Results and Strategy Settings
 

Here are the strategy settings that I am currently using. I am posting some results as well from an 1871-2018 game. IPouts/GS is the IPouts per Game Started for those seasons and IPouts/RA is the Reliever IPouts per Relief Appearance. Estimated Reliever Innings and Estimated Relief Appearances are for reference as well and are per team per season.

Here are some pitching results.

Thanks for this information!

Garlon, really appreciate all the work you do.
I do have one question about the OOTP 19 guide you have on steam. Maybe it was a misprint but I think you said you DO NOT have players miss season as in real life checked. I do not understand why you would have players play seasons that the missed in real life. Maybe I read it wrong. Seems like that would mess up the stats.
Again, thank you.

You have way too many relievers, up until around 1939. And too many pitchers in general until around 1920. A more historical breakdown would be:

1871-1880 2 SP, 0 RP
1881-1887 3 SP, 0 RP
1888-1895 4 SP, 0 RP
1896-1904 5 SP, 0 RP
1905-1914 6 SP, 0 RP
1915-1923 6 SP, 1 RP
1924-1928 6 SP, 2 RP
1929-1933 6 SP, 1 RP
1934-1938 6 SP, 2 RP
1939-1945 5 SP, 3 RP
1946-???? 5 SP, 4 RP

Those numbers are all approximate - I'd have to do a lot more research to get more accurate figures. In general, the 1920s saw staffs average around 7, but some clubs still used 5 pitchers on a regular basis. It appears that clubs trimmed rosters during the Depression. By 1939 most teams relied on 5 pitchers for the majority of their starts, although other pitchers on the staff could still be expected to pick up starts here and there (in OOTP terms, they'd be relievers who had "emergency starter" as a secondary role). I didn't look much beyond 1946, as pitching staffs started looking more like "modern" staffs by the 1950s.

These numbers assume that "allow starters in relief" would be checked for the entire period up to 1945. There should really be no relievers at all until around 1915, and even then most relievers were just failed starters until approximately 1930, with a few notable exceptions. The starting rotation mode should be "start highest rested" until about 1939, at which point it should go to "strict rotation, occasionally start highest rested."

You could probably go with 2 RP's from 1871-1910. I may update this. I do not suggest going less than that though because of game with extra innings, nor do you want a singe RP having to be used in every game in OOTP as they will perform very poorly.

As for having rotations beyond what I have recommended you will find that you simply do not have enough SP's in the league to make that work and end up with pitchers with 10 stamina in your rotation. Those guys will throw 4-5 IP per start and perform very poorly.

There is something wrong. :ohmy: Cy Young does not over 500 wins. :laugh:

OOTP handles RP and SP stamina differently, so you're right that, if you use one RP all the time, he'll be tired all the time. But if you allow SPs to relieve and have "start highest rested" as your rotation mode, you should have enough rested starters on the bench to fulfill your ordinary relief needs. Try that instead of using "strict rotation" as your rotation mode. You will probably find that you don't need relievers.

Also, if you're using historical schedules, your starters should have enough off-time to rest up before their next start. For example, if you have two starters on your team in 1877 and you're playing 60 games in 162 days, there's no need to carry any relievers - your starters will almost always be rested. On the other hand, if you're playing 162 games in 1877, you'll need a much bigger staff. My staff recommendations are based on using historical schedules.

I've never found that to be true. Just about every pitcher prior to around 1920 should have the qualities required to start. You just need to adjust the league settings to expand rotations league-wide in order to change relievers into starters. To illustrate: OOTP's default settings for 1916 are 4-man rotations and high stamina. So Rube Foster* of the Boston Red Sox is a reliever in the game (as he would be under your settings as well). But Foster started 19 games for the Sox that year, and his ratings in OOTP are clearly good enough to let him start:

Attachment 637477

He has four pitches - not spectacular, but good enough to be a back-end starter, and he has very high stamina. His current role is even listed as "borderline starter." He is, in other words, perfectly capable of starting. And when we change the league settings to six-man rotations, Foster turns into a starter and the game slots him in the fifth spot in Boston's rotation.

Attachment 637478

So if you keep the OOTP settings (or use your setting recommendations), you're correct - it will look like there aren't enough starters to give each team a six-man rotation. But if you change the game's rotation setting, the game will automatically convert two relievers on each club into starters. Voila! Problem solved!



*Not the Rube Foster - Negro League legend and Hall-of-Famer - but the other Rube Foster.

How many pitchers have a green stamina rating (60 if you are using the 1-100 scale for ratings) in your league? These are the pitchers who actually have the necessary stamina required to start games. Any pitchers with less than that can still be used, but they will struggle to average even 5 IP/GS.

I'll add that your roster sizes are too big, especially in the nineteenth century. In part, that's unavoidable. OOTP's minimum roster size is 15. Even that number is too big for the 1870s, when the optimal roster consisted of:

7 position players
1 or 2 catchers
1 primary starter
1 "change" pitcher/utility player

For example, here's a photo of the 1882 Providence Grays.

