OOTP10 Suggestions: Artificial Intelligence

[battists]

This suggestion thread is designed to collect suggestions for the next version of OOTP, on the following topic:

Artificial Intelligence

Includes
- Suggestions specifically related to the game's artificial intelligence, including situational AI in the in-game engine, and roster descision-making

Excludes
- Probably some overlap with other threads, but that's OK!

Instructions
We will use this thread to collect suggestions on the above topic. Later on, we hope to present this information back to the community in a way that would allow you to vote and prioritize certain features.

If you would like to participate, here are the rules!

1. Only 1 suggestion per post.
2. Suggestions must be on-topic for the thread.
3. Constructive, non-inflammatory suggestions only.
4. No comments on others' suggestions are necessary. That means, we don't need 15 posts in the thread saying "Great idea!" or "I agree!"
5. Please don't link to other forum threads as your way of making suggestions. ("My suggestion can be found HERE.") If it's important enough to you that you want it in this thread, then it should be important enough to put it in the format I've asked for here!
6. You can certainly post suggestions that conflict with or build upon earlier suggestions.
7. Mockups are welcomed, they really help Markus see what you're interested in accomplishing.

Format for Suggestions

Suggestion
What is your suggestion? Be as specific as possible; include the screen you're talking about, if applicable.

Reason
Why do you think this is important? If this is to correct what you perceive as a problem in OOTP 2007, what is that problem? Be as specific as possible.

Priority
In your opinion, how important is this suggestion?
High (MUST have), Medium (should have), Low (nice to have))


Handy Cut-and-Paste Suggestion Form:

Suggestion


Reason


Priority


Thanks for participating!

Steve

[SteveP]

Suggestion: Reduce the amount of pinch-hitting for position players, by giving the AI an additional parameter to consider in each pinch-hitting decision: the number of times a pinch hitter has already been brought in for one or more position players during that at bat. My idea is that the threshold for the AI deciding to use a pinch hitter would rise automatically each time one is brought into the game during at at-bat (exception: PHing for pitchers). The rarity of the first PH would be determined as it is now (subject to the various settings and sliders that exist in the game). However, a second such PH would be significantly rarer, and a third much rarer still (i.e., almost never).

Reason: the amount of pinch-hitting is excessive and unrealistic. It is quite common in close games for both teams to clear their benches by the end of the ninth inning, or the tenth. I have made all the game setting adjustments that are reasonable, including for AI-managed teams, and I still often see three or more PHs for position players in a late inning. One side-effect is that the AI often brings in batters who have not been hitting well to PH for batters who have better hitting statistics. The bigger problem, however, is that the AI is unable to plan ahead for problems with fielding substitutions. As a result, players often end up in absurd positions in late innings (e.g., catchers at 2nd base, OFs playing catcher, etc.). My guess is that this latter problem is beyond the ability to solve through reprogramming of the AI (exception: a bug I've already reported, BZ 1873, which often occurs when two or more PH-for-position-player moves happen during an at-bat), but the frequency with which it occurs would be greatly reduced by less PHing.

Priority: High. The overuse of pinch hitters for position players (and the consequent mis-positioning of fielders) is by far the most unrealistic feature of how OOTP recreates individual games. In fact, I can't think of anything else that I find significantly unrealistic about how individual games play out, other than this -- but this is a pretty big variance from true simulation of what happens in real games.

[Afino]

Suggestion

- AI Rosters tend to have way too many bench position players and never enough pitchers (which apparently is because they LOVE to pinch hit - see above - I don't like that either), and there's been times where a relief pitcher with horrible endurance ratings gets left in there for 70, 80, 90 pitches before being pulled simply because they HAVE no one else in the bullpen.

Basically, I'm suggesting a better allocation of pitchers and hitters on the 25 man roster regardless of era. Whether that's done with more relief pitchers or even better, bringing a starter into the game in that situation doesn't matter to me.


Reason

- I almost feel bad for those opposing pitchers left out to die by the AI.


Priority

- Low to Medium. Most of the time (80%, at least) it isn't an issue. But it's annoying when it happens.

[John C]

Suggestion: Have team chemistry, player chemistry (double play combinations, pitcher/catcher, etc), lineup chemistry, etc.

Reason: I think we're at the point now that, as GMs, we have to find the right players for our teams. FOF has had this forever...Wolverine's games all have this. If the player does not get along with teammates or coaches, the team will suffer. I think the ones listed will go the furthest towards adding another element to our universes. If we can grow that Tinker-to-Evers-To-Chance, or find someone who makes his pitchers better because of the way he handles the staff.

