[rasnell]

Drilling down in depth in a search for bugs with Patch #2, here is what I've found in a fictional league of 16 teams, with historical player creation and era modifiers clicked on: (I simmed 14 seasons from the 1920-40 era, fictional).

1. Stolen bases in the default settings are still way too high -- about five times higher than the historical setting and the historical modifiers do not reset or correct the problem in subsequent seasons. Teams average more than 500 stolen bases per year.

2. Home runs in the default settings are too low by about 15 percent, even using the historical modifiers in the default game.

3. Manager date-of-birth field is faulty. No matter what birth year you choose, it gets overridden and the age is incorrectly reported.

4. By mid- to late-season, or some sort of minimum at bat and IP, the AI is going to have to be tweaked to begin considering actual stats over the ratings. Exceptional players with average ratings should obviously be in the lineup. We're talking .380 hitters after 300 at bats who are #3 in the depth charts because someone out-ranks them in the ratings. Remember when we had this problem as far back as OOTP 3 and Markus gave us the ability to give a weighted average to ratings vs. actual stats. I think we need this option again.

5. Same problem with on-base vs. power for leadoff hitters. Markus gave us weighted averages that we could set and tweak and this game has the same flaws.

6. Wild pitches are 3 times higher than the MLB norm. This bug appears in exactly the same magnitude in historical or fictional sims so the problem is not simply the field in the Lahman database. If you show all pitchers after a 14-year sim and sort by wild pitches for an individual season, you'll find up numerous pitchers with 40-60 wild pitches, and some even exceed bases on balls for those individual pitchers. If you divided these totals by 3, it would be closer to realistic.

7. In selecting the leadoff hitter, the AI is wrong 10 out of 16 times and really, really wrong in 5 of those 10 times. I gave maximum benefit of the doubt to the AI. By looking at all 16 teams, sorting first by contact, then by steal, eliminating the top three hitters for power (assuming they should be in the middle of the lineup), and then considering eye and K's as a tie-breaker, the AI is really, really missing the mark. In three examples, after seeing the best contact and power combos are in the middle of the lineup, there was still a 90 contact, 80 eye, 100 steal available for leadoff and the AI selected a 45 contact, 15 eye, 80 steal for the leadoff position. Something is very wrong in what the AI is being told to look at in selecting the leadoff hitter. (I accounted for fielding, positions and gave benefit of the doubt to whom the AI would want defensively; still doesn't check out and is very wrong).

8. There are minimal problems in how AI is selecting the slugger. Somehow they are ending up in the second or sixth spots when they clearly should be #4, but this glitch is infrequent.

9. The AI is very reliable in choosing the best contact-power hitter for the #3 spot.