League Totals & Modifiers - Need Some Help

[Questdog]

PS. There are a lot more at bats in 2019 than in 1919, so if you just put in the 1919 HR totals, you should end up with a lot fewer home runs per team than in 1919.

[RchW]

Originally Posted by Syd Thrift

Honestly, I do nothing but fictional leagues set in "real" historical eras (for instance, right now I'm doing a league I started in 1946 which is now in 1948) and IME the way the game handles LTMs is already very, very accurate. The fictional league totals are generally very, very close to what actually happened, and that goes on down to stuff like complete games and GIDPs. IME actual run scoring is low for older leagues (not so much the 40s but in leagues up into the 20s I see this) and I think that's primarily a function of a. steals being a bit too costly (I think that from the data we have, the 55%-60% success rate is pretty close to accurate but it's my *guess* that there were more errors and extra base advances on them than IRL) and b. there not being enough advances on base hits compared to real life. But for primary stats, the numbers are often eerily accurate.

This is true for me. Frankly where I have totals off it’s generally where I made a mistake transcribing an LTM from my excel file or I changed one of the general strategic tendencies (say bunting) without changing the LTM accordingly.

[Syd Thrift]

Originally Posted by joefromchicago

I have no doubt, but then that's how OOTP is designed to work. I'm less concerned about making the league totals match the historical results and more interested in seeing the players perform their roles in a historically accurate fashion. If you've read my threads about pitcher usage in the pre-reliever era, you can get a good sense of what I'm focusing on right now.

This is going to be a thing that mostly goes beyond LTMs then, and I just don't think the AI is going to do what I think you're asking it to do, i.e. use pitchers in the 20s the way they'd be used today. This is especially the case with fictional leagues; if you have an environment with a 4 man rotation and 3 relievers per staff, the draft will generate 4 starters and 3 relievers for every 7 pitchers. Many of those relievers will be of the 2 pitch "relief only" variety and there really isn't a great deal you can do to change that except something crazy, like, I don't know monkeying with roster settings the day before the draft is announced or something.

As for the lack of real rotations at the time, yeah, the AI just doesn't know how to do that, in part because managers at the time didn't really have a set gameplan to go by either. Occasionally you'd have a Connie Mack using his ace as a stopper in big-game situations but otherwise sending him out to start, more often you'd have guys like the Tigers 'pen you mentioned having starters getting put back a day or two to relieve, say, 5-10 times a year, and some teams just went whole-hog in putting out their top 3 or so guys whenever they could. I think that the *general* notion was to just start the best player who was ready to go, which is kind of lowkey very, very far removed from how pitchers get used today.

Frankly, batting strategies area almost as fraught. I think the game winds up having the pitcher sacrifice way too often and position players not nearly often enough in the deadball era especially and it's my general sense that they don't employ the hit and run nearly often enough (although TBF batters are also generated with the assumption that hit and runs will get tried a "normal" amount, so upping the usage could make stuff really, really weird). Also of course even in the deadball era substitutions of any kind - pinch hitters, pinch runners, defensive subs - were super, duper rare and they just... aren't in OOTP.

All in all, it's built to look like 2020, which means it does pretty well (I think, anyway) from the 70s onwards, not too badly from the postwar era on up, and different shades of "you're gonna need to squint" for any time period before that. I don't know that I have good advice on that outside of "manage all the teams" and "try to show discrepancies through, like, dozens of seasons when you present them to the devs".

[Syd Thrift]

Originally Posted by RchW

This is true for me. Frankly where I have totals off it’s generally where I made a mistake transcribing an LTM from my excel file or I changed one of the general strategic tendencies (say bunting) without changing the LTM accordingly.

Sometimes, too, especially in deadball era leagues you can, for example, have Babe Ruth miss all of spring training and not be on the Opening Day roster but still play most of the season. A 1920, for instance, calibrated without Ruth but run with Ruth will very likely have way too many HRs (and Ruth himself will probably hit like 80).

[joefromchicago]

Originally Posted by Syd Thrift

This is going to be a thing that mostly goes beyond LTMs then, and I just don't think the AI is going to do what I think you're asking it to do, i.e. use pitchers in the 20s the way they'd be used today. This is especially the case with fictional leagues; if you have an environment with a 4 man rotation and 3 relievers per staff, the draft will generate 4 starters and 3 relievers for every 7 pitchers. Many of those relievers will be of the 2 pitch "relief only" variety and there really isn't a great deal you can do to change that except something crazy, like, I don't know monkeying with roster settings the day before the draft is announced or something.

I agree, but I think we're getting a little sidetracked here - I didn't start the thread to deal with the player-usage issue. I mentioned that because I was asked how I was playing the game. So the league totals/modifiers issue has nothing to do with that.

And I'd agree that the game can't do a very good job recreating the way pitchers were used in the pre-reliever era, but I think the game can do a better job than it's doing right now with a few adjustments. Of course, the best solution would be to reconfigure the entire OOTP pitching model, but I don't think there's much chance of that happening any time soon. My investigations, therefore, are focused on finding ways to make OOTP more realistic as it currently exists while advocating for small changes that are reasonably possible within the existing framework.