Supreme Court scholars and others — including Chief Justice John Roberts Jr. — have pointed to some statistical research about oral arguments that seems too simple to be accurate. Namely, if you want a good predictor of how a case will turn out, count up the questions from justices aimed at each side. Much more often than not, the party that gets the most questions loses.

“The hypothesis is strongly supported,” said Lee Epstein, William Landes and Richard Posner in a law and economics paper recently published on SSRN and available here. It’s not just a matter of more questions being needed to probe a weaker case, they concluded, but also a function of strategies by certain justices about the best way to persuade their colleagues to join their side.