with whom BAKER, Judge, joins (dissenting):
I do not understand how a military judge’s inartful comment regarding her subjective belief concerning the thoughts others might have about her presiding over a bench trial when she had been the military judge in companion cases warrants reversal in this case. First, it has never been the law that hearing a companion case alone is a basis for disqualification under Rule for Courts-Martial (R.C.M.) 902(a). See United States v. Rivers, 49 M.J. 434, 444 (C.A.A.F.1998) (rejecting a disqualification claim where the military judge had disclosed presiding over a companion case before the accused opted for a trial by judge alone). Yet, as Judge Baker’s separate dissent points out, the majority effectively creates a per se rule against military judges sitting on companion cases. United States v. McIlwain, 66 M.J. at 318 (C.A.A.F.2008) (Baker, J., dissenting). At the same time, the majority acknowledges that this is not the law. United States v. McIlwain, 66 M.J. at 314 (C.A.A.F.2008). Second, and relatedly, the test for disqualification under R.C.M. 902(a) is objective, not subjective. United States v. Norfleet, 53 M.J. 262, 270 (C.A.A.F.2000). Objectively, there was no basis for disqualification. Third, under well-settled case law from this Court, because the military judge was transparent and disclosed that she had heard two companion cases, it was trial defense counsel’s responsibility to “present evidence regarding a possible ground for disqualification.” R.C.M. 902(d)(2); see also United States v. Kincheloe, 14 M.J. 40, 50 (C.M.A. 1982) (stating that an allegation must be supported by facts or probative evidence warranting a reasonable inference of partiality). He did not. In light of this, the military *316judge did not abuse her discretion in denying Appellant’s motion for recusal especially where, as here, the record discloses nothing that shows Appellant was prejudiced by her failure to recuse herself.
Even assuming the military judge did err, application of the factors outlined in Liljeberg v. Health Services Acquisition Corp., 486 U.S. 847, 864, 108 S.Ct. 2194, 100 L.Ed.2d 855 (1988), to the actual facts of this ease yields no basis for reversal. The majority’s contrary result is incongruous; there is not a single fact in the record disclosing any prejudice to Appellant, and the only hypothetical bias suggested by the military judge was in Appellant’s favor.
A. No R.C.M. 902(a) Disqualification
Rule for Court’s Martial 902(a) states that “a military judge shall disqualify himself or herself in any proceeding in which that military judge’s impartiality might reasonably be questioned.” R.C.M. 902(a) (emphasis added). The test is objective, not subjective. Kincheloe, 14 M.J. at 50. A military judge’s impartiality is presumed. See United States v. Sherrod 26 M.J. 30, 33 (C.M.A.1988). And the fact that a military judge discloses a possible basis for recusal does not require disqualification under the rule:
We have emphasized that “where the military judge makes full disclosure on the record and affirmatively disclaims any impact on him, where the defense has full opportunity to voir dire the military judge and to present evidence on the question, and where such record demonstrates that appellant obviously was not prejudiced by the military judge’s not recusing himself, the concerns of R.C.M. 902(a) are fully met.”
Norfleet, 53 M.J. at 270 (quoting United States v. Campos, 42 M.J. 253, 262 (C.A.A.F. 1995)).
Disqualification based solely upon appearances is exceptional, and involvement in a companion case does not warrant such an exception. Sherrod 26 M.J. at 33; Rivers, 49 M.J. at 444. It is not derived from application of a simple formulaic. See Sherrod 26 M.J. at 33 (analyzing for prejudice); cf. United States v. Burton, 52 M.J. 223, 226-27 (C.A.A.F.2000) (reviewing military judge’s questions on sentencing for personal bias); Rivers, 49 M.J. at 444 (reviewing military judge’s judicial involvement in a companion case and concluding that involvement in a companion case alone is not sufficient to require disqualification under R.C.M. 902(a)).1
In this case, the military judge stated on the record that she had been involved in two companion cases, questioned whether an impartial person might suspect her impartiality if she served as the finder of fact in Appellant’s ease, and advised Appellant that she would recuse herself and get another military judge if he chose a bench trial, which he did not do. The military judge complied with the procedure outlined in Norfleet and Campos. Trial defense counsel had every opportunity to further voir dire the military judge and present evidence, but chose not to, and the record demonstrates no prejudice to Appellant from the military judge’s denial of his recusal motion.
The majority nonetheless holds that the military judge’s statement is the basis for finding that she abused her discretion in this case because she “was biased.” Of course, actual bias is a ground for recusal under R.C.M. 902(b). But no one suggests that this is an R.C.M. 902(b) case. The majority states that “[e]very time she ruled on evidence, asked questions, responded to member questions, or determined instructions, *317the military judge exercised her discretion, a discretion that she admitted an impartial person would conclude had not been exercised in an impartial manner.” Mcllwain, 66 M.J. at 314. But it fails to identify a single fact to support the conclusion that the military judge’s decisions throughout the trial were tainted, less than impartial, or that Appellant was prejudiced in any way.
