Opinion
CHIN, J.During jury selection, each party is entitled to a limited number of peremptory challenges. (Code Civ. Proc., § 231.) However, exercising peremptory challenges to remove prospective jurors solely because of group bias, for example, on racial grounds, violates both the California Constitution (People v. Wheeler (1978) 22 Cal.3d 258 [148 Cal.Rptr. 890, 583 P.2d *1306748] (Wheeler)) and the United States Constitution (Batson v. Kentucky (1986) 476 U.S. 79 [106 S.Ct. 1712, 90 L.Ed.2d 69] (Batson)). Both Wheeler and, later, Batson established procedures for courts to follow when one party objects to the other party’s peremptory challenges. Defendant contends that California’s procedures violate Batson in two respects.
First, although both Wheeler and Batson require the objector to establish a prima facie case of discriminatory use of peremptory challenges before the other party must explain its challenges, Wheeler used two terms— “strong likelihood” and “reasonable inference”—to describe the necessary showing of group bias; Batson used the single term, “an inference of discriminatory purpose.” (Batson, supra, 476 U.S. at p. 94 [106 S.Ct. at p. 1721]; Wheeler, supra, 22 Cal.3d at pp. 280-281.) Defendant argues that the “strong likelihood” standard states a different, and higher, requirement for establishing a prima facie case than Batson permits.
Second, we have observed that comparing, for the first time on appeal, the answers of excused jurors with those of jurors not excused to determine whether the trial court erred in denying an objection to the use of peremptory challenges is unreliable and fails to give due deference to the trial court’s ruling. (E.g., People v. Box (2000) 23 Cal.4th 1153, 1190 [99 Cal.Rptr.2d 69, 5 P.3d 130]; People v. Montiel (1993) 5 Cal.4th 877, 909 [21 Cal.Rptr.2d 705, 855 P.2d 1277]; People v. Johnson (1989) 47 Cal.3d 1194, 1220-1222 [255 Cal.Rptr. 569, 767 P.2d 1047].) Defendant contends that this rule against “comparative juror analysis” violates Batson.
We conclude that Wheeler’s terms, a “strong likelihood” and a “reasonable inference,” refer to the same test, and this test is consistent with Batson. Under both Wheeler and Batson, to state a prima facie case, the objector must show that it is more likely than not the other party’s peremptory challenges, if unexplained, were based on impermissible group bias. We also conclude that Batson does not require state reviewing courts to engage in comparative juror analysis for the first time on appeal. Finally, applying California’s procedures, which satisfy Batson, to this case, we uphold the trial court’s finding that defendant failed to establish a prima facie case that the prosecutor used his peremptory challenges improperly.
Accordingly, we reverse the judgment of the Court of Appeal, which reached contrary conclusions.
I. Facts and Procedural History
A jury found defendant guilty of the second degree murder of the 19-month-old daughter of his girlfriend and of assault resulting in the death of *1307a child under the age of eight. (Pen. Code, §§ 187, 273ab.) The issues before us solely involve jury selection, so we focus on that process.
The district attorney exercised 12 peremptory challenges. He used three of them to challenge all three African-American prospective jurors on the jury panel—C.T., S.E., and R.L. After the second of these challenges, defendant made a “Wheeler motion.” (Wheeler, supra, 22 Cal.3d 258.) He stated that his motion “[mjore specifically . . . concerns [S.E.], the last individual who was eliminated by the People.” He argued the prosecutor had no apparent reason to challenge this prospective juror “other than [her] racial identity.” He made no argument regarding C.T. The court responded that “based on the record that’s been made, [it] would find that there’s not been shown a strong likelihood that the exercise of the peremptory challenges were based upon a group rather than an individual basis. The Court has to start from the position of a premise that the exercises of the peremptory challenges were based on constitutional grounds.” The court also told the district attorney, however, that “we are very close.”
After the third of these challenges, defendant renewed his Wheeler motion. Focusing this time on the most recent challenge, he based his motion on the circumstance that the district attorney had removed all of the African-American prospective jurors. The court denied the motion in a detailed ruling. Regarding the most recent challenge, the court stated that it had had “concerns with regard to her qualifications in this matter based upon her answers on the questionnaire; specifically, the Court had noted that she had a sister who had had drug charges, although her answers in follow-up verbally were such that the Court would not have found that the issues were such to lead to a challenge for cause. May be sufficient to justify a peremptory challenge by the People, [f] Also, with regard to her answers generally on the questionnaire itself, there was an indication that she had difficulty understanding some of the issues, and specifically, her last response which was somewhat rambling on the questionnaire indicated that she herself felt that she had difficulty understanding things. Again, her verbal responses here in court were such that I would not have granted a challenge for cause on that basis, but the Court felt that the answers on the questionnaire were sufficient that they certainly would have justified a peremptory challenge by either side, frankly, based upon the concerns about her ability to understand the proceedings.”
The court noted the rest of the district attorney’s challenges were against “all other types of groups, including white women and white men as well.” Regarding S.E., the court stated that it had been concerned about her, “including the answer on question thirty that she gave verbally here in court *1308that she had not included in her questionnaire; a parent had a robbery or arrest, even though that was a number of years ago, thirty years ago. Her answers on the record were not as such that I would have granted a challenge for cause. Certainly, they could justify a challenge by the People, [f] She expressed on the record that she didn’t know if she could be fair. Her verbal follow-up responses were not such that I would have granted a challenge for cause. And also based upon the answers with regard to question sixty-three . . . her emotions and feelings, again, her answers were such that they may have caused concern for either side. Even though the answers tend to lean in the favor of the prosecution in the case, neither side would want a juror deciding a case based upon emotions, rather than the facts and the evidence.” “In summary,” the court said, “with regard to the jurors, while the Court would not grant the challenges for cause, there were answers ... at least on the questionnaires themselves such that the Court felt that there was sufficient basis for the peremptory challenge, [f] Even with the addition of [the most recent challenge], the Court will not find a prima facie case.”
The Court of Appeal reversed the judgment. It found that the “strong likelihood” standard the trial court applied violated Batson, supra, 416 U.S. 79. Based primarily on its own comparison of answers the challenged jurors gave with answers of nonchallenged jurors, the court concluded that “a prima facie case of group bias was established and that the judgment must therefore be reversed.” Justice Haerle dissented on all points.
We granted the Attorney General’s petition for review.
II. Discussion
A. Background
Exercising peremptory challenges because of group bias rather than for reasons specific to the challenged prospective juror violates both the California Constitution and the United States Constitution. (Batson, supra, 476 U.S. 79; Wheeler, supra, 22 Cal.3d 258.) Because Wheeler predated Batson, the Wheeler court obviously did not have the benefit of Batson in establishing the procedures to follow in California. Defendant argues that California’s procedures violate Batson in two respects: (1) Wheeler's “strong likelihood” standard, or at least the way it was understood and applied in later cases, is a higher standard than Batson permits; and (2) California law impermissibly restricts comparative juror analysis. We discuss these contentions in order below, then review the trial court’s rulings. But first, to fully understand these issues, it is necessary to review Wheeler and Batson in detail.
