concurring in part and dissenting in part.
I concur with the majority that the appeal should not be barred by the waiver in *521Boneshirt’s plea agreement. I strongly dissent as to the sentence.
Discretion in sentencing belongs with the district court, see United States v. Likens, 464 F.3d 823, 827 (8th Cir.2006) (Bright, J., dissenting), but that discretion is not unfettered, United States v. Burns, 577 F.3d 887, 896-97 (8th Cir.2009) (en banc) (Bright, J., concurring). As this court observes, “[a] sentence is substantively unreasonable if the district court fails to consider a relevant factor that should have received significant weight, gives significant weight to an improper or irrelevant factor, or considers only the appropriate factors but commits a clear error of judgment in weighing those factors.” United States v. Lozoya, 623 F.3d 624, 626 (8th Cir.2010) (quotation omitted). Stated another way, “ ‘substantive review exists, in substantial part, to correct sentences that are based on unreasonable weighing decisions.’ ” United States v. Kane, 639 F.3d 1121, 1136 (8th Cir.2011) (quoting United States v. Irey, 612 F.3d 1160, 1193— 94 (11th Cir.2010) (en banc)).
In en banc this court has explained that it will be the “unusual” case when we reverse a sentence as substantively unreasonable. United States v. Feemster, 572 F.3d 455, 464 (8th Cir.2009) (en banc). However, “[i]n sentencing, as in other areas, district judges at times make mistakes that are substantive. At times, they will impose sentences that are unreasonable. Circuit courts exist to correct such mistakes when they occur.” Rita v. United States, 551 U.S. 338, 354, 127 S.Ct. 2456, 168 L.Ed.2d 203 (2007). While we review the district court’s sentencing decision for an abuse of discretion, “[tjhere is a difference ... between recognizing that another usually has the right of way and abandoning one’s post.” Kane, 639 F.3d at 1136 (quotation omitted).
Boneshirt’s forty-eight-year sentence is substantively unreasonable because the district court unreasonably weighed the facts at issue in the case. The district court failed to give proper weight to the fact that Boneshirt was a juvenile when he committed the crime, especially when his age is considered with his background and upbringing. And further, the district court’s sentence placed too much weight on a plan to escape by Boneshirt when he was pending sentencing.
First, federal and state courts treat juvenile offenders, even serious offenders, differently than adults who commit serious crimes. Juveniles “have a ‘lack of maturity and an underdeveloped sense of responsibility’; they ‘are more vulnerable or susceptible to negative influences and outside pressures, including peer pressure’; and their characters are ‘not as well formed.’ ” Graham v. Florida, 560 U.S. -, 130 S.Ct. 2011, 2026, 176 L.Ed.2d 825 (2010) (quoting Roper v. Simmons, 543 U.S. 551, 569-70, 125 S.Ct. 1183, 161 L.Ed.2d 1 (2005)). The Supreme Court noted that “developments in psychology and brain science continue to show fundamental differences between juvenile and adult minds,” id., and “ ‘[i]t is difficult even for expert psychologists to differentiate between the juvenile offender whose crime reflects unfortunate yet transient immaturity, and the rare juvenile offender whose crime reflects irreparable corruption.’ ” Id. (quoting Roper, 543 U.S. at 573, 125 S.Ct. 1183). “[T]he character of a juvenile is ‘more transitory’ and ‘less fixed’ than that of an adult.” Id. at 2038 (Roberts, C.J., concurring) (quoting Roper, 543 U.S. at 569-70, 125 S.Ct. 1183). We still hold juvenile offenders responsible for their actions, but because of these fundamental differences a juvenile’s offense “ ‘is not as morally reprehensible as that of an adult.’ ” Id. at 2026 (quoting Thompson v. Oklahoma, 487 U.S. 815, 835, 108 S.Ct. *5222687, 101 L.Ed.2d 702 (1988) (plurality opinion)).
