Posted by lamps24 on 3/25/2020 1:51:00 PM (view original):
FWIW, my linear regression model with ERA as the outcome has a small but positive coefficient for P3, and very tiny negative coefficients for P4 and P5.
In other words, P3, P4, and P5 don’t have a huge effect on predicted ERA. It’s not a perfect model by any means, but I don’t think that they programmed 2 and 3-pitch pitchers to be “inconsistent” or significantly worse than 4-pitch pitchers, other than the fact that they miss out on 0.01 to 0.05 points in ERA.
IMO, the inconsistency observation is 1) partially true because higher ERA pitchers will actually have more variation because larger numbers vary more, and 2) confirmation bias.
So, I'll go on the other side of this then (because I love to Argue)
Danny Mullholland
Don't worry about the other seasons. Focus on SD in the NL (where baseballs go to die). He is by no means an elite pitcher. I had him as an SP3 to be honest. I figure the 3.90 ERA season is about where he belongs, but he was all over the place. 3.03, 3,33, 4,48, 4.40. The big thing here is there aren't too many seasons in the 'Middle' Other then the one 3.90 season there is nothing between a 3.50 ERA and a 4.20 ERA which is what I would expect.
Oddly enough too (I read every play by play) I always called him Danny 'One bad inning' Mullholland. Because it always seemed to happen. He would be dealing for 1-2 innings, then give up a 3 or 4 spot, then go back to pitching shutout ball.
One of the strangest players I have ever seen.