*13091. Wheeler
In Wheeler, we concluded “that the use of peremptory challenges to remove prospective jurors on the sole ground of group bias violates” the California Constitution. (Wheeler, supra, 22 Cal.3d at p. 276.) We then turned to the difficult question of what to do when one party objects to the other party’s peremptory challenges on this basis. We began “with the proposition that in any given instance the presumption must be that a party exercising a peremptory challenge is doing so on a constitutionally permissible ground. . . . [H] Yet it is only a presumption, and must be rebuttable if the foregoing constitutional right is not to be nullified even by honest zeal. The issue is what showing is necessary to rebut it. We must define a burden of proof which a party may reasonably be expected to sustain in meritorious cases, but which he cannot abuse to the detriment of the peremptory challenge system.” (Id. at p. 278.)
We adopted the following procedure: “If a party believes his opponent is using his peremptory challenges to strike jurors on the ground of group bias alone, he must raise the point in timely fashion and make a prima facie case of such discrimination to the satisfaction of the court. First, ... he should make as complete a record of the circumstances as is feasible. Second, he must establish that the persons excluded are members of a cognizable group within the meaning of the representative cross-section rule. Third, from all the circumstances of the case he must show a strong likelihood that such persons are being challenged because of their group association rather than because of any specific bias.” (Wheeler, supra, 22 Cal.3d at p. 280, fn. omitted.) We discussed types of evidence the objector may present to make this showing. “[T]he party may show that his opponent has struck most or all of the members of the identified group from the venire, or has used a disproportionate number of his peremptories against the group. He may also demonstrate that the jurors in question share only this one characteristic— their membership in the group—and that in all other respects they are as heterogeneous as the community as a whole. Next, the showing may be supplemented when appropriate by such circumstances as the failure of his opponent to engage these same jurors in more than desultory voir dire, or indeed to ask them any questions at all. Lastly, . . . the defendant need not be a member of the excluded group in order to complain of a violation of the representative cross-section rule; yet if he is, and especially if in addition his alleged victim is a member of the group to which the majority of the remaining jurors belong, these facts may also be called to the court’s attention.” (Id. at pp. 280-281, fn. omitted.)
We then discussed what the court must do. “Upon presentation of this and similar evidence—in the absence, of course, of the jury—the court must *1310determine whether a reasonable inference arises that peremptory challenges are being used on the ground of group bias alone. We recognize that such a ruling ‘requires trial judges to make difficult and often close judgments. They are in a good position to make such determinations, however, on the basis of their knowledge of local conditions and of local prosecutors.’ [Citation.] They are also well situated to bring to bear on this question their powers of observation, their understanding of trial techniques, and their broad judicial experience. We are confident of their ability to distinguish a true case of group discrimination by peremptory challenges from a spurious claim interposed simply for purposes of harassment or delay.
“If the court finds that a prima facie case has been made, the burden shifts to the other party to show if he can that the peremptory challenges in question were not predicated on group bias alone. The showing need not rise to the level of a challenge for cause. But to sustain his burden of justification, the allegedly offending party must satisfy the court that he exercised such peremptories on grounds that were reasonably relevant to the particular case on trial or its parties or witnesses—i.e., for reasons of specific bias as defined herein. He, too, may support his showing by reference to the totality of the circumstances: for example, it will be relevant if he can demonstrate that in the course of this same voir dire he also challenged similarly situated members of the majority group on identical or comparable grounds. And again we rely on the good judgment of the trial courts to distinguish bona fide reasons for such peremptories from sham excuses belatedly contrived to avoid admitting acts of group discrimination.
“If the court finds that the burden of justification is not sustained as to any of the questioned peremptory challenges, the presumption of their validity is rebutted.” (Wheeler, supra, 22 Cal.3d at pp. 281-282, fn. omitted.)
2. Batson
In Batson, the United States Supreme Court held that principles of equal protection “forbid discrimination on account of race in selection of the petit jury.” (Batson, supra, 476 U.S. at p. 88 [106 S.Ct. at p. 1718].) It, too, discussed what to do when one party objects to the other party’s peremptory challenges. “As in any equal protection case, the ‘burden is, of course,’ on the defendant who alleges discriminatory selection of the venire ‘to prove the existence of purposeful discrimination.’ [Citations.] In deciding if the defendant has carried his burden of persuasion, a court must undertake ‘a sensitive inquiry into such circumstantial and direct evidence of intent as may be available.’ [Citation.] Circumstantial evidence of invidious intent may be proof of disproportionate impact. [Citation.] We have observed that *1311under some circumstances proof of discriminatory impact ‘may for all practical purposes demonstrate unconstitutionality because in various circumstances the discrimination is very difficult to explain on non-racial grounds.’ [Citation.] For example, ‘total or seriously disproportionate exclusion of Negroes from jury venires,’ [citation] ‘is itself such an “unequal application of the law ... as to show intentional discrimination,” ’ [citations].” (Id. at p. 93 [106 S.Ct. at p. 1721].)
A party alleging discriminatory use of peremptories “may make out a prima facie case of purposeful discrimination by showing that the totality of the relevant facts gives rise to an inference of discriminatory purpose. [Citation.] Once the defendant makes the requisite showing, the burden shifts to the State to explain adequately the racial exclusion.” (Batson, supra, 476 U.S. at pp. 93-94 [106 S.Ct. at p. 1721].) In a footnote, the court stated that its “decisions concerning ‘disparate treatment’ under Title VII of the Civil Rights Act of 1964 have explained the operation of prima facie burden of proof rules. See McDonnell Douglas Corp. v. Green, 411 U.S. 792 [93 S.Ct. 1817, 36 L.Ed.2d 668] (1973); Texas Dept, of Community Affairs v. Burdine, 450 U.S. 248 [101 S.Ct. 1089, 67 L.Ed.2d 207] (1981); United States Postal Service Board of Governors v. Aikens, 460 U.S. 711 [103 S.Ct. 1478, 75 L.Ed.2d 403] (1983). The party alleging that he has been the victim of intentional discrimination carries the ultimate burden of persuasion. Texas Dept, of Community Affairs v. Burdine, supra, at 252-256 [101 S.Ct. at pp. 1093-1095].” (Id. at p. 94, fn. 18 [106 S.Ct. at pp. 1721-1722].)
The high court discussed' in detail the “standards for assessing a prima facie case in the context of discriminatory selection of the venire . . . .” (Batson, supra, 476 U.S. at p. 96 [106 S.Ct. at p. 1723].) “[A] defendant may establish a prima facie case of purposeful discrimination in selection of the petit jury solely on evidence concerning the prosecutor’s exercise of peremptory challenges at the defendant’s trial. To establish such a case, the defendant first must show that he is a member of a cognizable racial group [citation], arid that the prosecutor has exercised peremptory challenges to remove from the venire members of the defendant’s race.[1] Second, the defendant is entitled to rely on the fact, as to which there can be no dispute, that peremptory challenges constitute a jury selection practice that permits *1312‘those to discriminate who are of a mind to discriminate.’ [Citation.] Finally, the defendant must show that these facts and any other relevant circumstances raise an inference that the prosecutor used that practice to exclude the veniremen from the petit jury on account of their race. This combination of factors in the empanelling of the petit jury . . . raises the necessary inference of purposeful discrimination.
“In deciding whether the defendant has made the requisite showing, the trial court should consider all relevant circumstances. For example, a ‘pattern’ of strikes against black jurors included in the particular venire might give rise to an inference of discrimination. Similarly, the prosecutor’s questions and statements during voir dire examination and in exercising his challenges may support or refute an inference of discriminatory purpose. These examples are merely illustrative. We have confidence that trial judges, experienced in supervising voir dire, will be able to decide if the circumstances concerning the prosecutor’s use of peremptory challenges creates a prima facie case of discrimination against black jurors.” (Batson, supra, 476 U.S. at pp. 96-97 [106 S.Ct. at p. 1723].)
