prosecutors burden." 545 U.S. at 252 , 125 S.Ct. 2317 . If that new reason our court offered was just part of evaluating whether a prosecutor's stated reason was plausible "in light of all evidence," the Supreme Court would not have ruled it off limits.

Other circuits conducting comparative jury analysis have also read Miller-El II as requiring that the "validity of a strike challenged under Batson must 'stand or fall' on the plausibility of the explanation given for it at the time, not new post hoc justifications." Taylor , 636 F.3d at 902 ; see also Love v. Cate , 449 Fed.Appx. 570 , 572 (9th Cir. 2011) (refusing to consider the State's post-trial explanation that white jurors it accepted "had non-racial characteristics that distinguished them from the black venire-member" the State struck because "the prosecutor never stated to the state trial court that he relied on these characteristics, even though Batson required him to articulate his reasons"); McGahee v. Alabama Dep't Of Corr. , 560 F.3d 1252 , 1269 (11th Cir. 2009) (faulting the state appellate court for bolstering the prosecutor's reason with a new explanation when the "State never offered such a full explanation"). In Taylor , the only reason the prosecutor gave during jury selection for a strike was that the black juror was unwilling to impose the death penalty on a nonshooter, a position also taken by accepted white jurors. 636 F.3d at 903, 905 . After a remand because comparative juror analysis raised concerns about the strike, the district court credited a different justification: it concluded that the comparators' differing views about the death penalty-which "the prosecutor did not say a word about" at trial-explained their disparate treatment. Id. at 905-06 . Those after-the-fact explanations could be characterized as reasons for keeping the white jurors just as much as they could be treated as reasons for striking the black juror. Yet in the opinion reversing, Judge Sykes explained that it was clear error to accept "new, unrelated reasons extending well beyond the prosecutor's original justification." 8 Id. at 906 . So it is today.

None of these others circuits or the Supreme Court has said that Miller-El II 's stand-or-fall rule applies only at the second step of Batson when the challenged lawyer must state race-neutral reasons. Maj. Op. at 841. The Supreme Court said just the opposite about the placement of the pretext inquiry: "If a prosecutor's proffered reason for striking a black panelist applies just as well to an otherwise-similar nonblack who is permitted to serve, that is evidence tending to prove purposeful discrimination to be considered at Batson 's third step ." Miller-El II , 545 U.S. at 240-41 , 125 S.Ct. 2317 (emphasis added). The whole point of comparative juror analysis is to flush out pretext. As it is in the analogous framework for deciding summary judgment in employment discrimination cases, pretext is directly related to the ultimate question of discrimination. The second step requires nothing more than the assertion of a race-neutral reason; the third step tests the legitimacy of that offered reason with comparative analysis playing a key role. Batson , 476 U.S. at 98 , 106 S.Ct. 1712 . Of course, the ease of manipulation via post-trial rationalizations that motivates the stand-or-fall principle could happen just as easily in any of the Batson steps. If it is just a "step 2" concern as the majority says, why didn't the Seventh Circuit in Taylor allow consideration of the comparators' differing death penalty views as part of "step 3"? Why didn't Miller-El II allow consideration at step 3 of the differences our court and Justice Thomas identified among the comparators? If prosecutors and courts had free reign at Batson 's third step to compare jurors as to characteristics not cited as a contemporaneous reason for the strike, one would expect to see cases allowing that. There are none.

D.

As no other court applying Miller-El II has relied on reasons beyond those given at trial when comparing jurors, the majority is forced to somehow show that this case is unique. It emphasizes that defense counsel did not identify the comparison at trial. The glaring problem with this is that the same was true in Miller-El II , as well as in most of the subsequent cases faithfully applying its stand-or-fall command. As there is nothing unusual about this procedural posture of the Batson challenge-indeed it is the norm-there is no basis for the majority opinion's new rule that says a prosecutor "is allowed to explain why he accepted non-black comparators at the time the analysis is [first] undertaken." Maj. Op. 843 n.6.

