The district court initially intended to give DeMarrias two years of supervised release but, after reading the report of a psychological evaluation, instead sentenced him to supervised release for life. Reading the same report, I cannot construe it to justify a lifetime of punishment, and therefore respectfully dissent.
The law is clear: supervised release is a form of punishment. Samson v. California, 547 U.S. 843, 850, 126 S.Ct. 2193, 165 L.Ed.2d 250 (2006) ("Federal supervised release ... is meted out in addition to, not in lieu of, incarceration." (cleaned up) ); United States v. Jackson, 866 F.3d 982, 985 (8th Cir. 2017) ("Supervised release is a 'form of criminal sanction imposed by a court upon an offender after verdict, finding, or plea of guilty.' " (quoting Griffin v. Wisconsin, 483 U.S. 868, 874, 107 S.Ct. 3164, 97 L.Ed.2d 709 (1987) ) ). Thus, under the federal sentencing regime, the length of supervised release must be "sufficient, but not greater than necessary" to reflect the seriousness of the offense, deter further criminal conduct, protect the public, and provide the defendant with needed training and treatment. 18 U.S.C. § 3553(a).
The facts in the report are also clear. The forensic psychologist wrote that DeMarrias "was previous[ly] convicted of Sexual Abuse of a Minor. In addition, the PSR also contains information regarding another possible sexual abuse of a minor arrest. Therefore , the diagnosis of Unspecified Paraphillic Disorder appears to be an appropriate classification for this individual." (Emphasis added.) The report goes on to acknowledge that an "unspecified" diagnosis was selected "to communicate a degree *576of diagnostic uncertainty" because of "the limited collateral information available." These passages make clear that DeMarrias's diagnosis was based entirely on information that the district court already had when it said it intended to sentence DeMarrias to two years of supervised release. No other information in the report supports the diagnosis. The report simply affixed an "unspecified" diagnostic label to facts the district court already knew.
Given this dearth of new information, and the fact that none of DeMarrias's prior revocations were based on sexual misconduct, the key question is whether the report's evaluation of DeMarrias's prognosis-that his "difficulties are of a characterological and pervasive nature, and will likely prove, [and] have proven, resistant to change"-is sufficient to justify a lifetime of punishment. DeMarrias's prognosis, which reflects the evaluating psychologist's professional expertise and experience, may very well warrant an extended term of supervised release. But a lifelong sentence is different. DeMarrias was 28 years old at sentencing, so a lifetime of supervised release means he will likely spend many decades under the supervision of the United States Probation Office. And while all supervised release sentences are, broadly speaking, related to deterrence, protection of the public, and provision of treatment, the duration of this sentence is not reasonable given that it is based on such limited evidence. In my view, it exceeds the permissible bounds of discretion. See United States v. Feemster, 572 F.3d 455, 461-62 (8th Cir. 2009) (en banc).
For these reasons, I would vacate the term of supervised release and remand for resentencing.