"How can you look at those numbers and say he's not more or likely to suck in clutch stats than a guy like Franceour?" Um... because I know about probability distribution? If you dig around long enough, you will find players with split stats that would seem to indicate that there's some sort of effect going on - clutch, unclutch, or any set of things you can look at with splits. But given the size of the population you're looking at, you would expect there to be some unlikely/unexplainable outcomes in the data set. One player's splits don't prove anything, not even about that player. It's not enough to know whether or not your sample is significant, but whether or not your measurement is significant. When you look at major league players as a population, "clutch" tendencies don't seem to persist from season to season, the way that, say, platoon tendencies do.