There was no invocation of comparative analysis at Miller-El's trial. 545 U.S. at 241 n.1, 125 S.Ct. 2317 . He did not point out comparable white jurors until federal habeas review. Id. at 241 n.2, 125 S.Ct. 2317 . This did not go unnoticed. Justice Thomas objected that in state court the petitioner "did not even attempt to rebut the State's racially neutral reasons" and instead "presented no evidence and made no arguments" in response to the prosecutor's stated justifications. Id. at 278 , 125 S.Ct. 2317 (Thomas, J. dissenting). He protested that the majority's reliance on "theories that Miller-El never argued to the state courts" and, like Mississippi and the majority opinion does here, argued that "AEDPA does not permit habeas petitioners to engage in this sort of sandbagging of state courts." Id. at 279 , 125 S.Ct. 2317 . But the Miller-El II majority rejected this position, holding that the "comparisons of black and nonblack venire panelists" was a "theor[y] about th[e] evidence" properly raised for the first time on federal habeas review to support the petitioner's preserved Batson claim. Id. at 241 n.2, 125 S.Ct. 2317 . Importantly, although no objection put the prosecutor on notice that his reason applied equally to comparable white jurors, the Miller-El II court held the state to his reasons. Id. at 252 , 125 S.Ct. 2317 . The prosecutor gave specific reasons-for example a potential juror thought the death penalty was "too easy on some defendants," id. at 250 , 125 S.Ct. 2317 -so the state could not rely on other dissimilarities fixed in the record even though they related to the same general topic of views on the death penalty, see id. (not considering a juror's "general ambivalence about the [death] penalty"); id. at 290, 293-94 , 125 S.Ct. 2317 (Thomas, J., dissenting) (noting the majority refused to consider that one comparable white juror "was adamant about the value of the death penalty for callous crimes").

Until today, we have likewise recognized that Miller-El II 's command that prosecutors are stuck with the reasons they cited during voir dire applies when the defense does not identify comparators at trial. See Reed , 555 F.3d at 372-73 (holding that under Miller-El II and Snyder the federal habeas court was required to consider the comparative analysis no matter that the petitioner did not identify comparators at trial); see also Woodward , 580 F.3d at 338 (holding a comparative analysis was not waived although not raised at trial and conducting such an analysis focused on justifications the prosecutor offered at trial); United States v. Wilkerson , 556 Fed.Appx. 360 , 363-65 (5th Cir. 2014) (affirming this court "must consider only the [Government's] asserted reasons for striking the black jurors and compare those reasons with its treatment of the nonblack jurors" while noting the petitioner's failure to point to similarities at trial "robb[ed] the Government of the opportunity to demonstrate other meaningful distinctions"). In fact, when we have engaged in comparative juror analysis, more often than not the comparison was not raised at trial. See, e.g. , Reed , 555 F.3d at 369-75 ; Woodward , 580 F.3d at 338 ; Smith v. Cain , 708 F.3d 628 , 638 (5th Cir. 2013) ; Stevens v. Epps , 618 F.3d 489 , 497 (5th Cir. 2010) ; Hayes , 361 Fed.Appx. at 571 ; Wade v. Cain , 372 Fed.Appx. 549 , 553 (5th Cir. 2010). 9 And as we observed in Reed , Texas appellate courts also routinely conduct comparative juror analyses when defendants did not contemporaneously identify comparators. See , e.g. , Vargas v. State , 838 S.W.2d 552 , 556 (Tex. Crim. App. 1992) (en banc);

Blackman v. State , 414 S.W.3d 757 , 765, 765 n.31 (Tex. Crim. App. 2013) ; Blanton v. State , 2004 WL 3093219 , at *10 (Tex. Crim. App. June 30, 2004) ; Adair v. State , 336 S.W.3d 680 , 689 "(Tex. App.-Houston [1st. Dist.] 2010, pet. ref'd). Today's conclusion that Miller-El II 's stand-or-fall rule is inapplicable or weakened when a petitioner did not identify comparators during voir dire is unprecedented and contravenes that Supreme Court decision and federal and state cases applying it.