Instead, the majority appears to rest its decision solely on the military judge’s gratuitous statement about what people might have thought about her presiding in a judge-alone trial had the accused elected one. But there is simply no legal authority for the proposition that this conjecture alone supports the conclusion that R.C.M. 902(a) disqualified her from the members trial over which she in fact presided. And the problem, as Judge Baker explains in his separate dissent, with which I agree, is that applying a contrary rule may discourage judicial candor. Mcllwain, 66 M.J. at 318-19 (Baker, J., dissenting).
Since nothing in the record would lead a reasonable person to question the military judge’s impartiality, and since having heard a companion case is, as a matter of law, alone not enough to present a R.C.M. 902(a) issue, I would hold that the military judge did not abuse her discretion.
B. If Error, No Prejudice
Even if a military judge errs in failing to recuse herself, the reversal of a military judge’s decision is warranted only when the record supports the conclusion that there is: (1) a risk of injustice to the parties in the case; (2) the risk that denial of relief will produce injustice in other cases; and (3) the risk of undermining the public’s confidence in the judicial process. United States v. Butcher, 56 M.J. 87, 92 (C.A.AF.2001) (citing Liljeberg, 486 U.S. at 864, 108 S.Ct. 2194). In this case, even assuming the military judge erred, Appellant has not shown, and the majority fails to explain, how Appellant, future litigants, or public confidence were subjected to an increased risk of harm by the decision of the military judge in this case. See Liljeberg, 486 U.S. at 864,108 S.Ct. 2194. Reversal is not warranted.
There is no showing of injustice to Appellant. Traditionally, this Court has undertaken a record-intensive analysis in weighing the first Liljeberg factor in order to find an injustice. See, e.g., Butcher, 56 M.J. at 92-93 (examining the record to determine the risk of injustice based on the timing of specific rulings during the trial); Burton, 52 M.J. at 226 (examining the record in determining whether military judge’s questions reflected a personal bias against the appellant). Instead, in this case the majority summarily states that “[t]he risk of injustice to the parties is high when a military judge who has stated her bias nonetheless presides over a court-martial, even when she does not act as the trier of fact.” Mcllwain, 66 M.J. at 315. This truism is unrelated to the facts of this case.
First, the military judge did not state that she had a bias. She said only that in a bench trial, an impartial person might assume she did because she presided over two companion cases. Therefore, she advised Appellant during his forum election that, if Appellant chose a bench trial, she would recuse herself and get another military judge in order to avoid the perception that any ruling she made on witness credibility in the earlier companion eases might be perceived as favorable to her determination of Appellant’s guilt or innocence. The distance between the military judge’s statement and the reasoning of the majority on this point is marked.
Second, unlike in Sherrod, 26 M.J. at 31, Appellant does not argue that his forum election was tainted by the military judge’s statements or that the outcome of his ease would have been different had the military judge recused herself. Finally, neither Appellant nor the majority point to a single ruling, comment, or fact at trial indicating the military judge demonstrated bias or was prejudicial to Appellant. The absence of any fact on the record showing that there was an increased risk of injustice to Appellant makes it difficult to see how the first Liljeberg factor weighs in favor of reversal.
I agree with the majority that the second Liljeberg factor was not implicated in this case, but part ways on the analysis of the *318third. The majority holds that the military judge’s decision undermined public confidence because “[ejvery time the military judge made a decision, she exercised her discretion — a discretion which she herself had found was biased.” Mcllwain, 66 M.J. at 315. As noted above, this is not what the military judge said. Moreover, it conflates R.C.M. 902(a) and (b), and, even assuming error, ignores the ordinary rule that something beyond error is required for reversal in any disqualification case. Liljeberg, 486 U.S. at 864, 108 S.Ct. 2194; Butcher, 56 M.J. at 92.
Based on this record, where the only evidence in support of recusal under R.C.M. 902(a) was the fact that the military judge had heard two companion cases, there was no risk that her decision “undermin[ed] the public’s confidence in the judicial process.” Liljeberg, 486 U.S. at 864, 108 S.Ct. 2194. Instead, the record reflects full transparency on the part of the military judge, no partiality towards either party in the conduct of the trial, that the findings and sentence were entered by a panel of members, and that the only bias expressed by the military judge was possibly in favor of Appellant.
I respectfully dissent.
. Parenteau v. Jacobson, 32 Mass.App.Ct. 97, 586 N.E.2d 15 (1992), and Walbergv. Israel, 766 F.2d 1071 (7th Cir.1985), cited in the granted issue are inapposite. In Parenteau, the judge abused his discretion by subordinating concerns about bias to judicial economy. 586 N.E.2d at 18. Further, the Parenteau record was replete with facts showing a reason to question the judge’s impartiality, not least of which was the judge’s statement that he had previously assessed the appellant's credibility and determined "he was one of the biggest liars that I'd seen in a long time." Id. at 17. Walberg was an ineffective assistance of counsel case, not a disqualification case in which, similarly, the record showed the judge was so outwardly hostile towards the appellant that during the suppression hearing he answered the appellant’s questions before the appellant to demonstrate the predictability of his answers and implied in open court that the appellant was unworthy of the attorney's efforts on his behalf. 766 F.2d at 1073, 1077.