ANDERSON, Circuit Judge:

In this habeas corpus proceeding, we are faced with a challenge to the state's exercise of peremptory jury strikes in a criminal case. The challenge is brought under Batson v. Kentucky, 476 U.S. 79, 106 S.Ct. 1712, 90 L.Ed.2d 69 (1986), and its progeny. Petitioner was tried and convicted of felony murder in an Alabama state proceeding. During jury selection, the prosecution used nine of its fourteen peremptory strikes on black potential jurors. Petitioner's trial counsel raised an objection under Batson, which was rejected by the trial court. On appeal, the Alabama Court of Criminal Appeals remanded the case to the trial court for a hearing to determine whether the state's exercise of peremptory challenges was racially discriminatory under Batson and the then-recent state case of Ex parte Branch, 526 So.2d 609 (Ala. 1987). Hollingsworth v. State, 549 So.2d 110 (Ala.Crim.App. 1988). After the hearing, the trial judge found no violation of Batson or Branch; the Alabama Court of Criminal Appeals affirmed. Id. at 111.

After exhausting his state remedies, Hollingsworth filed the current action under 28 U.S.C. § 2254, reasserting the Batson issue and complaining of demonstrative evidence that was allegedly inflammatory. A magistrate judge reviewed the transcript of the state court Batson hearing and concluded that Petitioner failed to carry his burden of proving purposeful discrimination; in addition, the magistrate judge found that there was no error regarding the demonstrative evidence. The district court accepted the magistrate judge's recommendation that the petition be denied. Hollingsworth raises both his Batson and evidence claims in this court. We find that his argument regarding demonstrative evidence is without merit and warrants no discussion.

An analysis of the Batson claim must begin with a brief recitation of facts surrounding Hollingsworth's crime. The murder victim was the manager of the Ensley Grill, a popular restaurant in the Ensley area of Birmingham. The murder occurred during an armed robbery one night after the restaurant had closed. Hollingsworth was one of several participants in the robbery; under the state's theory of the case, a perpetrator other than Hollingsworth was the "trigger man" who actually killed the manager. The murder received a considerable amount of attention in Birmingham. Hollingsworth was charged with capital murder and found guilty of the lesser charge of felony murder.

During the Batson hearing in the state trial court, the prosecution articulated specific, non-racial explanations for each of its peremptory strikes. The state argued that it preferred jurors who were stable members of the community; to this end, potential jurors were asked if they had owned their home for at least five years and worked at the same job for at least five years. Additionally, due to the complexity of the case, the prosecution preferred jurors who had previous jury experience. The state also asked potential jurors if they had ever worked at a business that they had to close down at night. Other factors cited as relevant by the prosecution included familiarity with the Ensley Grill and the Ensley neighborhood in general. The prosecutor explained that many photographs of the restaurant would be introduced as evidence, so he wanted jurors who would depend on the photographs and not their personal recollections of the scene. Those who lived in the vicinity may have been exposed to more extra judicial discussions of the crime and may have had some familiarity with witnesses that had not yet been discovered at the time of jury selection. The prosecution attempted to show lack of racial bias by demonstrating that all potential jurors with enough "disqualifying" factors were stricken, regardless of race. In making this showing, the state analyzed only stricken potential jurors; it did not engage in any comparison of stricken black potential jurors with seated white jurors.

In response, defense counsel proffered two arguments: that many of the prosecution's claimed rationales, such as striking those who lived in the Ensley area, were pretexts for racial discrimination; and that the stated criteria were not applied equally because some seated whites had roughly the same characteristics as stricken blacks. The trial court later issued a written order denying Hollingsworth's Batson challenge. After affirmance by the Alabama Court of Criminal appeals, the Alabama Supreme Court denied certiorari.

Upon consideration of Hollingsworth's Section 2254 petition, the magistrate judge agreed with the Alabama state courts that no purposeful discrimination had been proven. The magistrate judge found the prosecution's proffered explanations of the strikes legitimate and nondiscriminatory; this conclusion was bolstered by the fact that whites with disqualifying factors were also stricken. The magistrate judge, like the prosecution and the state trial court judge, did not conduct an explicit comparison of stricken black potential jurors and seated white jurors.

