After a bench trial, the district court1 convicted Tracy Vaughn of conspiracy to distribute fifty grams or more of crack cocaine in violation of 21 U.S.C. §§ 841(a)(1) and (b)(1). Vaughn appeals, arguing there was insufficient evidence to support his conviction and challenging the drug quantity finding underlying his 360-month sentence. We affirm.

At trial, five admitted conspirators testified for the government that they had bought crack cocaine from or sold crack cocaine to Vaughn on numerous occasions during the alleged conspiracy period. The district court's detailed findings of fact reviewed this trial testimony, noted that the five witnesses had an incentive to cooperate with the government but not to testify falsely, further noted that some of their testimony was corroborated by others, and credited the testimony that they engaged in on-going crack cocaine transactions with Vaughn.2

On appeal, Vaughn argues that the district court erred in finding these witnesses credible because they had been convicted of drug offenses and agreed to testify against Vaughn in exchange for the possibility of receiving reduced sentences. In reviewing the sufficiency of the evidence after a bench trial, we apply the same standard that we apply when reviewing a jury verdict. See United States v. Matra, 841 F.2d 837, 840 (8th Cir.1988); United States v. Barletta, 565 F.2d 985, 991 (8th Cir.1977). It is well-established that "the uncorroborated testimony of an accomplice is sufficient to sustain a conviction if the testimony is not otherwise incredible or unsubstantial on its face." United States v. Dunn, 494 F.2d 1280, 1281-82 (8th Cir.), cert. denied, 419 U.S. 855, 95 S. Ct. 99, 42 L. Ed. 2d 87 (1974). Here, the district court credited the conspirators' testimony; these credibility findings are "virtually unreviewable on appeal." United States v. Candie, 974 F.2d 61, 64 (8th Cir.1992). After careful review of the trial record, we conclude that the evidence was more than sufficient to sustain Vaughn's conviction.

The district court's post-trial findings of fact included detailed findings as to the quantities of crack cocaine to which the five conspirator witnesses testified at trial. Based in large part on those findings, Vaughn's Presentence Investigation Report (PSR) recommended that he be held accountable for 2.17 kilograms of crack cocaine. Vaughn objected to this recommendation, arguing "that there was no direct evidence of the amount of drugs attributable to [Vaughn] and that amount of drugs attributed to [him] by the government's witnesses was vague and incorrect." At sentencing, the district court overruled this objection: "Having presided over the trial of this matter as the finder of fact and law, I agree that the amount of crack cocaine attributed to [Vaughn] in the PSR is fair and accurate."

On appeal, noting that the district court did not recite that its post-trial drug quantity findings were made beyond a reasonable doubt, Vaughn argues for the first time that the court violated his Sixth Amendment rights under the Supreme Court's post-sentencing decision in Blakely v. Washington, 542 U.S. 296, 124 S. Ct. 2531, 159 L. Ed. 2d 403 (2004). In considering this contention, we apply the Supreme Court's more recent clarifying opinion in United States v. Booker, ___ U.S. ___, 125 S. Ct. 738, 160 L. Ed. 2d 621 (2005). Because Vaughn failed to raise the issue at sentencing, we review for plain error. United States v. Pirani, 406 F.3d 543 (8th Cir.2005) (en banc).

We will assume that the district court made its pre-Blakely drug quantity findings under the preponderance of the evidence standard. But the remedial opinion in Booker held that such judicial fact-finding for sentencing purposes does not violate the Sixth Amendment when made as part of an advisory Guidelines regime.3 Thus, the "Booker error" in this case was the district court's understandable error in treating the Guidelines regime as mandatory. The plain error question, then, is whether Vaughn has satisfied his burden of demonstrating a "reasonable probability" that the district court would have imposed a more favorable sentence had the court understood the Guidelines to be advisory, as now mandated by Booker. We conclude Vaughn has made no such showing. In imposing the 360-month sentence, the court explained, "I've chosen a sentence at the low end of the guidelines range because I think ... a sentence of that substantial length is going to serve all of the goals of the criminal justice system." See Pirani, 406 F.3d at 553.

Vaughn argues in the alternative that the evidence at trial did not support the district court's drug quantity finding. He relies on the fact that the district court credited witness Chinn's testimony that Vaughn supplied Chinn with 13 ounces of crack cocaine, whereas the government advised the court some months after the trial, while this appeal was pending, that the government had not called Chinn as a witness in a subsequent trial because of concerns about his truthfulness in that unrelated matter. If Chinn's trial testimony were material to the district court's drug quantity finding, we would need to consider whether this new information casts sufficient doubt on Chinn's credibility to warrant a remand so that the district court could consider the question in the first instance. But we conclude that Chinn's testimony was not material to the drug quantity issue. The district court attributed 2.17 kilograms of crack to Vaughn for sentencing purposes. This resulted in a base offense level of 38. See U.S.S.G. § 2D1.1(c)(1). Base offense level 38 applies if the defendant was responsible for 1.5 kilograms or more of crack cocaine. Chinn's testimony attributed 13 additional ounces, or 368.55 grams (one ounce equals 28.35 grams). Subtracting Chinn's 368.55 grams from the total of 2.17 kilograms still leaves Vaughn responsible for well over 1.5 kilograms, so Chinn's testimony did not affect Vaughn's base offense level of 38.

The judgment of the district court is affirmed.