The heavy criminal sentence here does not recognize the possibility of rehabilitation for juveniles. Categorically, juveniles may not be sentenced to death, Roper, 543 U.S. at 578, 125 S.Ct. 1183, and they may not be sentenced to life in prison without the possibility of parole for a nonhomicide crime, Graham, 130 S.Ct. at 2034. Part of the rationale in those cases is that “[jjuveniles are more capable of change than are adults” and “ ‘[f]rom a moral standpoint it would be misguided to equate the failings of a minor with those of an adult, for a greater possibility exists that a minor’s character deficiencies will be reformed.’ ” Graham, 130 S.Ct. at 2026-27 (quoting Roper, 543 U.S. at 570, 125 S.Ct. 1183). “Maturity can lead to that considered reflection which is the foundation for remorse, renewal, and rehabilitation.” Id. at 2032. A sentence for a juvenile needs to take into consideration the transitory nature of youth and the possibility of rehabilitation.
Here the district court ignored these principles in sentencing Boneshirt, who was seventeen years old at the time of the crime, to forty-eight-years’ imprisonment.4 The district court calculated Boneshirt’s total offense level at 42 with a criminal history category of I, thus indicating an advisory guideline sentence of 360 months (30 years) to life imprisonment. The sentencing judge justified, in part, his forty-eight-year sentence based on the crime being “horrible” and his belief that Boneshirt needed a lengthy incarceration to protect the public. Every murder is a horrible crime. Yet despite the already severe sentence included in the low end of the guideline range of 360 months (30 years), the sentencing judge proceeded to disregard statistics indicating much less harsh adult sentences for murder and hand down a sentence more than double the mean sentence for murder in the Eighth Circuit over the last five years.5 While sentencing statistics are not determinative, they provide context for the gravity of the sentence imposed. And in imposing such a harsh sentence, the only consideration the district court gave to Boneshirt’s age was to note: “You are very young, 17 years old, now 18. It is a factor that the Court considers.”6
*523The district court put the proper sentencing for juveniles on its head in this case by not giving proper weight to Boneshirt’s age and treating him much harsher than a similarly situated adult. Notwithstanding the broad sentencing discretion vested in federal judges, see Gall v. United States, 552 U.S. 38, 50-51, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007), this sentence is wrong on its analysis of the facts and its application of sentencing requirements.
Second, the district court did not give proper weight to Boneshirt’s background and upbringing. Like many children born and raised in Indian Country, Boneshirt suffered abuse beginning in infancy. See, e.g., United States v. Deegan, 605 F.3d 625, 639-42 (8th Cir.2010) (Bright, J., dissenting). Boneshirt was born to an alcoholic mother and she and his step-father heaped physical and emotional abuse on this young boy. From ages three to seven, he was shifted out of his mother’s home on numerous occasions and spent time with nine temporary foster families. His mental health situation was bad; he was born with alcohol-related neurodevelopmental disorder, and was diagnosed with a mood disorder and attention deficit hyperactivity disorder.
In an attachment to the PSR, Dr. Stephen P. Manlove gave an insight into Boneshirt’s mental health, his alcoholism, and the relationship of these conditions to the crime. “It appears that the problems related to his Alcohol Related Neurodevelopmental Disorder, including mood swings and the impulsivity he had due to Attention Deficit Hyperactivity Disorder, were catalyzed by his intoxication resulting in the events leading to the death of Marquita Walking Eagle.” Dr. Manlove also noted that “[ajppropriate treatment for chemical dependency and psychiatric problems would certainly diminish the likelihood of future dangerous acts” and “Boneshirt’s good behavior at the ... Juvenile Services Center suggests that as he matures he may be more able to access and use the treatment process.”
As a longtime federal judge, I have seen and reviewed scores of cases in which Indian males have been convicted of assaultive crimes, including murder. Alcohol has played a substantial role in the crime in almost every one of those cases. See, e.g., United States v. Jensen, 423 F.3d 851, 853 (8th Cir.2005), United States v. LeClair, 338 F.3d 882, 884-85 (8th Cir.2003), United States v. Emeron Taken Alive, 262 F.3d 711, 712 (8th Cir.2001). And alcohol obviously played a major role in the unfortunate homicide of Ms. Walking Eagle.7 Yet the sentencing judge chose a sentence well above the average sentence for murder when, in context, the circumstances were not outside of the usual, serious crime by an adult male Indian, particularly on a reservation.
