Quote:
Originally Posted by Hoiles
Thanks for getting it back on point.
It's clear data quality is an issue. For example, in my test sim, Owen Smaulding is the dominant pitcher of his era. He has a nice Wikipedia article, but also a real-life 5.42 ERA according to his profile. Regardless, the relatively low imported HR and BB (and relatively respectable 3.68 FIP) credited him to have 80 movement/75 control ratings at his peak. In particular, the HR rate was probably unrealistically low. With the "three true outcomes" model, it might be more realistic to extrapolate the HR from the ERA than assume that all home runs were counted in his stats.
|
Where did those "in real life" stats come from? Seamheads and BBRef both have Smaulding as pitching 67 innings in his two-year career, whereas OOTP has him throwing 188. Seamheads and BBRef have him pitching in 15 games and starting 11. OOTP has him pitching 36 games and starting 32. That's a big difference. Does OOTP have better sources for the Negro Leagues than Seamheads? Seems unlikely.