Who is trying to count sentencing mistakes?

Writing in the New York Times, Adam Liptak's "Sidebar" column today is headlined "Consensus on Counting the Innocent: We Can't."  Here are snippets:

A couple of years ago, Justice Antonin Scalia, concurring in a Supreme Court death penalty decision, took stock of the American criminal justice system and pronounced himself satisfied. The rate at which innocent people are convicted of felonies is, he said, less than three-hundredths of 1 percent — .027 percent, to be exact. That rate, he said, is acceptable. “One cannot have a system of criminal punishment without accepting the possibility that someone will be punished mistakenly,” he wrote. “That is a truism, not a revelation.”

But there is reason to question Justice Scalia’s math.  He had, citing the methodology of an Oregon prosecutor, divided an estimate of the number of exonerated prisoners, almost all of them in murder and rape cases, by the total of all felony convictions. “By this logic,” Samuel R. Gross, a law professor at the University of Michigan, wrote in a response to be published in this year’s Annual Review of Law and Social Science, “we could estimate the proportion of baseball players who’ve used steroids by dividing the number of major league players who’ve been caught by the total of all baseball players at all levels: major league, minor league, semipro, college and Little League — and maybe throwing in football and basketball players as well.”

Joshua Marquis, the Oregon prosecutor cited by Justice Scalia, granted the logic of Professor Gross’s critique but not his conclusion. “He correctly points out,” Mr. Marquis, the district attorney in Clatsop County, Ore., said of Professor Gross, “that rape and murders are only a small percentage of all crimes, but then has absolutely no real data to suggest there are epidemic false convictions in, say, burglary cases.”

What the debate demonstrates is that we know almost nothing about the number of innocent people in prison. That is because any effort to estimate it involves extrapolation from just two numbers, neither one satisfactory....  We are left with an uneasy agreement between Professor Gross and Mr. Marquis on at least one point.  “Once we move beyond murder and rape cases,” Professor Gross wrote, “we know very little about any aspect of false conviction.” 

But a few general lessons can be drawn nonetheless. Black men are more likely to be falsely convicted of rape than are white men, particularly if the victim is white.  Juveniles are more likely to confess falsely to murder.  Exonerated defendants are less likely to have serious criminal records.  People who maintain their innocence are more likely to be innocent. The longer it takes to solve a crime, the more likely the defendant is not guilty.

Regular readers will not be surprised to hear that, though I am sympathetic to concerns about the rate and nature of wrongful convictions, I am even more concerned about the rate and nature of wrongful sentencing.  Unfortunately, I do not think anyone is trying to track or even understand the problem of wrongful sentences.