[joefromchicago]

Originally Posted by Garlon

This is interesting. I go with a 4-man strict rotation from 1920-1935 and 3 relievers and a roster of 21 players. I am noticing that even though I set limits to 3 relievers, if I allow there to be say 16 position players and 7 relievers in the strategy settings that the computer will sometimes ignore this and put 4 or 5 relievers on the roster. I think the computer will not attempt to use fewer than 14 position players though.

I wouldn't use the strict rotation setting before the 1950s. No team employed pitching rotations before then, and even then it was regarded as something of a curiosity. "Start highest rested" is, I think, a more historically accurate strategy for this era, although I plan on going back through the game records to see exactly what OOTP considers "rested." In 1922, for instance, there were a couple of instances where a Detroit pitcher started games on consecutive days. I don't think that can happen in OOTP, even if the pitcher has high stamina and pitches only one or two innings in the first game.

I would think that having four-man rotations would result in far too few starters pitching far too many innings. In 1922, only four pitchers started 38 games or more (roughly 25% of their team's games) and only three pitched more than 300 innings. I'm happier using six-man rotations and "start highest rested."

I haven't noticed the AI adding relievers beyond the bullpen limit set in the League Settings/Stats & AI page, but I wasn't really paying much attention to the AI teams. I'll have to go back and look at that. As a general rule, the AI used more pitchers than I did, but I think that is, in part, due to the AI aggressively recruiting free agents and promoting them when rosters expanded on September 1.

Originally Posted by Garlon

So you might want to try a roster of 21 players with 14 position players, 6 SP with start highest rested as the mode and 1 RP in the settings and allow starters in relief and see if you get better results like that.

I think a 23-man roster is more historically accurate for 1922, although I do think Detroit and probably some other teams used fewer players on a regular basis. I used 21-man rosters for my 1916 replay, although I used eight-man pitching staffs instead of seven.

I haven't seen much evidence on the number of players carried by teams in this era. There were roster limits in place and we know what those were, but teams routinely suited up fewer players than they were allowed (a roster minimum wasn't established until 1977). Just looking at the stats, it's pretty clear to me that the Tigers used eight pitchers for the vast majority of the 1922 season. Some teams may have used seven. I doubt any team used fewer than seven or more than eight, at least for any extended period of time. Of course, pitching staffs grew over time. In the early years of the twentieth century, six-man staffs were the norm.

Originally Posted by Garlon

You may want to try adjusting the Pitcher Stamina as well.

That's certainly one possibility. The problem that I encountered is that the results suggest that pitcher stamina is too high and too low - too high because too many pitchers are completing games, and too low because the AI is using too many relievers. There are a lot of moving parts to take into account here, and changing one might have an unintended effect somewhere else.