“Once the defendant makes a prima facie showing, the burden shifts to the State to come forward with a neutral explanation for challenging black jurors. . . . The prosecutor . . . must articulate a neutral explanation related to the particular case to be tried. The trial court then will have the duty to determine if the defendant has established purposeful discrimination.” (Batson, supra, 476 U.S. at pp. 97-98 [106 S.Ct. at p. 1723], fns. omitted.) In a footnote, the court also stated, “Since the trial judge’s findings in the context under consideration here largely will turn on evaluation of credibility, a reviewing court ordinarily should give those findings great deference.” (Id. at p. 98, fn. 21 [106 S.Ct. at p. 1724].)
In rejecting an argument that its holding could create serious administrative difficulties, the high court noted that “[i]n those States applying a version of the evidentiary standard we recognize today,” courts have not experienced these difficulties. (Batson, supra, 476 U.S. at p. 99 [106 S.Ct. at p. 1724].) “For example, in People v. Hall, 35 Cal.3d 161, 672 P.2d 854 (1983), the California Supreme Court found that there was no evidence to show that procedures implementing its version of this standard, imposed five years earlier, were burdensome for trial judges.” (Id. at p. 99, in. 23 [106 S.Ct. at p. 1724].)
B. The Necessary Showing for a Prima Facie Case
Wheeler used both the terms “strong likelihood” and “reasonable inference” in describing the standard for a prima facie case. (Wheeler, supra, *131322 Cal.3d at pp. 280, 281.) We believe it obvious that we considered the two terms to he different phrasing of the same standard. Language in a 1994 Court of Appeal decision, People v. Bernard (1994) 27 Cal.App.4th 458 [32 Cal.Rptr.2d 486], however, created some uncertainty. That case seemed to read the two terms as stating different standards, with “reasonable inference” being a lower standard than “strong likelihood.” (Id. at p. 465.)
In Wade v. Terhune (9th Cir. 2000) 202 F.3d 1190, the court found that, at least after People v. Bernard, supra, 27 Cal.App.4th 458, “California state courts have applied a lower standard of scrutiny to peremptory strikes than the federal Constitution permits.” (Wade v. Terhune, supra, at pp. 1196-1197.) The court noted also that, in recent years, this court has generally cited the “strong likelihood” language rather than the “reasonable inference” language. (Id. at p. 1197.) It believed that California courts give criminal defendants less protection than the United States Constitution requires “when they follow the Wheeler ‘strong likelihood’ test in determining whether a prima facie case has been established.” (Wade v. Terhune, supra, at p. 1197.) In the court’s view, “the Wheeler ‘strong likelihood’ test for a successful prima facie showing of bias is impermissibly stringent in comparison to the more generous Batson ‘inference’ test.” (Ibid) The court “therefore conclude[d] that California courts in following the ‘strong likelihood’ language of Wheeler are not applying the correct legal standard for a prima facie case under Batson.” (Ibid)
Shortly after Wade v. Terhune, supra, 202 F.3d 1190, and in response to it, we reiterated that “in California, a ‘strong likelihood’ means a ‘reasonable inference,’ ” and disapproved People v. Bernard, supra, 27 Cal.App.4th 458, “to the extent it is inconsistent with People v. Wheeler, supra, 22 Cal.3d at pages 280-281.” (People v. Box, supra, 23 Cal.4th at p. 1188, in. 7.) However, the Ninth Circuit Court of Appeals held that this action did not cure the problem it perceived, at least for cases tried after Bernard and before Box in which trial courts were “following Bernard's take on Wheeler.” (Cooperwood v. Cambra (9th Cir. 2001) 245 F.3d 1042, 1047.) It said that “regardless” of our action in People v. Box, supra, at page 1188, footnote 7, “whenever state courts use the ‘strong likelihood’ standard, . . . these courts are applying a lower standard of scrutiny to peremptory strikes than the federal Constitution permits.” (Cooperwood v. Cambra, supra, at p. 1047.) The majority below agreed and, accordingly, found that the trial court used an impermissible standard in determining whether defendant had established a prima facie case.
We reiterate what we implied in Wheeler and stated in Box: Wheeler's terms “strong likelihood” and “reasonable inference” state the same standard. (People v. Box, supra, 23 Cal.4th at p. 1188, fn. 7; Wheeler, supra, 22 *1314Cal.3d at pp. 280-281.) This has always been true, although we recognize that People v. Bernard, supra, 27 Cal.App.4th 458, may have created some uncertainty for a few years. But the important question for current purposes is not whether California courts have always viewed these terms as the same, but whether California’s “strong likelihood” standard violates Batson. The Wade v. Terhune court and the majority below assumed that Batson’s inference test is “more generous” than Wheeler’s strong likelihood test without examining what the United State Supreme Court itself has said on the matter. (Wade v. Terhune, supra, 202 F.3d at p. 1197.) But to address this question, it is necessary to delineate exactly what the high court meant when it equated the necessary prima facie case with “an inference of discriminatory purpose.” (Batson, supra, 476 U.S. at p. 94 [106 S.Ct. at p. 1721].) We believe the court has made its meaning clear.
The high court appears to have given other courts some flexibility in establishing the exact procedures to follow. “We decline, however, to formulate particular procedures to be followed upon a defendant’s timely objection to a prosecutor’s challenges.” (Batson, supra, 476 U.S. at p. 99 [106 S.Ct. at pp. 1724-1725].) Thus, to some extent at least, Batson seems to have “left to lower courts the task of determining the type and quantum of proof necessary for a defendant to establish a prima facie case.” (State v. Duncan (La. 2001) 802 So.2d 533, 545; see also Note, Batson v. Kentucky and the Prosecutorial Peremptory Challenge: Arbitrary and Capricious Equal Protection? (1988) 74 Va. L.Rev. 811, 817 [“The Supreme Court has left the lower courts to grapple with the proper implementation of the Batson test. They must determine not only the quantum of proof necessary for a defendant to establish a prima facie case but also the types of explanations sufficient to rebut the defendant’s claim”].) But whatever flexibility the high court has left state courts, the California standard is not less generous than the test the Batson court itself applies to establish a prima facie case.
In Batson, the court stated that its decisions under title VII of the Civil Rights Act of 1964 “explained the operation of prima facie burden of proof rules.” (Batson, supra, 416 U.S. at p. 94, fn. 18 [106 S.Ct. at pp. 1721-1722].) The court has since repeatedly cited title VII cases as authoritative in the Batson context. (Miller-El v. Cockrell (2003) 537 U.S. 322, 338 [123 S.Ct. 1029, 1041, 154 L.Ed.2d 931] (Miller-El); Purkett v. Elem (1995) 514 U.S. 765, 768 [115 S.Ct. 1769, 1771, 131 L.Ed.2d 834]; Hernandez v. New York (1991) 500 U.S. 352, 359 [111 S.Ct. 1859, 1866, 114 L.Ed.2d 395] (plur. opn.).) One of the decisions the court cited in Batson, McDonnell Douglas Corp. v. Green (1973) 411 U.S. 792 [93 S.Ct. 1817, 36 L.Ed.2d 668], made clear that the plaintiff carried the initial burden of showing actions “from which one can infer, if such actions remain unexplained, that *1315it is more likely than not that such actions were ‘based on a discriminatory criterion illegal under the Act.’ ” (Furnco Construction Corp. v. Waters (1978) 438 U.S. 567, 576 [98 S.Ct. 2943, 2949, 57 L.Ed.2d 957], italics added.) Another of the decisions that Batson cited explained that the burden is not “onerous,” but the party nevertheless had to prove “by the preponderance of the evidence a prima facie case of discrimination.” (Texas Dept, of Community Affairs v. Burdine (1981) 450 U.S. 248, 252-253 [101 S.Ct. 1089, 1093], italics added.) It stated that “the prima facie case ‘raises an inference of discrimination only because we presume these acts, if otherwise unexplained, are more likely than not based on the consideration of impermissible factors.’” (Id. at p. 254 [101 S.Ct. at p. 1094], italics added.)