Attachment 637511

That's not a representative sampling of the team's best players or just all the guys who bothered to show up to the photographer's studio that day - that's the entire team, including the manager.*

So that's ten or eleven men on the roster, which is simply impossible in OOTP.

The "change" pitcher would be expected to play a regular position in the field but also be available to relieve the starter and to start on occasion. That's because, according to the substitution rules in place in the 1870s and 1880s, a player pretty much had to be on the field in order to switch positions with the pitcher - the modern substitution rule wasn't put into place until 1891. That also means that, if you're playing nineteenth-century games, you should definitely make sure that two-way players are allowed in the league.

So roster sizes should be trimmed. I'd suggest:

1871-1880 2 SP, 0 RP, 13 position players
1881-1887 3 SP, 0 RP, 13 position players
1888-1895 4 SP, 0 RP, 13 position players
1896-1900 5 SP, 0 RP, 13 position players
1901-1904 5 SP, 0 RP, 14 position players
1905-1914 6 SP, 0 RP, 14 position players
1915-1919 6 SP, 1 RP, 14 position players
1920-1923 6 SP, 1 RP, 15 position players
1924-1928 6 SP, 2 RP, 15 position players
1929-1933 6 SP, 1 RP, 14 position players
1934-1938 6 SP, 2 RP, 15 position players
1939-1945 5 SP, 3 RP, 15 position players
1946-???? 5 SP, 4 RP, 15 position players




*And including a rare shot of Hoss Radbourne (back row, third from left) not giving the finger to the photographer

Again, that has not been my experience. Using either the game's or your pitcher stamina ratings, along with the game's relief and hook settings, you shouldn't have much trouble with starting pitchers completing games, and that includes pitchers at the back end of the rotation. In addition, you should find lots of relief pitchers with blue or green stamina ratings.

For instance, in my 1922 simulation where I used six-man rotations and two-man bullpens, Ole Olsen started 26 games and completed six, even though he has a stamina rating of 25 on the 20-80 scale. He pitched 179 innings in his starts, averaging 6.9 IP/start. The pitcher settings I used were:

Hook for starting pitcher: -1
Hook for relief pitcher: +2 (slow)
Use of relievers: very rarely*
Use of closers: very rarely
Pitcher stamina: high
Pinch hit for pitchers: very rarely

I think all of those are the OOTP defaults. I'll note that those differ from your recommended settings in that you have relievers set at "normal" and closers set at "very often." I'm not quite clear on why you have closers set that high when the closer "position" wasn't really invented until around the 1980s. Given those settings, though, I'm not surprised that many of your back-end starters aren't making it out of the fifth inning.


*I think that rating may be too low. If I ran this test again, I would probably set relievers at "rarely" instead of "very rarely."

What I have noticed is that the game works best in terms of pitcher performance and being able to win games with Relievers set to Normal for all seasons. The LTM will control IP/GS for SP's and if you look at the results at the end of any season you will see that SP's will always be throwing more IP/GS than they should be. If you set Relievers to less than Normal the computer will simply be throwing away wins because they let that SP stay in even longer only to lose the game, and if you go above normal then apparently the computer will make additional pitching moves that are unnecessary and not use the best RP's to finish games. I suggest looking at the Winning Percentages of the best pitchers when you change these settings.

As for Closers being set to Very Often, what this does is simply use the best RP as much as possible. So your relief ace will pitch more than the other relievers. As a note, if you set Relievers to Very Often and Closers to Very Often, the Closers will actually pitch less than if you set Relievers to the Normal setting.

Even though I may not always agree with you, I enjoy reading your posts, Garlon, because you've clearly given this topic a good deal of thought and your analyses are always thought-provoking. I don't think I've ever really considered the interplay between the reliever/closer settings and the way that the AI makes pitching moves, but I think you're right. I'm not quite prepared to change my thoughts on the use of relievers in the pre-reliever era (let's not get carried away here!) but you've definitely given me a lot to think about.

I'll add that, in my 1922 sims, I found that the top starters tended to throw too few innings in comparison to their real-life stats. In part, I think that's because they tended not to have as many relief appearances in the game as opposed to real life. The back-end starters, though, pitched too many innings because they ended up with more starts than in real life. I think much of that has to do with the way the game calculates fatigue.

Well, historically, the best reliever available was usually a starter. I'm not sure if the AI recognizes that fact when it has a guy in the bullpen who does nothing but relieve. I'd have to go through the box scores in my 1922 sims to see who the AI used in critical save situations, but in my estimation it usually turned to the mediocre reliever in the bullpen rather than the superior and rested starter on the bench. Except in some unusual circumstances, though, that's not how managers used their staffs in the pre-reliever era.