Snyder v. Louisiana , 552 U.S. 472 , 128 S.Ct. 1203 , 170 L.Ed.2d 175 (2008), also shows this error. In that case, Justice Thomas again voiced his view that comparative juror analysis should not be used to find a Batson violation when the defense "never mentioned [the inconsistent treatment] in the argument before the trial court." Snyder , 552 U.S. at 489 , 128 S.Ct. 1203 (2008) (Thomas, J., dissenting). Over this objection, the Court found a Batson violation based on a comparative juror analysis never raised in state court, focusing only on the reasons the prosecutor contemporaneously gave. Id . at 485-86, 128 S.Ct. 1203 .

The majority cites to Snyder 's cautioning "that a retrospective comparison of jurors based on a cold appellate record may be very misleading when alleged similarities were not raised at trial" so "an appellate court must be mindful that an exploration of the alleged similarities at the time of trial might have shown that the jurors in question were not really comparable." Id . at 483, 128 S.Ct. 1203 . But Snyder explains that the trial court had explored "the shared characteristic" that the prosecutor had stated was important-"concern about serving on the jury due to conflicting obligations"-with the relevant jurors, so the record enabled the appellate court to compare those jurors as to that cited characteristic . Id. The "cold appellate record" comment merely points out that how persuasive a comparative juror analysis is depends on how clear the record is about whether prospective jurors were similar as to the prosecutor's stated justification . Id . When the record is not clear about that similarity, a comparison is not helpful. See, e.g. , Puckett v. Epps , 641 F.3d 657 , 664 (5th Cir. 2011) (relying on Snyder in refusing to find a Batson violation because the record did not reveal whether jurors who "arguably would have fallen" into categories identified by the prosecutor as problematic actually did so). Indeed, the Sixth Circuit recently explained the lesson to draw from Snyder is that conclusions can fairly be made from an appellate record when:

(i) the government purportedly strikes a venireperson because of an answer to a question posed during voir dire ; (ii) venirepersons relevant to the comparison were asked the same question during voir dire ; (iii) the relevant venirepersons actually answered that question in similar depth; and (iv) the purpose of the analysis is to show that the government treated jurors with similar answers differently.

United States v. Atkins , 843 F.3d 625 , 636 (6th Cir. 2016). That is this case.

The Snyder concern about an undeveloped record on "substantial similarity" analysis is not present here. Because the prosecutor relied exclusively on three specific answers to questions on juror questionnaires, we are able to fully compare the jurors as to the only characteristics the prosecutor identified as relevant, and they are identical. This case thus does not confront the issue that is often the focus of the "substantial similarity" analysis and that Snyder 's "cold record" comment was addressing. The explanation a lawyer offers for a strike is usually much more general than the identical answers to specific questions identified here. A prosecutor may, for example, say a juror was struck because she "seemed to be anti-law enforcement." That explanation, and whether it applies equally to accepted jurors of a different race, likely does not just implicate answers to a single question. A host of matters in the jury selection record may inform whether the struck juror is similar to an accepted juror as to a general characterization like "anti-law enforcement." Relevant to that assessment might be answers to questions asking about one's views on criminal justice issues, as well as whether anyone in a person's family has been charged with a crime and the reaction to that experience, or whether a relative works in law enforcement. It is that situation in which Snyder raises a concern about being able to conduct a similarity inquiry on a cold record. As the D.C. Circuit has recognized, that concern is not present when prosecutors' stated reasons are narrow and specific as they are here. See United States v. Gooch , 665 F.3d 1318 , 1330 (D.C. Cir. 2012). Snyder demonstrates this: "There was only one alleged shared characteristic at issue in Snyder -jurors' concerns over having to commit to jury duty in the face of conflicting obligations. It was easy for the Court to sort out this one shared characteristic even on a cold appellate record." Id . (citing Snyder , 552 U.S. at 483 , 128 S.Ct. 1203 ). It is easy too for the three specific answers cited for striking Sturgis and Minor. 10