In one of the cases Batson cited, the high court recognized that the term “prima facie case” is, by itself, ambiguous, so it specifically stated what it meant in this context. “The phrase ‘prima facie case’ not only may denote the establishment of a legally mandatory, rebuttable presumption, but also may be used by courts to describe the plaintiffs burden of producing enough evidence to permit the trier of fact to infer the fact at issue. 9 J. Wigmore, Evidence § 2494 (3d ed. 1940). McDonnell Douglas should have made it apparent that in the Title VII context we use ‘prima facie case’ in the former sense.” (Texas Dept, of Community Affairs v. Burdine, supra, 450 U.S. at p. 254, fn. 7 [101 S.Ct. at p. 1094].) Thus, when it refers to the objector establishing “an inference of discriminatory purpose” (Batson, supra, 476 U.S. at p. 94 [106 S.Ct. at p. 1721]), the high court means establishing a legally mandatory rebuttable presumption, and not merely presenting enough evidence to permit the inference.
Wigmore, in turn, goes into greater detail. He explains that the term “ ‘prima facie case’ is used in two senses . . . .” (9 Wigmore, Evidence (Chadboume rev. ed. 1981) § 2494, p. 378.)2 The sense the high court does not mean in this context created the lower of the two burdens—merely the “duty of producing some evidence” sufficient to allow a matter to go to the jury. (Wigmore, at p. 379, italics added.) Wigmore also describes the sense in which the high court uses the term. “[T]he term ‘prima facie’ is sometimes used as equivalent to the notion of a presumption, even in the strict sense of a ruling of the judge putting upon the opponent the duty of producing evidence.” (Ibid.) Thus, it applies “where the proponent, having the burden of proving the issue (i.e., the risk of nonpersuasion of the jury), has not only removed by sufficient evidence the duty of producing evidence to get past the judge or the jury, but has gone further, and, either by means *1316of a presumption or by a general mass of strong evidence, has entitled himself to a ruling that the opponent should fail if he does nothing more in the way of producing evidence.” (Ibid., italics added.)
Thus, Batson permits a court to require the objector to present, not merely “some evidence” permitting the inference, but “strong evidence” that makes discriminatory intent more likely than not if the challenges are not explained. Nothing suggests the high court has since modified Batson's approach. Indeed, in its most recent decision on this subject, it said it was considering “the three-step framework mandated by Batson and reaffirmed in our later precedents” and, in describing the prima facie case requirement, echoed Batson's “inference” language. (Miller-El, supra, 537 U.S. at pp. 338, 347 [123 S.Ct. at pp. 1040, 1045] [“the inference of discrimination to support a prima facie case”].)
Other states, although not all,3 have reached similar conclusions regarding Batson's meaning. Citing Texas Dept, of Community Affairs v. Burdine, supra, 450 U.S. at pages 252-253 [101 S.Ct. at page 1093], the Connecticut Supreme Court has said that under the first stage of the Batson inquiry, the “defendant must establish by a preponderance of the evidence a prima facie case of purposeful discrimination.” (State v. Gonzalez (1988) 206 Conn. 391 [538 A.2d 210, 213].) The Maryland Court of Appeals recognized that Batson cited title VII cases “for an explanation of the operation of prima facie burden of proof rules.” (Stanley v. State (1988) 313 Md. 50 [542 A.2d 1267, 1271].) The Maryland Court of Appeals also concluded that “to make the prima facie showing Batson requires, the moving party must ‘prove by a preponderance of the evidence that the peremptory challenges were exercised in a way that shifts the burden of production to the State and requires it to respond to the rebuttable presumption of purposeful discrimination that arises under certain circumstances.’ ” (Mejia v. State (1992) 328 Md. 522 [616 A.2d 356, 361], quoting Stanley v. State, supra, at p. 1277.)
Wheeler’s term “strong likelihood” signals that it meant “prima facie case” in a similar sense.4 This burden is not onerous (Texas Dept, of Community Affairs v. Burdine, supra, 450 U.S. at p. 253 [101 S.Ct. at pp. pp. 1093-1094]), but it is substantial. It is also appropriate to compel the objecting *1317party to meet this burden before it can force the other party to explain the reasons for its peremptory challenges. The burden is one “which a party may reasonably be expected to sustain in meritorious cases, but which he cannot abuse to the detriment of the peremptory challenge system.” (Wheeler, supra, 22 Cal.3d at p. 278.) The high court itself has noted that the Batson analysis “permits prompt rulings on objections to peremptory challenges without substantial disruption of the jury selection process.” (Hernandez v. New York, supra, 500 U.S. at p. 358 [111 S.Ct. at pp. 1865-1866] (plur. opn.); see also id. at p. 374 [111 S.Ct. at p. 1874] (cone. opn. of O’Connor, J.) [warning against “unacceptable delays in the trial process” that would be “antithetical to the nature and purpose of the peremptory challenge”].)
The term “strong likelihood” has never set a higher standard than Batson permits. Wheeler “define[d] a burden of proof . . . .” (Wheeler, supra, 22 Cal.3d at p. 278.) “Except as otherwise provided by law,” the default burden of proof in California is “proof by a preponderance of the evidence.” (Evid. Code, § 115; see Conservatorship of Wendland (2001) 26 Cal.4th 519, 546 [110 Cal.Rptr.2d 412, 28 P.3d 151].) Here, no other law has provided for a different standard. This court has never suggested that Wheeler imposed the burden of either clear and convincing proof or proof beyond a reasonable doubt, the two higher burdens of proof that exist in this state. (Evid. Code, § 115.) Moreover, the unqualified word “likelihood” would have been insufficient to signal the correct standard. That word, or the equivalent “likely,” has a range of meanings, including some lower than more likely than not. (See People v. Superior Court (Ghilotti) (2002) 27 Cal.4th 888, 917, 922 [119 Cal.Rptr.2d 1, 44 P.3d 949] [interpreting a statute using the word “likely” as not requiring a showing of a “better than even” chance].) Thus, qualifying “likelihood” with “strong” is appropriate and consistent with Batson.
Although not dispositive, the Batson court itself considered Wheeler's procedures comparable to its own. It did not specifically cite Wheeler's “strong likelihood” language, but it referred to this court’s “procedures implementing its version of [Batson's evidentiary] standard” without suggesting there was anything wrong with those procedures. (Batson, supra, 476 U.S. at p. 99, fn. 23 [106 S.Ct. at p. 1724].)