After a defendant has made his required prima facie showing of discriminatory use of peremptory strikes and the prosecution has articulated a race-neutral explanation, the ultimate burden in a Batson case remains with the defendant to show purposeful discrimination. Hernandez v. New York, 500 U.S. 352, 358, 111 S.Ct. 1859, 1866, 114 L.Ed.2d 395 (1991). The trial court's finding of no discrimination is a fact finding, ordinarily entitled to great deference on review. Batson, 476 U.S. at 98 n. 21, 106 S.Ct. at 1724 n. 21. This deference is bolstered by the fact that 28 U.S.C. § 2254(d) requires us to accord a presumption of correctness to state court factual findings. Hernandez, 500 U.S. at 364, 111 S.Ct. at 1869. Although Batson teaches that peremptory strikes cannot be exercised on the basis of race, we keep in mind that the state's justification is not required to reach the level of a challenge for cause. United States v. David, 844 F.2d 767, 769 (11th Cir. 1988). While a comparison of stricken whites with stricken blacks is relevant to a Batson claim, a comparison of stricken blacks to seated whites also is appropriate. See, e.g., United States v. Bennett, 928 F.2d 1548, 1551-52 (11th Cir. 1991); United States v. Alston, 895 F.2d 1362, 1374-75 (11th Cir. 1990) (Hatchett, J., concurring); Reynolds v. Benefield, 931 F.2d 506, 512 (8th Cir.), cert. denied, 501 U.S. 1204, 111 S.Ct. 2795, 115 L.Ed.2d 969 (1991). Because none of the courts that have heard this case have explicitly engaged in this latter type of comparison, our analysis will focus on the similarities and differences between stricken black potential jurors and seated white jurors.

The prosecution's primary areas of concern can be classified in three categories: stability, jury experience, and familiarity with the scene of the crime. Our review of the record leads to the conclusion that none of the questionable jurors had previous jury experience; therefore, we need not discuss this factor. The prosecution's primary proxies for stability were whether potential jurors had owned their own homes for five years and had the same jobs for five years. Familiarity with the scene encompassed those who had eaten at the Ensley Grill, along with those who had lived in the general area. Although these categories were those primarily relied upon, a great deal of potentially relevant information was elicited during voir dire.

White jurors Hoggle and Champion, and black potential juror Canady, had not held the same jobs for five years nor owned their homes for five years. None of them had eaten at the Ensley Grill, and they did not report living within five miles of the crime scene. However, both Hoggle and Champion had worked in business that closed at night and owned handguns, qualifications that made them more attractive to the prosecution. In addition, Hoggle's home had been burglarized; although the actual burglar was not caught, law enforcement had apprehended the person who purchased the stolen merchandise. The prosecution had rational, nondiscriminatory reasons to prefer Hoggle and Champion as jurors over Canady.

The two questionable jurors presenting the closest issue are seated white juror Bowman and stricken black juror Studmire. Bowman had not owned her own home for five years, but had held the same job for five years; she had eaten at the Ensley Grill. Studmire had both owned her own home for more than five years and held the same job for more than five years. She also had eaten at the Ensley Grill; in addition, she lived in the general vicinity and carpooled in Ensley. Bowman, however, answered affirmatively the prosecution's voir dire question asking jurors to indicate if they had owned, operated, or worked in a small business that closed after dark, indicating that the clinic at which she works closes at 10 p.m. Studmire's home had been burglarized, but the thief was never caught. This is ambiguous from the perspective of the prosecution; the burglary may create sympathy for the victim, while the failure of law enforcement in apprehending the culprit may lead to antipathy toward the prosecution. Additionally, Studmire's close connection to the area in which the crime was committed is highly relevant. Living and carpooling in the area created both an increased risk of familiarity with the scene of the crime and a heightened likelihood of being subjected to conversations relating to the crime.

After a careful review of the record, we cannot conclude that the state trial judge's ultimate finding of fact was in error. The record supports the finding that the same factors used to strike African-American potential jurors were also used to strike white jurors. Although a few white jurors (most notably juror Bowman) also had some of the characteristics on the basis of which black jurors were stricken, a careful comparison of Bowman with the strongest stricken black potential juror (Studmire) does not leave us with the impression that potential juror Studmire would have appeared to the prosecution as a more desirable juror than Bowman. To the contrary, as the above discussion indicates, the record reflects that white juror Bowman would appear to be slightly preferable as a juror from the perspective of the prosecution to black juror Studmire, regardless of race.

Our conclusion is bolstered by the fact that the ultimate burden of proof lies with the petitioner, and the fact that in the evidentiary hearing in state court, counsel for petitioner had an opportunity to adduce evidence to carry the burden of proof, and failed to do so. The actual record developed by counsel, although presenting a fairly close question, did not ultimately demonstrate that any stricken black potential juror did in fact possess the same or better (from the perspective of the prosecution) characteristics as those of the seated white jurors, leading to the conclusion that the prosecution's strikes were not intentionally racially discriminatory.

In sum, the evidence shows that the prosecution used the same criteria to strike white and black potential jurors, that the criteria were racially neutral, and that (as above analyzed) there were significant non-race-based differences between seated white jurors and stricken black potential jurors. Under all the evidence, we conclude that the state court's finding was supported by the record. For the foregoing reasons, the judgment of the district court is AFFIRMED.