There is a great, great article on Pat Gillick on ESPN.com that describes the sort of thing I am talking about. You can read it HERE.

Priority: I'd say high. I have never played anything SI...but if there is one thing I'd take out of that game and the games I listed, it is this. This above 2D, this above some online component...it's not a game-killer. But can you imagine how incredible this would be if we can embellish on the personalities of the players, and have to become cognizent of ALL players in our universe, so we can find the right fits for our teams? Hell, the Phillies were awful under Larry Bowa's fiery personality...and yet, the general core of this team has thrived under the more congenial, subtle Charlie Manuel.

There are an unbelievable amount of examples I can pull out. We should be able to make this a reality in our game, too.

[JoseRijo]

SUGGESTION
I think the AI should be smart enough to wait until spring training to call up minor leaguers to the big leagues (also to do other minor league promotions/demotions).

REASON
When I manage my team, during the offseason my 25 man roster drops to like 20 due to FA's leaving and demotions of marginal players. That way I dont have to turn around and promote and re-demote players unnecessarily as I trade and sign free agents. The need of the AI to constantly fill its 25 man during the offseason causes unneeded use of waivers and sometimes the release of quality players.

PRIORITY
Medium. It has a number of downstream effects including overuse of waivers and options and subsequent loss of talent which makes the AI significantly disadvantaged, and can make for easy pickings for humans.

[SteveP]

Suggestion: Allow the AI to use players who are not starting because they are tired (as a result of the "If starter tired" option in depth charts) in the later innings of a game if necessary.

Reason: Right now, the AI appears to treat these players as if they are disabled. They should be available to act as pinch hitters. Even more importantly, they should be available to field their regular positions, if need be. This would reduce the times when players end up in the wrong (for them) positions owing to excessive pinch hitting by the AI.

Priority: Important, both because it is unrealistic for these players to locked to the bench, and because of problem of excessive pinch hitting.

[Raidergoo]

Suggestion
Fix usage of SP in the All Star Game

Reason
CL start getting used in the 3rd inning. That's not good.

Priority
Medium

[Vengeance]

Suggestion: Add an option in player strategy that is similar to "Never use as pinch hitter when tired" but instead make it "Never use as pinch hitter when injured".


Reason: When a player has a day-to-day injury and you don't want him risking further injury by pinch hitting.


Priority: Medium

[Tony M]

Suggestion

Greatly expand the scope of strategy settings and situations

Reason

In online leagues for example strategy can't be set correctly for every match over a period of 7 days (if that's the simming period).

I'll knock together a PDF of changes and attach it later

Priority

High

[ervinn]

Suggestion

Modify 'Minor League Signings/Releases', 'Minor League Promotions/Demotions' in 'Team Control Settings' to allow Minor League Managers to control them but - as an added option - would now be required to ask for approval from the GM (Human Manager) on minor league signings/releases and promotions/demotions decisions.


Reason

I don't trust the AI to make decisions 'behind my back'. I would like their decisions to become recommendations or requests only and have control over the final decisions.


Priority

I would say medium priority because it would make gameplay so much better for me and I hope others as well.

[KGAGE]

Suggestion
Create umpires with different ratings. Some ratings could be strike zone size, temperment, accuracy. Would tie in to things such as ejections/suspensions. Also pitchers might have a better day with umpires who have a large strike zone.

Reason
Adds more unpredicatability to the game, immersion.

Priority
Low

[markprior22]

Suggestion

If possible, please make AI more aware of pitching when end of season rolls around. Basically, the way it is now, a team can have a 5 game lead going into the final weekend and it will still pitch whatever pitchers happen to be next in line. I'm sure it would be difficult but it would be nice if teams could set up pitching for the playoffs when that luxury is available. I also just had a one game playoff for the wildcard where team A pitched a 22-11 pitcher against a 4-7 pitcher who had thrown less than 100 innings and had a 5+ ERA. They had a solid starting pitcher at 85% (threw 43 pitches 4 days before) that should definitely have been given that start. After playing 162 games, it is a shame that the teams don't "realize" that it is all or nothing for one game. Even have starters in the bullpen for those games.


Reason

I think this is very important for reasons stated above. A team plays 162 games and has to A) play a one game playoff or B) go through the last week or so with the division wrapped up and the computer doesn't handle the pitching any differently than it would in mid June.

Priority

Medium to high IMO

[lt2cents]

Suggestion

In making the batting lineup the AI needs to take a better overall look at speed, power, bat skills to more sensibly place players. The AI needs to take a more global look at defensive positions, and shift players to maximize team defense whenever a new lineup is made or substitutions are made. And the AI needs to take a more global look whenever it makes a PH/PR/reliever move.

