League Total Modifiers: Finding the Right Files and Numbers

[David Watts]

Originally Posted by Rain King

I did check that setting (it was set to Often), but after the game had adjusted the modifiers. I can test again with setting that ahead of time...but that probably isn't something that users should generally be expected to do to get acceptable modifiers.

Thank you.

[Rain King]

FYI, that did make a HUGE difference.

I set the Bunting tendency to be "Very Often" before advancing to Opening Day and the Sac Bunts modifier was then adjusted to be 1.462 as opposed to 4+.

It sounds like the community could use some kind of helper on how the Tendency settings and Auto-Calc system work together.

[David Watts]

Originally Posted by Rain King

FYI, that did make a HUGE difference.

I set the Bunting tendency to be "Very Often" before advancing to Opening Day and the Sac Bunts modifier was then adjusted to be 1.462 as opposed to 4+.

It sounds like the community could use some kind of helper on how the Tendency settings and Auto-Calc system work together.

I agree. Plus, if we set it to very often to make the modifier come out right, what does that mean in terms of game play? Will sac bunt attempts be close to that year's actual total, or will it instead turn into buntfest central?

[Garlon]

The bunt strategy setting are basically sone static modifier then the league totals modifier on top of that is able to target the appropriate value. These two things work in tandem so boosting the strategy setting to very often should not have any detrimental effect.

[actionjackson]

Originally Posted by David Watts

I agree. Plus, if we set it to very often to make the modifier come out right, what does that mean in terms of game play? Will sac bunt attempts be close to that year's actual total, or will it instead turn into buntfest central?

Buntfest central sounds like a festival where you gorge on Bundt cake the whole time. Delish, but hazardous to your health.

[Charlie Hough]

Originally Posted by Rain King

Charlie, have you seen this issue when starting in a year other than 1979?

Yes, I just ran three tests with 1978, and the sac bunt modifiers were as high as 3+ or 4+. Those were on initial saved game creation. Then, as an experiment, with these initial numbers in place and the game on opening day, I tried using the auto-calc modifiers button to see what the sim would do. It calculated all modifiers again, and the sac bunt modifier started coming in as high as 9.745 and 10.000!

Other modifiers were generally pretty similar between new saved games or auto-calculations, but occasionally there are differences that can add up to a fairly significant impact on statistical output, such as the error modifier being 0.2 higher for a particular outfield position. But I can live with those kinds of variants.

I also ran some 1914 tests, and the fielding line drive double-plays modifier came in at 6.456, 8.841, and 9.407! ! I'm not sure what the correct number should be for 1914, but wild swings like those certainly don't make sense. Incidentally, in my last test with 1914, the sac bunts modifier was 10.000. So, as I've been trying to convey all along, we have problems across multiple modifiers, and whether certain modifiers are set to an extreme at all, or how extreme they are, seems down to randomness and specific test instances.

In contrast, I've run about five tests with 1985, and the modifiers are all pretty similar between saves. There are still differences every time, which I understand is normal for OOTP, but they're typically minor. Sometimes the sac bunt number can be about 0.5 different between saves, and sometimes there's still more movement in the outfield errors modifiers than I'd like, which seems odd, but I haven't seen any major issues.

At this point, things don't look good. Since there is no consistency between saves, years, or calculations, and the extreme numbers aren't always happening with the same modifier, I don't see how to work around this very easily.

[Rain King]

Auto-Calc works by doing (I believe 3) simulations and adjusting the modifiers after each simulation.

So, it isn't expected to end up with the same values each time.

However, for Sac Bunts, I do think that as Garlon recommended setting the Sac Bunt tendency to "Very Often" before the Opening Day modifier calculation happens seems to be a good work-around for that issue, but it certainly is not great that this is all kind of an unknown to the average user as to how it works and some year's definitely seem to have large issues "out of the box".

[Charlie Hough]

At this point, since this affects modifiers other than sac bunts, and since it's affecting various seasons throughout history and different modifiers in completely unpredictable and extreme ways, I'm going to leave this to the OOTP Developments bug report.

Lukas gave the bug report a preliminary look and replied, but now I've linked him to this thread for more details and verification.

I hope this gets some serious attention because no one should have to adjust tendencies to get the game to calculate proper modifiers, and it would be impossible to chase this across different years and different modifiers, especially with no way to know whether your modifiers are "correct" and which ones are going to be affected from year to year.

[Garlon]

When the modifiers get maxed out it us because the baseline strategy setting is too low. Set sac bunts to often or very often. The game has a historical accuracy report you can reference. It does not have every category in the report but it shows your garbage modifiers accurately target and produce the correct ratios. I have been saying for years that the default strategy settings should be revised. For example, use relievers should probably be set to Normal for every season. Use closers can be customized to the season though.

[Charlie Hough]

Originally Posted by Garlon

When the modifiers get maxed out it us because the baseline strategy setting is too low. Set sac bunts to often or very often.

