dissenting.
A district court may revoke a. defendant’s term of supervised release if the court, “pursuant to the Federal Rules of Criminal Procedure applicable to revocation or probation or supervised release, finds by a preponderance of the evidence that the defendant violated a condition of supervised release.” 18 U.S.C. § 3583(e)(3). Though not all rules applicable to criminal trials apply strictly to revocation hearings, the defendant is entitled to “question any adverse witness unless the court determines that the interest of justice does not require the witness to appear.” Fed.R.Crim.P. 32.1(b)(2)(C). If the government asserts that confrontation is “undesirable or impractical,” the district court must balance the defendant’s “right *426to confront a witness against the grounds asserted by the government for not requiring confrontation.” United States v. Bell, 785 F.2d 640, 642-43 (8th Cir.1986). Only if “the government demonstrates that the burden of producing live testimony would be inordinate and offers in its place hearsay evidence that is demonstrably reliable” has it made a sufficient showing of good cause to overcome the defendant’s right to confront the witnesses against him. Id. at 643. If the government fails to make such a showing, the defendant retains the right to confront the witnesses whose testimony may result in the revocation of his supervised release. United States v. Johnson, 710 F.3d 784, 789 (8th Cir.2013).
Here, the government failed to present any evidence to support a finding that producing live testimony would have been unreasonably burdensome or impractical. As the court notes, at the revocation hearing the government offered no explanation for its failure to produce the live testimony of either the Virginia probation officer or the Virginia police officers. There is no evidence of which, or how many, witnesses would have been required at the hearing,3 exactly how far they would have had to travel, the time required for their travel and testimony, the actual costs associated with their travel, or any professional obligations that would have prevented their travel. See Johnson, 710 F.3d at 789 (“Balancing the lack of an explanation for a witness’s unavailability against the reliability of the police report, we hold that the district court erred” in allowing introduction of a police report without the testimony of the arresting officer). Perhaps we can take judicial notice of the geographical distance between Bristol, Virginia (assuming all relevant witnesses were located there — a fact not in the record), and Springfield, Missouri. But, there is no evidence in the record of how that distance created an inordinate burden on the government’s ability to call live witnesses at the hearing. See Bell, 785 F.2d at 644 (declining to decide whether the “considerable expense” of travel between Wichita, Kansas, and the Eastern District of Arkansas would constitute good cause not to require live testimony of arresting officers). With no such evidence presented, I respectfully disagree that the facts were sufficiently developed at the district court such that we can properly conduct the required balancing test — weighing Harrison’s “right to confront a witness against the grounds asserted by the government for not requiring confrontation” — on appeal.
The government’s failure to demonstrate that producing live testimony would be unreasonably or inordinately burdensome is, in my view, fatal by itself to its argument that' there was good cause to overcome Harrison’s right to confrontation. However, I also am not persuaded that the hearsay evidence admitted bore sufficient indicia of reliability to justify the lack of live testimony. Probation officer Bradley Cox’s testimony was entirely based on his review of the police report and arrest warrant, as well as his conversation with Virginia federal probation officer Robert Hagy. Hagy, in turn, apparently had received his information about the incident from Virginia state probation officer Eddie Blevons. Though the police report does contain first-person narratives of the incident, we have “previously questioned the reliability of police reports as evidence of *427criminal conduct.” Johnson, 710 F.3d at 789 (citing Bell, 785 F.2d at 644). “While police reports may be demonstrably reliable evidence of the fact that an arrest was made, they are significantly less reliable evidence of whether the allegations of criminal conduct they contain are true.” Id. (quoting Bell, 785 F.2d at 644). Here, the officers who wrote the reports were also victims of the alleged assault. Whether the reliability of a police report is bolstered when the report’s author is also the alleged victim — or whether it is, instead, undermined — is not a question that has a readily apparent answer.
I also disagree that Harrison’s own statements meaningfully corroborated the police report and arrest warrant. Harrison admitted to several of the key facts described in the reports, such as “[running] out of the probation board” and continuing to run even after the officers had grabbed his shirt. His characterization of the incident was that the officers slipped and fell while pursuing him. Harrison’s statements at the revocation hearing, however, did not corroborate the whole of the police report, which included that Harrison “struggled and fought” with the officers and “continued to fight with [them], pulling away and swinging at [them].” The report also included the legal conclusion that Harrison “assaulted” the police officers as they were attempting to detain him. Thus, Harrison disagreed with both factual assertions and legal conclusions contained in the police report, and it was the whole of the police report upon which the district court relied when revoking Harrison’s supervised release and imposing sentence.
The government failed to demonstrate that producing one or more live witnesses at the revocation hearing was unreasonably burdensome or impractical, and the evidence offered in its place lacked sufficient indicia of reliability to warrant denying Harrison the right to confrontation. As a result, I would conclude that the district court abused its discretion in admitting the disputed evidence. I would therefore not reach the question of whether the district court properly conducted the required categorical analysis to determine if the alleged offense was a crime of violence under USSG § 4B1.2(a).
. Given the nature of the hearsay evidence in this case, it seems entirely plausible that one officer’s testimony could have corroborated the evidence and provided sufficient indicia of reliability to justify its admission without the live testimony of all four Virginia officers involved.