Generated players who only have experience at premium positions

[BadluckinOOTP]

When OOTP creates center fielders and shortstops, it sometimes makes excellent defensive players who only have any experience playing center. Numerous times (I sim huge chunks of time) I've seen excellent outfielders be forced to play awful centerfield because the game won't convert them to a corner spot because they have zero experience there, especially when they age and their defensive abilities start to deteriorate, but their bat doesn't.

Mike Trout is an excellent example of this. He's only rated at CF, even though eventually he'll shift to a corner spot or first base because his bat will keep him in the lineup. OOTP doesn't really simulate this very well.

I've taken to trying to identify these guys and editing to add other position But I wish there was another way. My solution would be that any natural centerfielder or shorstop would come with at least a little experience at any position to the left or right of them and maybe that would stimulate the AI playing them at new positions. But, that would probably open a new can of problems.

[andyhdz]

Your right AI isn't savvy enough to convert player positions to extend bat life

[Rain King]

Yeah, we've had a few other threads about this in the past. I don't think the issue is their initial generation. These type of players aren't playing corner spots in high school/college. However, the AI should absolutely understand that they can be moved to those types of positions as they age (or even if the circumstances dictate it i.e. Alex Rodriguez/Marcus Semien).

[Matt Arnold]

One of the changes this year is that the AI will be a little more aggressive at shifting players down the defensive spectrum when they start losing their ability. They still don't overly agressively look for outside options (ie. you will likely still rarely see an AI running a sub-par Trout in CF trade for another CF), but if they happen to have one in their system, they will at least consider shifting him to a corner.

It is always a tough balancing act, since IRL, there are also many cases of players NOT shifting even though their abilities decline. Plenty of cases like Jeter or Tulo who never moved off their main position, despite declining D in their later careers.

[Rain King]

Good to hear!

[Brad K]

The original question was about fictional players and the answer used real players as examples so does this apply to real players too? Presently (on 21) I can train a CF to play a corner but if he didn't real life AI managers don't do it.

1B should be the LAST position a player can play yet historical doesn't work this way. A guy who played 1B early or mid career loses is ability to do it if he switched to LF real life. Guys who end their careers as DHs are totally screwed with someone like me who doesn't allow that abomination in the game.

[Rain King]

Originally Posted by Brad K

The original question was about fictional players and the answer used real players as examples so does this apply to real players too? Presently (on 21) I can train a CF to play a corner but if he didn't real life AI managers don't do it.

1B should be the LAST position a player can play yet historical doesn't work this way. A guy who played 1B early or mid career loses is ability to do it if he switched to LF real life. Guys who end their careers as DHs are totally screwed with someone like me who doesn't allow that abomination in the game.

It sounds like you are referring to how Recalc works in historical games but that is completely different from what this thread is referring to.

The real player used in the example is a current player (Trout) so the end of his career within OOTP is essentially fictional and run by the AI/Development Engine which is typically disabled during Historical play.

[Brad K]

The OP talks about AI created players and then switches to Trout so I'm not sure what the subject is. Anyway, in historical AI fails to train players in related positions also. So its the same thing.

[BadluckinOOTP]

Originally Posted by Brad K

The OP talks about AI created players and then switches to Trout so I'm not sure what the subject is. Anyway, in historical AI fails to train players in related positions also. So its the same thing.

The subject is both, I suppose. I see it a lot because I pretty much only play fictional. I was using Trout as an example, because I've seen the AI trot him out to CF long after his glove was able to handle it.

[Rain King]

Originally Posted by Brad K

The OP talks about AI created players and then switches to Trout so I'm not sure what the subject is. Anyway, in historical AI fails to train players in related positions also. So its the same thing.

Well, I guess there are a few possible things going on here based on what you are talking about and what the OP has brought up.

1.) AI generating players who play primary positions and do not have experience at "corner-type" positions.

