There's been much attention paid lately to how appellate judges make decisions, what the influences are, and the perils of predicting Supreme Court decisions -- whether, for example, the left-leaning liberal legal academy blew the predictions about how the Court would treat Obamacare because, well, it is comprised of left-leaning liberals. The question of how judges decide cases takes on greater seriousness as the debate about the implications of "dark money" in judicial campaigns intensifies. In Michigan, a soon-to-be-published tome by former Michigan Supreme Court justice Elizabeth Weaver appears poised to add fuel to the flames.
If you're up for a whole book on the subject of judicial appellate decisionmaking (albeit of non-elected judges), you can read The Behavior of Federal Judges, A Theoretical and Empirical Study of Rational Choice by Lee Epstein (political scientist), William Landes (economist), and Richard Posner (federal judge), just published by Harvard University Press. Or, read Cass Sunstein's piece about the book, "Moneyball for Judges," in The New Republic. Among other things, the new book addresses the importance of ideology on the decision-making of a Supreme Court justice, explaining that advancing ideology "enhances the internal satisfactions of the judgeship” and also likely helps justices to “reap prestige, exert power and influence, and achieve celebrity, from attempting to align the law with [their] ideological commitments.”
If you're looking for another thought-provoking analysis of appellate decision-making, it's time to dust off an old (2004) instant classic, the Supreme Court Forecasting Project, by another interdisciplinary team -- Theodore W. Ruger (law), Pauline T. Kim (law), Andrew D. Martin (political science), and Kevin M. Quinn, (political science). You'll find their essay, The Supreme Court Forecasting Project: Legal And Political Science Approaches To Predicting Supreme Court Decisionmaking, in Columbia Law Journal, Vol. 104, No. 4, May, 2004. The study compared the results of experts' opinions to the results of a statistical model that used only six variables: (1) circuit of origin; (2) issue area of the case; (3) type of petitioner (e.g., the United States, an employer, etc.); (4) type of respondent; (5) ideological direction (liberal or conservative) of the lower court ruling; and (6) whether the petitioner argued that a law or practice is unconstitutional. Looking at the cases argued in the U.S. Supreme Court term, the model they constructed predicted 75% of the Court’s affirm/reverse results correctly. Collectively, legal experts had only a 59.1% success rate.
The study acknowledges the model's limitations:
There will always be some cases that depart from the general pattern, and the skillful advocate will be able to exploit distinctive facts, or make novel connections between legal principles in ways that reinforce or counteract these general trends, to the client’s advantage. Moreover, a statistical model such as this one, which assumes that past behavior best predicts future behavior, is a necessarily limited way of modeling the judgments of real human beings who are capable of evolving over time. What the model can do is to assess systematically what good lawyers know already in a rough sense: that some cases are better shots than others, and relatedly, that on particular cases some Justices’ votes will be easier to get than others.
The study also acknowledges an important truth -- that the reasoning of a decision is just as important, if not more important, than the outcome because the ways in which a court addresses disputes through its rationales, its analytical framework, and its language, "both gives voice to certain values and influences public understanding of these issues."