This post considers two categories of Supreme Court reversals of rulings from the court of appeals: reversals of rulings that had favored defendants, and reversals of rulings that had favored the state. The discussion covers 15 terms (1998/99 through 2012/13), which have provided a total of 316 criminal cases for consideration.
The 316 cases include 102 in which the court of appeals (or trial courts) had ruled in favor of defendants. In most instances, the phrase “ruled in favor of” meant something well short of dismissal of charges. Defendants commonly requested such things as a new trial (claiming that evidence was obtained improperly, or that they had not received competent legal representation, or that instructions to the jury were flawed, and so forth). After accepting these 102 cases, the Supreme Court reversed in 77% of them (79/102).
In the remaining 214 cases, the Supreme Court reviewed rulings in which the court of appeals or trial courts had sided with the state by approving the behavior of police and other authorities, rejecting the defendants’ arguments, and upholding their convictions. Here, the Supreme Court reversed in 24% of the cases (52/214).
Change Over Time
The Court assumed its current composition in the 2008-09 term, when Justice Gableman replaced Justice Butler. During the five terms from 2008-09 through 2012-13, the justices voted as follows regarding the issues outlined above. They reversed 80% (20/25) of rulings that had favored defendants and 12% (8/68) of rulings that had favored the state.
See Table 1 for year-by-year figures covering all 15 terms.
One may also tabulate the votes cast by individual justices in these cases, and the results highlight some striking differences among the members of the Court.
For instance, a number of justices displayed comparatively small gaps between their reversal percentages in the two categories described above. Thus Justice Bablitch voted to reverse in 49% of cases where rulings had favored defendants and in 41% of cases where rulings had favored the state. For Justice Bradley the figures were 45% and 49%; for Justice Abrahamson, 40% and 54%, and for Justice Butler, 53% and 49%. In contrast, the Court’s three most recent members have been much more inclined to reverse rulings favoring defendants (87%, 90%, and 92% for Justices Roggensack, Ziegler, and Gableman respectively), while very reluctant to reverse decisions favoring the state (13%, 11%, and 9% for the three justices).
Meanwhile, Justice Sykes generated the most unusual combination of percentages, as she voted to reverse 86% of decisions that had favored defendants—but also voted 34% of the time to reverse decisions that had sided with the state.
Table 2 provides figures for all 12 justices who served on the Court during the period under consideration.
 In most instances these rulings were decisions issued by the court of appeals, though 18 percent of the cases involved trial-court rulings that court-of-appeals judges certified directly to the Supreme Court.
 The total of 316 cases is obtained by subtracting 16 cases from the set of 332 cases described in the first footnote of the previous post (“Race of Defendants in OWI, Drug, and Sex Cases”). The 16 excluded cases consist mainly of those in which both the state and the defendant appealed different portions of a ruling, along with a handful of cases in which it was difficult to say whether the Supreme Court’s decision favored the state or the defendant.
 This happened in one of three ways. (1) The trial court granted the defendant’s motion(s), and the court of appeals affirmed. (2) The trial court granted the defendant’s motion(s), and the court of appeals passed the case up to the Supreme Court without issuing its own ruling. (3) The trial court denied the defendant’s motion(s) and was then reversed by the court of appeals.
 This also happened in one of three ways—following courses analogous to those that yielded rulings favoring defendants.
 Justice Steinmetz’s last year on the Court coincides with the first year of this study, and thus the figures on his line in Table 2 amount to a tiny (and quite possibly unrepresentative) sample of his voting throughout a tenure of two decades on the Court.