Add new comment

Just saying that it needs to be accounted for to grasp the amount of change a pitch caller affects on the pitcher.
Do you break down pitch types and counts etc.like i was explaining as well? I don't think it would be really accurate without doing it but I don't see a way to accurately do it. Maybe using expectations of situations in some way would be better than figuring based on the exact outcomes. I don't know. Its mostly in the difference of runners on base when an extra base hit occurs that it seems like a big crap shoot, but then in most of those cases both the pitcher and catcher are responsible for the baserunners. It's just that the outcome of a called pitch don't necessarily reflect the likely outcome and it seems much larger samples would be required to come up with an expected outcome. You can throw the right pitch for the situation, including count and hitter, and just get beat on any given day. Every start includes usually a handful of mistake pitches the catcher is not at all responsible for, but they often get away with it anyway. A catcher shouldn't be responsible for a curve just not breaking, should they? Many a grand slam has been hit from breaking pitches not breaking.
Pitch framing seems much easier to figure, but even with that the umpires ultimately control a variable thatseems incalculable.
CERA doesn't have to be a perfect stat, but I find it very difficult to get rid of enough noise in the base model to make it worth even considering. CERA is something I ignore because of the problems I see with samples, but a stat with the same idea could be very valuable down the road. It's going to take a lot of work though.