Fielding with 2B and SS problems

[Garlon]

I imported a game from OOTP20 to OOTP21. The player fielding ratings were not changed, however, now players who were ok fielders are putting up -40 ZR at 2B and SS in OOTP21. Has the defensive scales been changed for OOTP21 requiring higher Range and Arm to play infield positions? If so, it seems that any game being imported from OOTP20 will have these issues.

The game is also changing the strikeouts of some pitchers in the editor but not others. Kimbrel for instance in my league was boosted from 136 K per 300 AB to 153 K per 300 AB. Why it affects some pitchers but not others I do not know, other than perhaps the game is adding strikeouts to pitchers who are already very good but leaving other the same.

[Matt Arnold]

We have opened up the in-game cap on stuff, so I'm guessing a guy like Kimbrel should have exceeded those ratings before but got capped by it, whereas now can use the larger range.

As for the defense, we have tweaked and adjusted some of the in-game fielder model. So while the explicit ratings scale hasn't really changed, it's possible that more marginal players are getting hit harder. Or if it's just a one season sample, it's possible that some players got really unlucky. Hard to say exactly, though.

[stealofhome]

Originally Posted by Matt Arnold

We have opened up the in-game cap on stuff, so I'm guessing a guy like Kimbrel should have exceeded those ratings before but got capped by it, whereas now can use the larger range.

This seems to break with starters, though I've had a few pitchers on an import from 20 to 21 that exceed the stuff rating and turn into 12 WAR monsters.

[Garlon]

That may not be what is happening. This was a file where Kimbrel originally had 300 AB against 31 BB 3 HBP and 136 K (keep in mind that the 136 was already a boost that OOTP20 applied to him because he was a reliever)...and his K went to 153 in OOTP21. I really think something else might be at work here. Pstrickert used my file with OOTP21 and Hanley Ramirez struck out over 300 times in the season, while when I play the same file on OOTP20 with the same settings he strikes out less than 200 times in a season.

For Kimbrel the stats that I had originally put in the editor for OOTP20 were 2221 AB, 42 HR, 227 BB, 19 HBP, and 836K

OOTP scaled this to 300 AB, 6 HR, 31 BB, 3 HBP, and 136 K. All of tthose values besides K are correct scaling in OOTP 20. The K's should be 113, but OOTP give relievers a boost to stuff by design because they rarely face an opposing pitcher so they earn tougher strikeouts and this is fine. So originally Kimbrel's strikeout rate before the boost was (836)/(2221+227+19) = 0.340. He struck out 34% of his Batters Faced. After the OOTP boost to stuff he receives (136)/(300+31+3) = 0.407. So the game boosted him from 34% strikeouts to 40% strikeouts...ok, that's quite a bit but it is fine. In OOTP21 when importing the league file over he receives 153 K in the editor. This gives him (153)/(300+31+3) = 0.458. This should not be happening here. I originally entered 34% strikeouts for him and he ends up at 45%. 45/34 = 1.34, so in total he is getting a 34% boost to his stuff. So consider the fact that instead of Kimbrel striking out 113 batters per 300 AB against, he now ends up with 23 more from the extra boost in OOTP20 and then another 17 more with an additional boost from moving to OOTP21. He original strikeout rate was just under 34% and now it is at 45%. This is not because he had maxed out the scale, rather the game has made him substantially better for some reason. I think for the upper end of the scale perhaps you opened it up but when you convert from a previously saved league instead of going to the editor to create the new rating you are having OOTP look at the stuff rating and reconvert this to the editor. The actual strikeout rate that represented a given value in OOTP20 must be different for some values and this is not being accounted for when converting to OOTP21.

This same issue may be affecting some batters with high strikeout rates as well as people have been posting issues with players striking out over 300 times with the OOTP21 major league season file.

This may only be happening for players in extreme ranges. I noticed that Pedro Strop and other pitchers had the correct values in the editor when moving from OOTP20 to OOTp21, so this is really an issue converting the upper and perhaps lower ranges of the scales that you mentioned were adjusted for the new game. The math just needs to be undone before it is redone when converting.

Please find a way to correct this issue. The old stuff and Avoid K ratings need to be reconverted back to a strikeout rate and then reconverted to the new values for OOTP21. Perhaps this was being attempted but there is a fraction in the wrong place in the conversion process.

Any changes you have made with fielding is pretty much going to trash any leagues coming in from OOTP20 now because players who previously had ratings that worked are now not usable without missing 25% of the plays they should be making.

If the major league file for this season that was launched with the game was built using OOTP20 or the old scales then that entire file has these problems and that file needs to be reconverted too. This will even cause problems for baseball reference which is using this game to produce simulated results for the season.

[Garlon]

So I just now tested Pedro Strop with OOTP20 and OOTP21 by typing his original values from OOTP20 that I used into the editor in OOTP21.


First I did this in Kimbrel's editor and just put in the Pedro Strop values:
The original values I typed in were 2020AB, 36 HR, 221BB,32HBP, 469K.
OOTP20 scaled this to 300 AB, 5HR, 33BB, 5HBP, and 83K
I retyped these original values into OOTP21 in the editor and it scales to:
300 AB, 5HR, 33BB, 5HBP, and 95K

So you can see here that the game is now applying something wrong in the scaling process. Strop has a raw strikeout rate of 20.6% being typed into OOTP20 and the game boosts him to 24.5% .245/.206 = 1.189, a significant boost even in OOTP20 for a reliever.

In OOTP21 he now ends up with 28.1% for his strikeout rate. .281/.206 = 1.36. OOTP gives him a 36% boost or basically double what it should be boosting him. However, the issue is that Pedro Strop will actually import correctly and end up with the same 83k in the editor as OOTP20. So the conversion is ok for some pitchers in the part of the range that probably was not adjusted.

Then I went into Pedro Strop and typed in those same original values and it converted it to 83 strikeouts as it originally did in both OOTP20 and the conversion to OOTP21.

This suggests that something happens in the boost process that is hidden and is attached to some pitchers but not others. Maybe there is something in the code that is applying the reliever boost twice to certain imported relief pitchers but not others. Maybe these are pitchers whose stuff value was originally part of the range that was adjusted for the new game.

Please look into this further as there is a bug here that needs to be found as you shouldn't be able to get two different editor results using the same stats for different players.