Reason

Playing out games it is frustrating to see the wierd lineups and lineup moves the AI managers make. Although human managers will vary in how they place players, its hard to imagine any coming up with the lineups often made by the AI. AI substituting needs a lot of work. The AI roster usage is very inefficient in often using too many players. It too often pinch hits for the wrong players, with the wrong players, and too often uses a third player in the field, who sometimes would have been the better pinch hitter.


Priority

High. OOTP has established a great platform. Now its time to go back and refine the details that will make it so much better.

[phillosopherp]

Suggestion:

Have the AI for players have more of an understanding of there capabilities and have them react accordingly. By this I mean that you should not have guys that will hang around in A ball until they are in their 30's most would "retire" before that.

Reason:

It would help get rid of the huge number of pages of FAs in leagues and such, as well as make the game much more believable.

Priority:

Very High.

[ovccsteve]

Suggestion: Give a "Maximize Defense" option in the Depth Charts screen. This would force the AI when simming games to move players around to the best possible defensive alignment when starting subs. (The "Start at __ pos" on the Strategy screen could override the option for certain players.)

Reason: In real life, managers often rest a great defensive player by putting a good-hit/poor-field player at a less demanding position and moving another player already into the lineup into the vacant position. For example, if the 4th OF is a good hitter but lousy defensive player, the real-life manager may sit the CF for a game and move his starting LF to center and stick the stiff in left (or move the starting DH to left, letting the lug just hit). As it is, the only way to sub the hitter for the fielder is to put the big, slow guy in CF. (Seven-day lineups can be used this way, but not as elegantly as "start every X games.")


Priority: I think it should be fairly high. It's normal in real life and it wouldn't seem to be too hard to code.

[Eugene Church]

Suggestion:
Starting pitchers are left in the game too long and give up too many runs if the score differential in the game stays close in a relatively high-scoring game.

Reason:
For instance I just played a game where both starters gave up far too many runs. They would not have been kept in the game except that the game score differential was close the entire game.
One starter gave up 9 runs in 7 innings and the other gave up 6 runs in 8 innings. Realistically, the pitchers were not effective and should have been pinched hit for earlier in the game or they would have been removed for a more effective pitcher.
In Game Strategy I think the AI will pull a pitcher when the differential gets four or more. It never happened in this game. It should be coded to pull a starter when he gives up 6 runs or more.

Hartsdale Hellcats ---100 130 011 - 7 14 1
South Fork Stallions - 200 113 20x - 9 9 2

Priority:
High
This is not a rare occurence in the game. It happens regularly.
This would make the game more realistic in the pulling of ineffective starting pitchers.

[Eugene Church]

Suggestion:
Remove a pitcher once he walks 6 batters in a game.

Reason:
I have seen quite a few times where the pitcher will walk 6-11 batters in a game and not be pulled. This is not realistic. Please code it so that a pitcher will be yanked when he walks over 6 batters in a game.

Priority:
Medium.
It makes the game unrealistic when this occurs. It wouldn't happen in real baseball. At least not as much as it happens in OOTPB.

[Eugene Church]

Suggestion:
Do not take a pitcher out when he has a big lead late in the game, until he has the lead cut in half.

Reason:
If a pitcher has a 10-0 lead late in the game, don't take him out until he gives up 5 runs. Up until the 1980s or 90s, a manager would allow the pitcher to complete the game, if it was not in danger of being lost.

Priority:
Low
But this would make the game more realistic. A pitcher with a 10-run lead should be able to give 5 runs before being relieved. A pitcher with a 15-run lead should be allowed to give up 6 or 7 runs before being removed.

[Eugene Church]

Suggestion:
Remove a pitcher after he gives up 10 hits in a game, except when he has given up less than 4 runs.

Reason:
When a pitcher gives up double-digit hits, as a rule, he is not very effective and should be removed.

Priority:
Medium
It would add a more realistic element to the game.

[PhillieFever]

Suggestion

When playing out games,the AI seems to need a complete overhaul,things like not moving the defense in with a runner on third and less than two outs late in a tie game is unacceptable,there are other things as well,attempting to steal a base with runners on first and third and one out in a tie game in the 8th inning.This are basic baseball strategies that are not being implemented.I'm sure there are other things as well but I think you get the idea.



Reason

Mainly for competitive reasons,I've won countless games simply because of bad managerial decisions on the part of the AI.


Priority

Extremely high,without this being right,nothing else matters