We disagree that the “strong likelihood” standard became too high in the interval between People v. Bernard, supra, 27 Cal.App.4th 458, and People *1318v. Box, supra, 23 Cal.4th 1153. Bernard did not somehow transform “strong likelihood” from a correct to an incorrect standard. Like the court in Wade v. Terhune, supra, 202 F.3d 1190, and the majority below, the Bernard court did not examine what Wheeler or the high court meant by the word “inference” in this context. It misunderstood that meaning. The Bernard court said the presumption that peremptory challenges are being used properly should be rebuttable “only by a strong showing, not a mere inference,” because a different standard “might easily transform removal of each and every prospective juror belonging to a cognizable group into a Wheeler hearing.” (People v. Bernard, supra, at p. 465.) It also referred to “a reduction of the prima facie standard to a ‘reasonable inference’ test.” (Ibid.) This use of the word “inference” was akin to Wigmore’s lower sense of evidence merely permitting one to draw an inference, not the higher sense of Batson and Wheeler. Thus, to the extent Bernard erred, it was in assuming “inference” was lower than it really is in this context, not in raising “strong likelihood” to an impermissibly high standard.
Accordingly, Wheeler’s standard for establishing a prima facie case of discriminatory use of peremptory challenges is, and always has been, compatible with Batson. It merely means that to state a prima facie case, the objector must show that it is more likely than not the other party’s peremptory challenges, if unexplained, were based on impermissible group bias. The trial court here properly cited the Wheeler standard in determining whether defendant had established a prima facie case.
C. Appellate Use of Comparative Juror Analysis
The majority below compared the answers of challenged jurors with those of nonchallenged jurors, a comparison not done at trial, in order to overturn the trial court’s finding of no prima facie case. Defendant argues that doing so was proper and that our cases, which he interprets as prohibiting all comparative juror analysis, violate Batson, supra, 476 U.S. 79. Defendant misunderstands our jurisprudence. We have observed that engaging in comparative juror analysis for the first time on appeal is unreliable and inconsistent with the deference reviewing courts necessarily give to trial courts, but we have never prohibited trial courts from doing so or the party objecting to the challenges from relying on such analysis in seeking to make a prima facie case.
In People v. Johnson, supra, 47 Cal.3d 1194, we revisited People v. Trevino (1985) 39 Cal.3d 667 [217 Cal.Rptr. 652, 704 P.2d 719], which had engaged in comparative juror analysis for the first time on appeal. We found that Trevino had “placed undue emphasis on comparisons of the stated *1319reasons for the challenged excusáis with similar characteristics of nonmembers of the group who were not challenged by the prosecutor. First, we note . . . that the comparison is one-sided since it ignores the characteristics of the other . . . jurors against whom the prosecutor also exercised peremptory challenges. [Citation.] Moreover, we fail to see how a trial judge can reasonably be expected to make such detailed comparisons mid-trial.” (People v. Johnson, supra, at p. 1220.)
We explained that “use of a comparison analysis to evaluate the bona tides of the prosecutor’s stated reasons for peremptory challenges does not properly take into account the variety of factors and considerations that go into a lawyer’s decision to select certain jurors while challenging others that appear to be similar. Trial lawyers recognize that it is a combination of factors rather than any single one which often leads to the exercise of a peremptory challenge. In addition, the particular combination or mix of jurors which a lawyer seeks may, and often does, change as certain jurors are removed or seated in the jury box.” (People v. Johnson, supra, 47 Cal.3d at p. 1220.)5 We found it apparent “that the very dynamics of the jury selection process make it difficult, if not impossible, on a cold record, to evaluate or compare the peremptory challenge of one juror with the retention of another juror which on paper appears to be substantially similar. [Attempting] to make such an analysis of the prosecutor’s use of his peremptory challenges is highly speculative and less reliable than the determination made by the trial judge who witnessed the process by which the defendant’s jury was selected. It is therefore with good reason that we and the United States Supreme Court give great deference to the trial court’s determination that the use of peremptory challenges was not for an improper or class bias purpose.” (Id. at p. 1221.)
Accordingly, we disapproved People v. Trevino, supra, 39 Cal.3d 667, to the extent it was inconsistent with these views and returned to a “standard of truly giving great deference to the trial court in distinguishing bona fide reasons from sham excuses.” (People v. Johnson, supra, 47 Cal.3d at p. 1221.) Like the high court, we stated we “ ‘may assume that trial judges, in supervising voir dire in light of our decision today, will be alert to identify a prima facie case of purposeful discrimination.’ ” (Ibid., quoting Batson, supra, 476 U.S. at p. 99, fn. 22 [106 S.Ct. at p. 1724].)
Since then, we have not engaged in comparative juror analysis for the first time on appeal. “If the trial court makes a ‘sincere and reasoned *1320effort’ to evaluate the nondiscriminatory justifications offered, its conclusions are entitled to deference on appeal. In such circumstances, an appellate court will not reassess good faith by conducting its own comparative juror analysis. Such an approach would undermine the trial court’s credibility determinations and would discount ‘ “the variety of [subjective] factors and considerations,” ’ including ‘prospective jurors’ body language or manner of answering questions,’ which legitimately inform a trial lawyer’s decision to exercise peremptory challenges.” (People v. Montiel, supra, 5 Cal.4th at p. 909.)
Defendant argues that any rule against comparative juror analysis is invalid but, as a backup position, he also argues that, at least, the rule is limited to the second stage of the trial court’s duty—judging the validity of the reasons for the exercising of challenges after it has found a prima facie case—and does not apply to the first stage—judging whether the objector has established a prima facie case to begin with. We disagree. Although some of our decisions, including People v. Johnson, supra, 47 Cal.3d 1194, did involve the second stage, others reviewed trial court findings of no prima facie case. (E.g., People v. Box, supra, 23 Cal.4th at pp. 1188, 1190; People v. Turner (1994) 8 Cal.4th 137, 167, 169-170 [32 Cal.Rptr.2d 762, 878 P.2d 521].) Differences do exist in the two procedural postures. At the first stage, the party making the challenges is not asked to explain them, so the trial court does not have to judge that party’s credibility. But the concerns about the inability of a reviewing court to judge the dynamics of jury selection on a cold record apply to both stages. A comparison of the jurors’ answers is unreliable when divorced from the context of the trial. A trial court, but not a reviewing court, is able to place the answers into context and draw meaning from all the circumstances, including matters not discemable from the record.
Even the Ninth Circuit Court of Appeals, which has a different practice in this regard than we (Burks v. Borg (9th Cir. 1994) 27 F.3d 1424, 1427), reviews a trial court’s finding of no prima facie case deferentially. (Tolbert v. Page (9th Cir. 1999) 182 F.3d 677 (in bank).) Its reasoning for giving such deference echoes ours both in giving similar deference and in not engaging in comparative juror analysis for the first time on appeal. “[T]he trial court is better positioned to decide the Batson prima facie issue, which involves a ‘ “factual inquiry” that “takes into account all the possible explanatory factors” in the particular case.’ [Citation.] Whether or not ‘all the relevant circumstances’ ‘raise an inference’ of discrimination will depend on factors such as the attitude and behavior of the challenging attorney and the prospective jurors manifested during voir dire. As a purely practical matter, the trial judge’s unique perspective of voir dire enables the judge to have *1321first-hand knowledge and observation of critical events. [Citation.] The trial judge personally witnesses the totality of circumstances that comprises the ‘factual inquiry,’ including the jurors’ demeanor and tone of voice as they answer questions and counsel’s demeanor and tone of voice in posing the questions. [Citation.] The trial judge is able to observe a juror’s attention span, alertness, and interest in the proceedings and thus will have a sense of whether the prosecutor’s challenge can be readily explained by a legitimate reason. ... In addition, the trial court is ‘experienced in supervising voir dire.’ [Citations.]