The State and majority opinion further contend that Miller-El II 's rule against after-the-fact justifications creates an unfair asymmetry in which the prosecution is held to the reasons it offered at trial whereas the defendant can wait until the appeal to identify jurors like Cooper who have the same answers as people who were struck. Whatever the soundness of this complaint, it is rejected by the leading decisions applying comparative juror analysis. See Miller-El II , 545 U.S. at 240-41 , 125 S.Ct. 2317 (conducting comparative analysis on habeas review despite no such analysis being presented to state courts); see also Reed , 555 F.3d at 372-75 (same); Woodward , 580 F.3d at 338 (same); Smith , 708 F.3d at 638 ("[A]lthough Smith did not point to specific jurors for comparative analysis, we have conducted an in-depth review of the record ...."). A fear that the prosecution will frequently have to explain "why it kept [a] white juror," Maj. Op. at 842, also ignores that the prosecutor only has to offer reasons of any type after a court has found a prima facie case of discrimination. At that stage, after the serious accusation of racial discrimination has been leveled and a preliminary case to support it recognized by the court, it does not seem too much to ask prosecutors to list all the reasons motivating their strikes. Miller-El II , 545 U.S. at 252 , 125 S.Ct. 2317 ("But when illegitimate grounds like race are in issue, a prosecutor simply has got to state his reasons as best he can and stand or fall on the plausibility of the reasons he gives."). If a concern about a black juror was important enough to be cited as a reason for the challenged strike, a white juror with the same problematic characteristic should also be on the prosecutor's mind or, even more easily detectable when the Batson claim was raised in this case, subject to the prosecutor's highlighter.

* * *

To sum up the discussion of comparative juror analysis, every one of the grounds on which the majority opinion tries to avoid the inescapable conclusion of pretext that flows from a comparison properly limited to the reasons stated at trial was also true of Miller-El II . It was a case involving AEDPA deference. It was a case in which the comparative juror analysis was not advanced by the defense at trial or on direct appeal. It was a case in which there were many differences between the struck and accepted jurors not cited at trial but that appellate judges could identify and speculate were reasons for either striking a comparator or for keeping another one. Yet the Supreme Court still found that our application of comparative juror analysis was in error. It is once again.

III.

Beyond its fundamental error of repeating our violation of Miller-El II 's stand-or-fall principle, the majority opinion does not even follow the new approach it creates. It says the prosecutor should be given a "chance to respond whenever the court engages in a comparative juror analysis." Maj. Op. 844. That opportunity can include providing a "supported basis for keeping a nonblack juror" that was not articulated at trial. Id .

But the prosecutor who exercised the challenged strikes at Chamberlin's trial has never responded to the comparison of the jurors who are identically situated as to the reason stated at trial. The majority opinion instead slams the door on the Batson claim based on speculative reasons offered in a brief by appellate attorneys who work in a different office than the trial lawyer who picked the jury. That is at odds with the majority opinion's explanation that a court would have to assess if the new, post-trial justification "provides an adequately redeeming reason to override the strike-worthy characteristics the non-black juror shares with the black jurors who were struck." Id . As with any inquiry into intent, that determination would seemingly have to include a credibility assessment of the new reasons the prosecutor cites for "keeping" the white juror. That evaluation of credibility has never happened in this case. Nor will it ever. The majority opinion does not remand for a hearing on the supposed reasons for "keeping" the white juror who gave identical answers to the struck black jurors on Questions 30, 34, and 35. It just accepts what is said in an appellate brief without the prosecutor who made the strikes ever having to provide an explanation or without any explanation ever having been tested in an adversarial process and then evaluated by a factfinder. As a result, there is nothing to ensure that the new, post-trial justification is anything more than an "afterthought." Miller-El II , 545 U.S. at 246 , 125 S.Ct. 2317 .

IV.

