|
||||
|
01-29-2009, 02:05 PM | #1 |
Moderator
Join Date: Aug 2008
Posts: 3,109
|
Authentic gameplay: CG frequency
I’ve been analyzing historical gameplay (both outcomes and AI decision-making) by reading box scores and game logs from real life games while playing the same games in OOTP (my test so far has only been with a 1960 league). My objective is to figure out exactly what OOTP does differently from real life, in individual games, and to experiment with different settings to get the closest possible match that I can. That is, to have games in OOTP play out in ways that a GM, manager or fan of that time would recognize. I thought I would share some of my findings with respect to specific areas of gameplay. My apologies for the length of this post, but this is aimed at players who are particularly interested in fine-tuning gameplay (as distinct from those who focus on seasonal or career statistics).
One area where OOTP games diverge significantly from historical experience is CG frequency. I’ve found that, in non-DH leagues, the most important reason is pinch-hitting. Generally speaking, starting pitchers in the past were not taken out for pitch hitters as long as they were pitching effectively. Of course, this varied somewhat from team to team, and by the situation in a given game. But, all things being equal, starters were allowed to bat far more often in late innings, in close games, than would be true today. There are two adjustments that affect this. One is the League strategy setting for PHing for pitchers. The other is the same setting in each team strategy page. Between these two settings it is possible to get much more appropriate CG frequency for a given era. In the 1960 league, for example, I found that setting League strategy to “Rarely”, while having the team strategy setting in the range of middle to frequent produces the best results (I tend to vary the latter from team to team, so you don’t get everyone doing the exactly same thing). An early era league might set the League strategy to “Very Rarely.” A more recent era might set League strategy to “Normal” but use lower settings in the team strategy page. Players who make this adjustment have to be prepared for the result. We are used to seeing pinch-hitters for starting pitchers routinely in late innings (and even sooner). You may find it disconcerting to see an SP coming up to bat in the eighth inning with his team down 2-0. However, that is what often happened IRL (depending on the era, of course). It is why that pitcher could end up with a 3-2 CG win, which would otherwise not happen. Note: as near as I can tell, these adjustments do not produce unrealistic behavior in pinch-hitting for relievers, when I compare to what happened IRL. |
01-29-2009, 02:05 PM | #2 |
Moderator
Join Date: Aug 2008
Posts: 3,109
|
Other Settings:
User of closers: the League strategy setting for “use of closers” is a factor in CG frequency. If set to “Normal,” OOTP will bring in a closer whenever there is a save opportunity, which greatly reduces CG frequency. I find that “Rarely” works well for the 1960 era. “Very Rarely” is probably best for early eras. Unfortunately, there is no parallel setting in the team strategies page, so it is impossible to fine-tune this.
Endurance: endurance is a factor in getting good CG frequency (and is the most important factor in DH leagues). However, I don’t believe that this is because the game’s endurance modeling is wrong (at least, not that I could “prove”). Still, an adjustment is needed for another reason: the rigidity of starter rotations in OOTP. IRL, managers were more flexible in the use of pitching staff than the AI can be (also the AI can’t plan ahead for future games). In OOTP, starting rotations often get over-used, especially in seasons with double-headers (the AI is completely at sea in dealing with these). So, I concluded that an adjustment to endurance was essential to compensate for this rigidly in OOTP. In other words, it has to be reasonably possible for an SP to complete a game even if at 80-90% rest, in order to get authentic CG frequency. The problem with adjusting the League setting for endurance to “High” is that it affects all pitchers, causing relief pitchers to pitch too long. It also causes AI managed teams to reduce the size of the pitching staff to unauthentic levels. The better adjustment is to the League totals modifier for starting pitchers. I find that setting this to between 1.100 and 1.200 (I actually use 1.500) seems to produce the best compensation for the rigidity of the pitching rotation, without distorting any other aspects of authentic gameplay. The hooking SPs setting: People have tried to fix the problem of CG frequency by modifying this setting in team strategies. I believe this is a mistake. Example: in the 1960s era, SPs were routinely lifted when they gave up 4-6 runs (varying somewhat by situation and team). Therefore, I set the hooking SP option in the team settings to get this result (which is roughly in the lower middle range, varying it by team). I would not want SPs getting hooked only after 6-8 runs, just to increase CG frequency. Use of RPs setting: For similar reason, I don’t believe this setting should be adjusted just to increase CG frequency, because it distorts other behavior. Adjusting this setting also causes the AI to adjust the size of pitching staffs inappropriately. FWIW |
01-30-2009, 12:11 AM | #3 |
Hall Of Famer
|
I can also add that trying to fix the CG problem by adjusting the "hook" slider is problematic for reasons you've mentioned.
But I have to ask - why is it an issue if RP's pitch "too long"? OOTP doesn't model RP's endurance correctly as it is (for example, if a RP pitches 2 days in a row, he will be "tired" for sometimes 3-4 days after that, which creates even MORE AI problems. Somewhat unrealistic.) I've actually found it better that RP's can pitch longer and more often under "High" or "Very High" endurance, IME. |
02-01-2009, 03:09 PM | #4 | |
Moderator
Join Date: Aug 2008
Posts: 3,109
|
Quote:
Second: If you modify pinch-hitting for pitchers as I recommend (which is essential to get realistic CG frequency), and at the same time increase endurance for all pitchers, you may get relief pitchers lasting longer in games than they typically did. IRL, once a manager got into the bullpen, he was also much quicker to pull an RP for a PH. You can't model this behavior in OOTP given the available tools. You can increase "hooking" RP frequency, but that's not exactly the same thing, because it depends on the opponent doing something (e.g., scoring runs) to trigger the hook. On a related note: RP stats for this era can be misleading, because for many RPs, a lot of their IP were gained when they filled in as starters. If you take that into account, OOTP's endurance modeling seems pretty reasonable, at least to me eye. |
|
02-01-2009, 11:59 PM | #5 |
OOTP Historical Czar
Join Date: Dec 2001
Location: Bothell Wa
Posts: 7,254
|
Fwiw
Setting Closer usage to rarely or very rarely also results in said closer (usually the best reliever on the staff) pitching very few innings.
|
02-02-2009, 12:31 AM | #6 | |
Moderator
Join Date: Aug 2008
Posts: 3,109
|
Quote:
One option is to increase the frequency of "hooking relievers" somewhat in the team strategy pages. Closers will come in more often once the AI manager has gone to the bullpen. Beyond that, I don't know any way to fine-tune this. |
|
Bookmarks |
Thread Tools | |
|
|