In addition, the district court did not take into consideration Boneshirt’s life expectancy. The sentencing judge specifically expressed the desire to protect the public by ensuring that Boneshirt would be an old man when he was released: “the Court believes that there is just too much of a risk with Mr. Boneshirt being a part of society before the point where he’s of a very mature age.” However, the district *524court did not properly consider Boneshirt’s background as a Native American male, who has a life expectancy of fifty-eight years. See Christopher J.L. Murray et al., Eight Americas: Investigating Mortality Disparities across Races, Counties, and Race-Counties in the United States, 3 PLoS Med. 1513, 1514 (2006) (“Native American males in the cluster of Bennet, Jackson, Mellette, Shannon, Todd, and Washabaugh Counties in South Dakota had a life expectancy of 58 [years] in 1997-2001----”).8 Even if he earns all of his good time credit, which the district court was not optimistic about, he will still serve more than forty years in prison. The district court anticipated Boneshirt would be an old man when he was released, but in reality he may be a dead man.
Finally, a significant event that increased the guideline range was an alleged plan for an escape from the Hughes County Jail. Without that finding, Boneshirt’s guideline for sentencing would have been much less. Boneshirt would have received credit for acceptance of responsibility and no added points for obstruction of justice (the escape issue), resulting in an advisory guideline range of 210 months (17 years, 6 months) to 262 months (21 years, 10 months). His final sentence was more than twenty-six years over the maximum guidelines range without the escape attempt.
If Boneshirt had attempted to escape after sentencing, he would have been charged for attempted escape under 18 U.S.C. § 751(a), or instigating or assisting an escape under 18 U.S.C. § 752(b), and faced a statutory maximum sentence of five years. In this case, planning an escape attempt before sentencing added over twelve years to Boneshirt’s minimum guideline range. Such an increase in Boneshirt’s sentence is substantively unreasonable for one poorly-timed, impetuous act of a teenager that, if attempted after sentencing, would carry a statutory maximum of five years. For a similar adult offender for second-degree murder in category I, even with obstruction of justice and no acceptance of responsibility, a thirty-year sentence should be an ample punishment. But adding eighteen years, in part for an escape conspiracy, which was unsuccessful and did no one any harm, seems clearly excessive.
In sum, in failing to give fair and adequate consideration to a juvenile’s lack of judgment and the likelihood of change for the better of a youthful offender, especially in light of Boneshirt’s upbringing, and in placing too much emphasis on a plan to escape, the district court was substantively unreasonable in weighing the facts and sentenced this defendant far more harshly than a comparable adult criminal. This excessive sentence should not stand regardless of the views of the majority. Thus, I dissent, and I would remand for reconsideration and possible reduction of the sentence.
. By comparison, Graham was thirty-four days short of his eighteenth birthday when he committed the home-invasion robbery that lead to his sentence. Graham, 130 S.Ct. at 2019. As the Supreme Court noted, "[t]he age of 18 is the point where society draws the line for many purposes between childhood and adulthood,” Id. at 2030 (quotation omitted).
. The mean sentence for murder in the Eighth Circuit over the past five years is just under 250 months (20 years, 10 months), as calculated from the statistics tracked by the United States Sentencing Commission.
Fiscal Mean Number
Year (in months) of Cases
2010 285.3 10
2009 264.7 6
2008 150.2 5
2007 247.2 9
2006 276 3
U.S. Sentencing Comm’n (USSC), Statistical Information Packet, Fiscal Year 2010, Eighth Circuit 10 (2010); USSC, Statistical Information Packet, Fiscal Year 2009, Eighth Circuit 10 (2009); USSC, Statistical Information Packet, Fiscal Year 2008, Eighth Circuit 10 (2008); USSC, Statistical Information Packet, Fiscal Year 2007, Eighth Circuit 10 (2007); USSC, Statistical Information Packet, Fiscal Year 2006, Eighth Circuit 10 (2006). I also note that the 2010 statistics were not available at the time of sentencing and, therefore, the mean would have been even lower. To reflect life expectancy, life sentences and sentences over 470 months are capped at 470 months. USSC, 2010 Sourcebook of Federal Sentencing Statistics 161 (2010). Murder includes first and second degree murder, felony with death resulting, and conspiracy to murder with death resulting. Id. at 165.
. The district court was also "careful when considering juvenile history and considering history from tribal court proceedings because of due process issues,” but only in the context of discussing Boneshirt’s previous criminal conduct as a juvenile.
. It’s quite a paradox. In the record in this case, alcohol flowed everywhere at the places Boneshirt and Ms. Walking Eagle visited and enjoyed. But they could not buy cigarettes at the convenience store on the reservation.
. The offense occurred in Todd County, South Dakota.