But they grant cert in only a truly tiny percentage of all cases they get petitioned to hear, and the thing we're interested in looking at is the inter-circuit

But they grant cert in only a truly tiny percentage of all cases they get petitioned to hear, and the thing we're interested in looking at is the inter-circuit reversal differential. I wish I could find more fine-grained statistics than what I know off the top of my head is available (StatPacks from ScotusBlog, but there seems to be no aggregation of them, just individual ones for like 25 years; Empirical Scotus, but I see no data available to play with, just posts). I also wish I were better at statistics.