“The appellate court, on the other hand, must judge the existence of a prima facie case from a cold record. An appellate court can read a transcript of the voir dire, but it is not privy to the unspoken atmosphere of the trial court—the nuance, demeanor, body language, expression and gestures of the various players. [Citation.] . . . [T]he prima facie inquiry is so fact-intensive and so dependent on first-hand observations made in open court that the trial court is better positioned to decide the issue . . . .” (Tolbert v. Page, supra, 182 F.3d at pp. 683-684.)
“While [People v. Johnson, supra, 47 Cal.3d 1194] limits the scope of appellate review of a trial court’s Batson determination, it does not preclude comparative analysis by the trial court.” (Burks v. Borg, supra, 27 F.3d at p. 1428.) We do not, and cannot reasonably, require trial judges to perform such analysis itself (People v. Johnson, supra, 47 Cal.3d at p. 1220), but the objecting party may make the argument for the trial court to evaluate. For example, in People v. Crittenden (1994) 9 Cal.4th 83 [36 Cal.Rptr.2d 474, 885 P.2d 887], the defendant presented at trial “evidence, compiled from the biographical information of 50 prospective jurors,” and compared those jurors’ answers in support of his argument that the prosecutor improperly challenged a particular juror. (Id. at p. 116.) As the reviewing court, we considered this and other evidence and concluded that “[i]n light of all the relevant circumstances, the trial court properly could find that defendant had not made a prima facie showing . . . .” {Id. at p. 119.)
In support of his position, defendant cites Miller-El, supra, 537 U.S. 322 [123 S.Ct. 1029], but that case merely provides another example of a reviewing court considering evidence of comparative juror analysis after it had been presented to the trial court. Miller-El was tried before the decision in Batson, supra, 476 U.S. 79. At trial, the defendant had objected that the prosecution was using peremptory challenges discriminatorily, but the Bat-son procedures had not yet been established. Batson was decided while the case was on appeal. The Texas Court of Criminal Appeals remanded the case to the original trial court for new findings in light of Batson. Defendant *1322presented to the trial court several types of evidence supporting a prima facie case. Before the original trial, he had presented “evidence relating to a pattern and practice of race discrimination in the voir dire.” (.Miller-El, supra, 537 U.S. at p. 331 [123 S.Ct. at p. 1036].) Then, two years later, “he presented, to the same state [trial] court, evidence that directly related to the conduct of the prosecutors in his case.” (Ibid) Some of the latter evidence was comparative juror analysis of the type offered for the first time on appeal here. (Id. at p. 343 [123 S.Ct. at p. 1043].) After the state courts denied relief, the federal district court denied relief on federal habeas corpus, and the appellate court denied a certificate of appealability which, under federal law, prevented the defendant from appealing the matter. The United States Supreme Court reversed the denial of a certificate of appealability. It did not directly address the Batson question, but rather determined that the issue was “debatable” and, under federal habeas corpus law, the appellate court should have considered it on the merits. (Id. at p. 348 [123 S.Ct. at p. 1045].) Its discussion made clear that the comparative juror analysis presented to the trial court was among the evidence reviewing courts should consider. (Id. at p. 343 [123 S.Ct. at p. 1043].) This conclusion is consistent with our practice. Nothing in Miller-El, however, suggests that a reviewing court must engage in comparative juror analysis for the first time on appeal.
Both Wheeler and Batson place the burden of making the prima facie showing on the objecting party. (Batson, supra, 476 U.S. at p. 96 [“the defendant must show that these facts and any other relevant circumstances raise an inference that the prosecutor used that practice to exclude the veniremen from the petit jury on account of their race”]; Wheeler, supra, 22 Cal.3d at p. 280 [the objecting party must “make a prima facie case of. . . discrimination to the satisfaction of the court,” should “make as complete a record of the circumstances as is feasible,” and “from all the circumstances of the case . . . must show a strong likelihood” of improper challenges].) Neither decision requires a reviewing court to search the record itself for evidence that might have supplemented the objector’s showing. Nor must the trial court consider arguments not made and evidence not presented, although nothing prevents it from doing so in judging all of the circumstances.
Defendant argues that our statement that we cannot expect a trial judge “to make such detailed comparisons mid-trial” (People v. Johnson, supra, 47 Cal.3d at p. 1220) “overlooks the trial lawyers’ role.” He argues that the attorneys can make these arguments to assist the trial court “because, under the adversary system, the job of marshaling the relevant evidence is performed in the first instance by the lawyers. Accordingly, comparative juror analysis can easily and realistically be employed by state trial judges, too.” We agree. Indeed, as noted, People v. Crittenden, supra, 9 Cal.4th 83, *1323provides an example of the attorneys doing just this for the trial court’s consideration. But defendant did not make this kind of showing at trial, and we cannot expect the trial court to do so itself.
We have also said that comparative juror analysis is “largely beside the point” because of the legitimate subjective concerns that go into selecting a jury. (People v. Arias (1996) 13 Cal.4th 92, 136, fn. 16 [51 Cal.Rptr.2d 770, 913 P.2d 980].) But it is not irrelevant. Although such analysis, by itself, proves little, properly presented to the trial court, it can be among the “all relevant circumstances” (Batson, supra, 476 U.S. at pp. 96-97 [106 S.Ct. at p. 1723]) or “all the circumstances of the case” (Wheeler, supra, 22 Cal.3d at p. 280) that the trial court must consider in making its determination.
In People v. Howard (1992) 1 Cal.4th 1132 [5 Cal.Rptr.2d 268, 824 P.2d 1315], we did not “limit[] our review . . . solely to counsel’s presentation at the time of the motion. This is because other circumstances might support the finding of a prima facie case even though a defendant’s showing [was itself inadequate]. Nor should the trial court blind itself to everything except defense counsel’s presentation. Indeed, we have emphasized that such rulings require trial judges to consider ‘all the circumstances of the case’ [citation] and call upon judges’ ‘ “powers of observation, their understanding of trial techniques, and their broad judicial experience.” ’ [Citations.] The trial judge in this case, for example, obviously knew that defendant belonged to the same group as the challenged jurors and that his victims did not. Clearly these are relevant factors [citation], and they were apparent to the trial court even though defendant did not mention them during his Wheeler motion.” (Id. at p. 1155.) “For these reasons,” we said, “when a trial court denies a Wheeler motion without finding a prima facie case of group bias the reviewing court considers the entire record of voir dire.” {Ibid.) Certainly, the trial court should consider obvious matters, and it can consider any other circumstances it finds relevant in the particular case. But, midtrial, we cannot expect, and do not demand, trial courts to engage sua sponte in the sort of comparative juror analysis that appellate lawyers and courts can do after scouring the often lengthy appellate record during the appeal. And, given the inability of reviewing courts to reliably conduct such analysis on a cold record, those courts are not required to do so for the first time on appeal.
Defendant cites a number of federal cases that engage in comparative juror analysis in a different way than we do. The Ninth Circuit, for example, has stated that its practice is different than ours. (Burks v. Borg, supra, 27 *1324F.3d at p. 1427.)6 Other courts may certainly adopt different procedures than we. But we do not believe that comparative juror analysis for the first time on appeal is constitutionally compelled. The Batson court itself stated that in deciding whether the defendant has made the necessary showing, “the trial court should consider all relevant circumstances.” (Batson, supra, 476 U.S. at pp. 96-97 [106 S.Ct. at p. 1723], italics added.) It relies heavily on “trial judges, experienced in supervising voir dire,” to make this determination. (Id. at p. 97 [106 S.Ct. at p. 1723].) “During jury selection, the entire res gestae take place in front of the trial judge. Because the judge has before him the entire venire, he is well situated to detect whether a challenge to the seating of one juror is part of a ‘pattern’ of singling out members of a single race for peremptory challenges.” (United States v. Armstrong (1996) 517 U.S. 456, 467-468 [116 S.Ct. 1480, 1488, 134 L.Ed.2d 687].) The “entire res gestae” do not take place in front of an appellate court.
