Law prof claims computer model predicts SCOTUS decisions with 70% accuracy
Updated: A South Texas College of Law assistant professor who developed a Supreme Court fantasy league says he and two colleagues have developed a computer model that can predict decisions of the court and individual justices.
Law professor Josh Blackman, writing at his blog, says his computer model, applied to cases since 1953, correctly identifies 69.7 percent of the court’s affirmances and reversals, as well as 70.9 percent of the votes of individual justices. A paper at SSRN has details.
The predictions are based on data that was available before the court’s decision. Ninety variables are used, including the party of the appointing president, the court era in which the decision is written, the court and justice’s ideological direction, and the agreement level of the court. The model compares predictions for each case to what actually happened, learning which variables work and which don’t.
The computer model categorizes cases using a Supreme Court database at Washington University created with a grant from the National Science Foundation, though Blackman and his colleagues will have to code cases for the upcoming term. Working on the project with Blackman are Michigan State University law professor Daniel Martin Katz and Michael James Bommarito II of Bommarito Consulting.
“While other models have achieved comparable accuracy rates,” Blackman writes, “they were only designed to work at a single point in time with a single set of nine justices. Our model has proven consistently accurate at predicting six decades of behavior of thirty justices appointed by 13 presidents.”
Blackman says he will be hosting a tournament where fantasy SCOTUS players compete against the algorithm. “What IBM’s Watson did on Jeopardy, our model aims to do for the Supreme Court,” he writes.
Blackman introduced the project at his blog on April 1. The April Fool’s joke, he tells the ABA Journal, was partly inspired by the actual project. Beginning the first Monday in October, Blackman says, a new website will allow users to pull down a pending Supreme Court case for a detailed prediction of what will happen.
Asked if the database would have real-world applications, Blackman said the model could be used by lawyers weighing whether to settle or litigate a case. “If you have intelligence that’s reliable about how the court will decide the case, you can make a more informed litigation decision,” Blackman says.
Prior coverage:
ABAJournal.com: “Databases Will Help Lawyers Craft Arguments, Make Lawsuit Decisions”
Updated on July 30 to include information about the Supreme Court database used in the project and to list colleauges working on the project with Blackman.