Chamberlin's crime was horrific. But for even the most gruesome of crimes with the most culpable of defendants, there are certain trial errors that so fundamentally infect the process ("structural error" is the legal term) that a new trial is required regardless of how strong the evidence against the defendant is. Sullivan v. Louisiana , 508 U.S. 275 , 281, 113 S.Ct. 2078 , 124 L.Ed.2d 182 (1993). Discrimination in jury selection is one. Scott v. Hubert , 610 Fed.Appx. 433 , 434 (5th Cir. 2015) ("[A] Batson violation would be a structural error" (analogizing to Vasquez v. Hillery , 474 U.S. 254 , 261-64, 106 S.Ct. 617 , 88 L.Ed.2d 598 (1986) )). Eliminating discrimination from jury selection does even more than guarantee a fair trial as important as that goal is; it also promotes confidence in the criminal justice system by ensuring that people of all backgrounds have the role in our courts the Constitution gives them.

Comparative juror analysis plays a crucial role in rooting out this discrimination under the Batson framework, which the Supreme Court has recognized may not fully capture discrimination:

Although the move from Swain to Batson left a defendant free to challenge the prosecution without having to cast Swain 's wide net, the net was not entirely consigned to history, for Batson 's individualized focus came with a weakness of its own owing to its very emphasis on the particular reasons a prosecutor might give. If any facially neutral reason sufficed to answer a Batson challenge, then Batson would not amount to much more than Swain.

Miller-El II , 545 U.S. at 239-40 , 125 S.Ct. 2317 ; see also id. at 270-71 , 125 S.Ct. 2317 (Breyer, J., concurring) (citing numerous sources in concluding that "the use of race- and gender-based stereotypes in the jury-selection process seems better organized and more systematized than ever before"). Comparative juror analysis is an attempt to rectify this weak link in the Batson framework: the risk that it "would become a 'mere exercise in thinking up any rational basis.' " Davis, 268 S.W.3d at 525 (Jefferson, C.J.) (quoting Miller-El II , 545 U.S. at 252 , 125 S.Ct. 2317 ). As mentioned at the outset, comparative juror analysis is the only tool that has ever enabled this court to find a Batson violation. It is largely neutered if an appellate court can come up with "any rational basis" that distinguishes jurors to undo a clear implication of pretext drawn from the reasons the lawyer provided at trial. With a precarious framework like Batson , any loosening of the reins can result in an empty harness.

More than mere loosening results from today's decision that defies precedent on the following important questions:

1. Whether the racial makeup of the overall strikes is relevant to the ultimate Batson discrimination inquiry concerning a particular strike.

2. Whether Miller-El II's stand-or-fall principle applies only at step 2 of Batson or also in making the final assessment of discriminatory intent.

3. Whether Miller-El II 's stand-or-fall principle only bars new post-trial reasons for striking a minority juror but allows new reasons for accepting white jurors.

4. Whether Miller-El II 's stand-or-fall principle applies only when defense counsel identified the comparison at trial.

Correction on these questions that are essential to the Batson framework is needed given the number of these claims raised in our circuit, often in capital cases.

In one of a series of criminal procedure cases the Supreme Court has recently decided that address discrimination in our justice system-three involving either jury selection or deliberations-it observed that "[t]he Nation must continue to make strides to overcome race-based discrimination." Pena-Rodriguez v. Colorado , --- U.S. ----, 137 S.Ct. 855 , 871, 197 L.Ed.2d 107 (2017) (jury deliberations); see also Tharpe v. Sellers , --- U.S. ----, 138 S.Ct. 545 , 199 L.Ed.2d 424 (2018) (same); Buck v. Davis , --- U.S. ----, 137 S.Ct. 759 , 197 L.Ed.2d 1 (2017) (racial testimony of an expert witness); Foster v. Chatman , --- U.S. ----, 136 S.Ct. 1737 , 195 L.Ed.2d 1 (2016) (jury selection). Today's decision strides in a different direction.