The Batson court rejected the argument that its holding would “create serious administrative difficulties” and noted that California had not found its own version to be “burdensome for trial judges.” (Batson, supra, 476 U.S. at p. 99 & fn. 23 [106 S.Ct. at p. 1724], citing People v. Hall (1983) 35 Cal.3d 161 [197 Cal.Rptr. 71, 672 P.2d 854].) However, requiring trial courts to engage in comparative juror analysis sua sponte in the middle of the trial would be burdensome. Moreover, permitting appellate courts to overturn trial court decisions based on their own comparative analysis of a cold record, divorced from the nuances of trial not apparent from the record, is inconsistent with the deference reviewing courts necessarily give trial courts. We see nothing in the high court decisions requiring us to defer less to trial courts or engage in our own comparative juror analysis for the first time on appeal.7
Accordingly, we maintain our long-standing practice. When the objecting party presents comparative juror analysis to the trial court, the *1325reviewing court must consider that evidence, along with everything else of relevance, in reviewing, deferentially, the trial court’s ruling. When such an analysis was not presented at trial, a reviewing court should not attempt its own comparative juror analysis for the first time on appeal, especially when, as here, the record supports the trial court’s finding of no prima facie case. While we decline to prohibit the practice outright, we are hard pressed to envision a scenario where comparative juror analysis for the first time on appeal would be fruitful or appropriate.
D. The Trial Court’s Ruling in This Case
We have already alluded to the standard of appellate review of a trial court’s finding of no prima facie case. Because these rulings call upon trial judges’ personal observations, we review them with considerable deference, (People v. Jones (1998) 17 Cal.4th 279, 294 [70 Cal.Rptr.2d 793, 949 P.2d 890].) Of course, “deference is not abdication . . . .” (People v. Scott (1997) 15 Cal.4th 1188, 1212 [65 Cal.Rptr.2d 240, 939 P.2d 354]; see also Miller-El, supra, 537 U.S. at p. 340 [123 S.Ct. at p. 1041] [“deference does not imply abandonment or abdication of judicial review”].) Accordingly, we review the entire record for evidence supporting the trial court’s ruling. But if the record suggests grounds on which the prosecutor might reasonably have challenged the jurors, we affirm that ruling. (People v. Howard, supra, 1 Cal.4th at p. 1155.)
The record here suggests grounds for the prosecutor to have reasonably challenged these jurors. The trial judge, “who had performed much of and observed the remainder of the voir dire, [and thus] was in the best position to determine under ‘all the relevant circumstances’ of the case” whether a prima facie showing existed (People v. Box, supra, 23 Cal.4th at p. 1189), mentioned reasons as to two of the jurors. The court noted that these reasons did not warrant a challenge for cause, but peremptory challenges need not be based on grounds for a challenge for cause. (People v. Jones, supra, 17 Cal.4th at p. 294.) So far as we can review the court’s reasons on *1326the cold record, the record supports them. Defendant has a different view of the significance of these jurors’ answers and courtroom statements, but we see no basis on which to overturn the trial court’s determinations.
As the majority below and defendant point out, the trial court did not specifically discuss the third of the challenged jurors, C.T., who was actually the first one challenged. But at trial, defendant did not argue that no reason existed to challenge that juror. His first Wheeler motion was “specifically” directed towards the challenge of S.E. His second Wheeler motion also never referred to C.T. The court acted reasonably in discussing the arguments defendant made, not arguments he did not make. Moreover, as the dissent below noted, the record discloses race-neutral grounds for challenging C.T.: “(1) she was childless (this case involved the death and alleged abuse of a minor), (2) the police had made no arrest after the robbery of her home five or six years ago, and (3) she omitted to answer the two questions in the questionnaire dealing with her opinions of prosecuting and defending attorneys.” (Fn. omitted.) As that dissent also noted, “ Tack of family may have appeared relevant to the prosecutor in a case involving child abuse and reasonably could be deemed to constitute a non-discriminatory basis for striking the venireman.’ ” (Quoting U.S. v. Lewis (9th Cir. 1988) 837 F.2d 415, 417.)
Defendant stresses that the district attorney used three of his 12 peremptory challenges to remove all three African-American prospective jurors, and this case involves an African-American defendant charged with killing “his White girlfriend’s child.” These circumstances are obviously highly relevant to whether a prima facie case existed. (Wheeler, supra, 22 Cal.3d at pp. 280-281.) They definitely warranted the trial court’s careful scrutiny, which that court gave. The court considered the question close but found no prima facie case under all the circumstances. We will not second-guess its determination by attempting a comparative juror analysis for the first time on appeal.
An additional note about the circumstances of this case is in order. Viewing these circumstances in isolation, it certainly looks suspicious that all three African-American prospective jurors were removed from the jury. But viewing a case like this in isolation is all a reviewing court can do. When this issue comes before a reviewing court, the circumstances often will be akin to those here. A reviewing court does not review the ordinary, nonsuspicious cases of jury selection. (Nor does it review those cases in which the trial court grants a Wheeler motion, so it also never sees those rulings.) A reviewing court does not see the big picture; it cannot place a case like this into perspective. It cannot know whether a case like this is typical, thus suggesting a real problem, or merely a statistical aberration of *1327the type that will inevitably occur occasionally given such a small sampling. The trial court, however, is capable of seeing the big picture. It can place a specific trial in a county into perspective. Trial judges “ ‘are in a good position to make such determinations ... on the basis of their knowledge of local conditions and of local prosecutors.’ ” (Wheeler, supra, 22 Cal.3d at p. 281.) This is another reason we must, and can, rely on trial courts to determine, from all the relevant circumstances, whether a prima facie case of discriminatory use of peremptory challenges exists.
Defendant cites Miller-El, supra, 537 U.S. 322 [123 S.Ct. 1029], in support of his argument that a prima facie case existed here. In Miller-El, the prosecution conceded in the high court that the defendant had established a prima facie case. (Id. at p. 338 [123 S.Ct. at p. 1040].) But the evidence the defendant presented in that case was far stronger than here. The defendant there presented several types of evidence to the original trial court: (1) the statistical disparity of peremptory challenges of African-Americans (10 of 11 were challenged) compared to others (four of 31 were challenged); (2) a comparison of the manner in which the prosecution questioned African-Americans and others, which suggested “that the manner in which members of the venire were questioned varied by race” (id. at p. 332 [123 S.Ct. at p. 1037]); (3) a comparison of the answers of prospective jurors akin to the comparative juror analysis defendant attempts here to present for the first time on appeal (id. at p. 344 [123 S.Ct. at p. 1043]); (4) the prosecution’s use of a procedural practice unknown in California called “jury shuffling,” which can be used to manipulate which prospective jurors are actually called;8 and (5) evidence that the district attorney’s office historically had a pattern and practice of racial discrimination in voir dire.