This does not seem like an issue to me. Generally, the players that are playing primary positions (i.e. SS and CF) have not had much experience playing corner positions as they have always been the best athletes on their teams.

2.) Historical games generating players that play primary positions and do not have experience at corner-type positions.

Historical games have information on where a player has played throughout their career and you have multiple options on how you want it to look at that and decide where to give a player ratings. However, I don't think it would make sense to have the game give the player experience at positions they did not play in real life.

3.) Current year games generating players that play primary positions and do not have experience at corner-type positions.

The current year games MLB players are manually created (well, most of their ratings are based on a set of projections). However, there is some subjectivity here especially, I believe, when it comes to defensive ratings. Also, they do sometimes give players ratings at places they have not played much as the goal is to get players into the most likely positions for the current MLB season (so Semien was rated as a 2Bman). However, again, I'm not sure in most cases they should be giving players experience at positions they have never played.


4.) The AI moving players to positions they do not currently have experience at as their skills at the "primary" positions diminish in fictional games (as well as the current MLB game which are essentially fictional as it moves into the future). Trout at the end of his career is a fictional player in OOTP terms.

This, IMO, is the main issue that the OP seemed to be getting at and in older versions of the game the AI definitely did not do a good job of this. It would let a player with a 1 defensive rating in CF play there until they retire. However, Matt's statement seems to mean they have improved this in the current version. I have not yet played enough to verify that.


5.) The AI moving players to positions they do not currently have experience at within Historical games.

This is a completely different question than #4 IMO most of the time, although it can depend on your Historical settings. For example, there are historical settings where you can tell the game how to determine where a player gets rated as a fielder, whether they develop using the OOTP development engine and whether the AI should consider the positions they played in real life when determining where they play in your historical game.

If you choose to use the "fictional" type of options here then the AI, from what I have seen, does do a reasonable job of moving players at times. I've seen guys at 1B who only ever played corner OF in real life, for instance. However, you have to tell the AI it is OK to do that kind of thing before it will do so and those definitely are not the default settings.

[Brad K]

Historical games have information on where a player has played throughout their career and you have multiple options on how you want it to look at that and decide where to give a player ratings. However, I don't think it would make sense to have the game give the player experience at positions they did not play in real life

As soon as a person deselects real lineups and real transactions it makes perfect sense. What position a player played RL is a function not just of his skills but also of the skills of his team mates.

AI needs help and letting players get enough of a rating at another position so AI will used them at that position if needed (and the player getting better) is a perfect way of giving AI help. I train IFs to play other positions which gives me an advantage over an AI controlled team. Often when I look for trades I'm offered RF with no LF rating, SS with no 2B rating, etc.

If its off season I can trade for these guys and get them up to competent in another position during spring training. If its during the season I have to pass on the trade.

[Rain King]

There is a Historical setting for "Base player roles/positions on:" which you can set to either "AI evaluation" or "Real life stats".

If you set that to AI evaluation you should see the AI do some of what you are describing.

[OutS|der]

We can't forget a lot of RL players won't switch positions so it's realistic that it doesn't happen as often in game. I'd like to see players personality tie into how willing they are to switch positions. As well switching a popular players position should have consequences with the fans.

The AI does need to be more aggressive when doing to extend bat life though.

[Brad K]

Originally Posted by Rain King

There is a Historical setting for "Base player roles/positions on:" which you can set to either "AI evaluation" or "Real life stats".

If you set that to AI evaluation you should see the AI do some of what you are describing.

My comments are based on observations while playing AI Evaluation on.

[Brad K]

Originally Posted by OutS|der

We can't forget a lot of RL players won't switch positions so it's realistic that it doesn't happen as often in game. I'd like to see players personality tie into how willing they are to switch positions. As well switching a popular players position should have consequences with the fans.

The AI does need to be more aggressive when doing to extend bat life though.

I'd rather the devs work on the nuts and bolts that need some adjusting rather than spending time on increasing the effect of the moan and groan feature.