Yes, you've mentioned this several times, but there are other modifiers being affected depending on the year and sheer randomness, so adjusting the sac bunt strategy is not a solution. Plus, there is always the question of whether changing the strategy settings is realistic. If you're having to change something to a higher or lower frequency than real life, just to get the modifier to be correct, then that's not an acceptable solution. There should be a proper balance where both the strategy setting and the modifier are accurate, realistic, and correct. And the user can't be expected to know what to do for every setting and every modifier for every season in baseball history, in order to override OOTP's problems.

If there needs to be an overhaul of the strategy settings and modifiers to get this right, then OOTP Developments should definitely get to work on it. I'm all for that, so hopefully they'll heed your suggestion for an overhaul this time.

[Alex Murray]

I have been spending some time looking into this and trying to nail down what the game is doing and if something is broken in the game. I'll let you all know what we find and update as the team works with me on what we already have from you all and my personal findings.

Hopefully we will have an update soon.

Thanks
Alex

[Alex Murray]

I was able to do a review of two historical saves to try and narrow down what we can dissuade as normal in historical saves and what is indeed having issues.

So we did confirm that the sac bunt modifier is having an issue that will be looked into. I was able to easily replicate the 10.000 modifier bug and have passed that along to our team.

I was unable to replicate the line drive double play modifier bug that was mentioned as well. I had found some preliminary hints at it but could never duplicate it to be anywhere as serious as the sac bunt modifier was doing.

In regards to the season total modifiers those are working as intended. Now we do have small variations for every single historical season when it comes to the imported season stat modifiers. So you will notice that the season totals on the left never change but the modifiers on the right can be slightly different as those do have that variation. Now those variations should not be very wide in terms of how low and high it can be randomized to. So if you are seeing major jumps in those please do let us know.

I will be doing another historical save to try and recreate that 1915 line drive double-play situation you saw Charlie. Otherwise, I wasn't able to see anything on the other modifiers that would be cause for alarm and of course, we will review the sac bunt modifier at the very least and get that looked into.

Thanks
Alex

[Charlie Hough]

I wanted to follow up on this and mention that league total modifiers for balks are also a big problem, and in general I continue to see problems with league total modifiers not having appropriate starting pitcher and relief pitcher stamina, plus other issues.

With balks, I frequently see OOTP set the league total modifier to over 1.200 or more, which results in an unusually frequent number of balks. While playing out games, this became so pervasive that I finally had to start setting the balks modifier to 1.000 for all historical games, once the game computes modifiers for each season.

Similarly, I've had to adjust starting pitcher and relief stamina because I think OOTP is trying to compensate for my historical pitcher stamina being based on total career and not a 3-year or 1-year recalc. Yes, pitcher stamina ratings are probably higher than usual when basing them on total career stats, but when the game lowers starter or reliever stamina modifiers to compensate, it sometimes does it too much, resulting in pitchers getting fatigued too soon in games.

There needs to be a balance of adjusting for current player ratings and maintaining realism for the season and era in question. Otherwise, you can end up with starters getting tired an inning or so early, and you can see complete games and total innings pitched quite a bit lower than they should be. I would recommend a review and revamp of league total modifiers to make sure they work well with different recalc and historical player ratings settings.

[Garlon]

There is no issue with pitcher stamina. The file has the actual IPouts per GS for every year. The modifier adjusts to your league setting for pitcher stamina. If you look at your starting pitching IPouts let gas and compare it to the real value it will be accurate.

As for balks it also targets the correct amount of balks per IPout for the season. When the modifier increases it does not mean you are getting more balks or more of anything, it is just finding a calibration value do that you get the correct ratios of everything.

There can possibly be an issue of individual pitchers getting more balls than expected though. This is because many pitchers will load with 0 balks in the editor because the editor sample size will round it down to 0, whole you have other pitchers who end up with 1 in the editor. Those pitchers with 1 may end up getting more than their share if balks.

I can ask if they l can update this so that any pitcher who actually did have at least 1 balk in a season historically ends up with 1 in the edition rather than getting rounded down to 0 which can cause some issues with balancing. The overall balks for the league though should still be correct.

[Garlon]

The modifiers will adjust to whatever you are using as imports. They are generated dynamically. I do suggest going into the game engine file and changing the balance if the stamina calculation. Currently it weights part of the calculation on IPouts per GS and part in CG percentage. This can cause an imbalance between the SP and RP. I set the CG weight to 1 in that file.

[Charlie Hough]

Originally Posted by Garlon

The modifiers will adjust to whatever you are using as imports. They are generated dynamically.

Yes, I'm aware of this, but the developers have already acknowledged that league total modifiers are not being calculated the way they're supposed to, at least for some of these categories. So that's the problem, and it's affecting multiple categories in sometimes wildly different ways, across different historical years and eras, and even for the exact same season with identical databases.