Here, defendant’s showing in the trial court consisted primarily of the statistical disparity of peremptory challenges between African-Americans and others. Regarding this type of evidence, Miller-El stated that “the statistical evidence alone raises some debate as to whether the prosecution acted with a race-based reason when striking prospective jurors. The prosecutors used their peremptory strikes to exclude 91% of the eligible African-American venire members, and only one served on petitioner’s jury. In total, 10 of the prosecutors’ 14 peremptory strikes were used against African-Americans. Happenstance is unlikely to produce this disparity.” (Miller-El, supra, 537 U.S. at p. 342 [123 S.Ct. at p. 1042].) Here, the number of strikes is smaller, and thus perhaps more explainable by happenstance: three of the *1328prosecutors’ 12 peremptory strikes were used against African-Americans. In one respect, however, the percentages are even worse: the prosecutor challenged all three (100 percent) of the eligible African-Americans. But the high court did not cite the statistical evidence to show that it alone necessarily established a prima facie case, but only to support the overall conclusion that the merits of the habeas corpus petition were sufficiently debatable that the federal appellate court should have permitted an appeal. Here, both the trial court and this court have considered the issue on the merits. The statistics here are indeed troubling and, as the trial court stated, the question was close, but, for the reasons explained, we see no basis to overturn the trial court’s carefully considered ruling.
Defendant and the dissent (dis. opn., post, at p. 1340) also argue that the district attorney asked no questions of the African-American jurors he challenged. This circumstance may also be relevant. (Wheeler, supra, 22 Cal.3d at p. 281.) But this trial occurred in 1998, at a time the trial court had primary responsibility for conducting voir dire. (Code Civ. Proc., former § 223, as added by Prop. 115, approved by voters at Prim. Elec., June 5, 1990.) The district attorney asked no questions of any prospective juror, including the nine of other ethnic groups he also challenged. Thus, asking no questions was of little or no significance here.
III. Conclusion
California’s reviewing courts are as ready as any to combat the pernicious effects of racial and other improper discrimination, in jury selection as in other matters. (E.g., People v. Silva (2001) 25 Cal.4th 345, 386 [106 Cal.Rptr.2d 93, 21 P.3d 769] [reversing a sentence of death because of error as to a single prospective juror].) But, like the United States Supreme Court, we necessarily, yet confidently, rely primarily on the good judgment of our trial courts to make the difficult determinations the procedures established in Batson, supra, 476 U.S. 79, and Wheeler, supra, 22 Cal.3d 258, require. The record in this case provides no reason to believe the trial court here failed to perform its duty properly.
Accordingly, we reverse the judgment of the Court of Appeal and remand the matter for further proceedings consistent with our opinion.
George, C. J., Baxter, J., Brown, J., and Moreno, J., concurred.
Both Wheeler, supra, 22 Cal.3d 258, and Batson, supra, 476 U.S. 79, like this case, involved criminal defendants claiming prosecutors impermissibly challenged members of the defendants’ own ethnic group. Other cases have expanded the prohibition into other circumstances. (E.g., J.E.B. v. Alabama ex rel. T.B. (1994) 511 U.S. 127 [114 S.Ct. 1419, 128 L.Ed.2d 89] [gender]; Georgia v. McCollum (1992) 505 U.S. 42 [112 S.Ct. 2348, 120 L.Ed.2d 33] [defendant’s exercise of peremptory challenges]; Edmonson v. Leesville Concrete Co., Inc. (1991) 500 U.S. 614 [111 S.Ct. 2077, 114 L.Ed.2d 660] [civil cases]; Powers v. Ohio (1991) 499 U.S. 400 [111 S.Ct. 1364, 113 L.Ed.2d 411] [defendant need not be member of the *1312excluded group]; People v. Willis (2002) 27 Cal.4th 811 [118 Cal.Rptr.2d 301, 43 P.3d 130] [defendant’s exercise of peremptory challenges].) We need not consider these cases here.
We are quoting the most recent revision of this work rather than the earlier version the high court cited in Texas Dept, of Community Affairs v. Burdine, supra, 450 U.S. 248; the two versions are identical as relevant.
The Colorado Supreme Court, for example, has interpreted Batson as not requiring the defendant “to prove by a preponderance of the evidence that discrimination occurred. Rather, the defendant must present evidence sufficient to raise an inference that discrimination occurred.” (Valdez v. People (Colo. 1998) 966 P.2d 587, 590.) But the court did not examine what the high court itself has stated on the question.
The dissent claims Evidence Code section 600, subdivision (b), which defines an inference as “a deduction of fact that may logically and reasonably be drawn from another fact or group of facts found or otherwise established in the action,” somehow supports its position. (Dis. opn., post, at p. 1334.) The Wheeler court certainly understood that an inference is a logical *1317deduction of fact. But the question here, which Evidence Code section 600 does not address, is how strong the inference must be. In a criminal trial, for example, an inference of guilt must be beyond a reasonable doubt. The dissent’s own unique proposed test—“a substantial danger that the prosecutor’s challenges were based on improper grounds”—is not found in Evidence Code section 600. (Dis. opn., post, at p. 1334.)
This case presents an example. The district attorney accepted the jury four times with C.T. on it before he peremptorily challenged her, apparently due to a change in the jury’s composition.
That court also noted that the United States Supreme Court “has not yet ruled on the role of comparative analysis on appellate review, so no one is quite sure whether our circuit or the California Supreme Court is right.” (Burks v. Borg, supra, 27 F.3d at p. 1427.)
The dissent claims that our reliance on People v. Johnson, supra, 47 Cal.3d 1194, is misplaced because Johnson barred using comparative juror analysis on appeal whether or not the parties raised that approach in the trial court. (Dis. opn., post, at pp. 1330-1332.) Contrary to the dissent, Johnson’s underlying rationale was predicated on the fact that the defendant made no comparative juror argument at trial. We reasoned that (1) a trial judge could not reasonably be expected to make detailed comparisons of jurors midtrial sua sponte, and (2) without explanation at trial from the party who exercised the challenge, comparative juror analysis could not accurately take into account the various factors that influence the decision to make a peremptory challenge. {Johnson, supra, at pp. 1220-1221.) We then concluded, “It should be apparent, therefore, that the very dynamics of the jury selection process make it difficult, if not impossible, on a cold record, to evaluate or compare the peremptory challenge of one Juror with the retention of another juror which on paper appears to be substantially similar.” {Id. at p. 1221, italics added.) In other words, the trial and appellate courts are in no *1325position to engage in comparative juror analysis on their own, without the issue first being raised by the parties at trial. Our approach is entirely consistent with Johnson.
Contrary to the dissent, People v. Johnson, supra, 47 Cal.3d 1194, is also entirely consistent with Miller-El. In that case, the high court had “no difficulty in using comparative juror analysis” (dis. opn., post, at p. 1332) because it was simply reviewing the state court record, i.e., comparative juror evidence that the defendant had first presented to the trial judge in support of his Batson motion. {Miller-El, supra, 537 U.S. at pp. 328, 344-347 [123 S.Ct. at pp. 1035-1036, 1043-1045].) Evidently, none of the parties disputed that a comparative juror analysis is a proper or necessary procedure in deciding a Batson claim. Thus, that issue was not before the high court, which only addressed whether the underlying evidence before the state trial court supported the issuance of a certificate of appealability. {Id. at pp. 341-347 [123 S.Ct. at pp. 1042-1045].)
“On at least two occasions the prosecution requested shuffles when there were a predominate number of African-Americans in the front of the panel,” with the result that the African-Americans were spread out more randomly and thus less likely to be called. (MillerEl, supra, 537 U.S. at p. 334 [123 S.Ct. at p. 1038].)