The problem now too big to ignore - playing out of place...

[david limbaugh]

Originally Posted by Matt Arnold

The only times I've seen people start at a position they're not rated for is if there's nobody else on the roster who can play there, or maybe on the off-chance where the roster is weird (ie. better to have an 18/20 2B and a 0/20 3B than a 7/20 at both positions).

I did want to have a look through our logic to determine starters and which positions to put guys at (I have seen cases of playing a clearly inferior lineup, but haven't got around to fixing that), so when I come around to that I'll definitely poke around more and make sure this will only happen if truly nobody can play the position.

Matt, this AI manager is brutal at making a lineup!

Catcher Nat Hicks, not qualified to play 1B, Only rating is at C, makes 20 errors in 7 games playing 1B, 20 errors in 73 chances at 1B! .726 fielding.

Team has 2 players rated at 1B.... On that same team Jim Devlin, one of those 1B, qualified at 1B, played 6 games at 3B, and made 17 errors in 31 chances, a .474 pct. he is not qualified to play 3B. Roster has 2 players who are rated at 3B....

Same manager puts John Radcliff at SS - his ratings are a 34 at 3B and 40 at RF. But Manager puts him at SS over 2 other rated SS's. In 16 games he has 36 errors on 115 chances - a .687 pct.... Somehow this team is 6-10

http://www.ootpdevelopments.com/boar...1&d=1433365154

http://www.ootpdevelopments.com/boar...1&d=1433365315

http://www.ootpdevelopments.com/boar...1&d=1433365432

[Doc_Brown]

Noticed this happening a lot and it's annoying.

Latest example:

Bottom of the (9th tied game);

AJ Pierzynski replaces J.Peraza who was playing 2nd and batting 8th then
J.Johnson who's hitting in the 9 hole is taken out of the game, the Braves use P.Gosselin to pinch hit.

Now instead of keeping Gosselin in the game, the AI keeps Pierzynski in and plays him at 2nd.

[David Watts]

Originally Posted by Doc_Brown

Noticed this happening a lot and it's annoying.

Latest example:

Bottom of the (9th tied game);

AJ Pierzynski replaces J.Peraza who was playing 2nd and batting 8th then
J.Johnson who's hitting in the 9 hole is taken out of the game, the Braves use P.Gosselin to pinch hit.

Now instead of keeping Gosselin in the game, the AI keeps Pierzynski in and plays him at 2nd.

I wonder if AJ changed mitts?

[Doc_Brown]

Originally Posted by David Watts

I wonder if AJ changed mitts?

Not sure, but he did start an inning ending DP to finish the top of the 10th.

[david limbaugh]

Orator Shafer benched 16 games in for Buttercup Dickerson. This occurred after monthly scouting appraisals....

http://www.ootpdevelopments.com/boar...1&d=1433720219

http://www.ootpdevelopments.com/boar...1&d=1433720233

Orator HOT!

http://www.ootpdevelopments.com/boar...1&d=1433720252

Buttercup not as good:

http://www.ootpdevelopments.com/boar...1&d=1433720252

than Orator:

http://www.ootpdevelopments.com/boar...1&d=1433720252

For some reason manager Harry Wright benched Orator for Buttercup?

[The Wolf]

It's pretty obvious that this needs fixing.

[david limbaugh]

Originally Posted by The Wolf

It's pretty obvious that this needs fixing.

1885 - 18 games in - new month, re-scouted....

AI Manager benches previous year's League Leader in HRs, who is also leading the league this year:

http://www.ootpdevelopments.com/boar...1&d=1434056434

For another player who is far inferior....

http://www.ootpdevelopments.com/boar...1&d=1434056434

Reilly is far superior, yet benched?

http://www.ootpdevelopments.com/boar...1&d=1434056434

This is just frustrating....

[RISP!]

Yea, this is definitely something the AI is terrible with. You can see it on AI teams all the time, where some guy with very poor defence is playing at a certain position for no reason when they have a better guy at that position to play defence. I think AI just has trouble generating appropriate depth charts for teams.

So it is important to look over your lineups and depth charts yourself rather than relying on AI alone, OR play out your games and make your own substitutions that make sense.

[Biggio509]

Originally Posted by david limbaugh

I agree errors were common back then Gloves weren't quite the same. I have posted several examples in Future Release thread - please see link above...

Gloves? Only sissies wear gloves.

Biddie McPhee

(I made that up and attributed it to him but he may have really said something similar).

[David Watts]

[QUOTE=RISP!;3881124]Yea, this is definitely something the AI is terrible with. You can see it on AI teams all the time, where some guy with very poor defence is playing at a certain position for no reason when they have a better guy at that position to play defence. I think AI just has trouble generating appropriate depth charts for teams.

So it is important to look over your lineups and depth charts yourself rather than relying on AI alone, OR play out your games and make your own substitutions that make sense.[/QUOTE

nm

[Biggio509]

Originally Posted by RISP!

Yea, this is definitely something the AI is terrible with. You can see it on AI teams all the time, where some guy with very poor defence is playing at a certain position for no reason when they have a better guy at that position to play defence. I think AI just has trouble generating appropriate depth charts for teams.

So it is important to look over your lineups and depth charts yourself rather than relying on AI alone, OR play out your games and make your own substitutions that make sense.

I agree with you last statement. You can't put any AI good or bad on autopilot and expected to act like you. That being said, my question is this consistent or just a few games. In other words does the inferior guy start most of the season or does he come in for a few games. The former is a problem with AI logic. The later is what should be happening to rest players. The question is not does an inferior player play but how often.

I sim my games and still find my backups getting about the right amount of bats due to the manager subbing in the back ups to rest the starters or when there is a day to day. Sometimes if I click let AI to setup the depth chart. I only sometimes find the AI disagrees with the starters on the ML level. Usually if it does it is because it is trying to take advantage of splits. Backups can have a strong disagreement but is usually because I want the lighter hitting better SS or CF to be the backup there over a better hitting poor fielder.

Are the settings traditional or sabr for lineups? If it is SABR the overall numbers may not matter much. The AI will try to take advantage of splits. With SABR selected I find a lot more platooning with a bad fielder who hits right or lefties better. I find a lot less of this with traditional settings.

Also note that ratings are only part of AI evaluation. The lineup code is pretty black box but if it is paying attention to fielding stats a guy not used much might have great fielding stats just because he hasn't had the opportunity to make errors.

[david limbaugh]

Originally Posted by Biggio509

I agree with you last statement. You can't put any AI good or bad on autopilot and expected to act like you. That being said, my question is this consistent or just a few games. In other words does the inferior guy start most of the season or does he come in for a few games. The former is a problem with AI logic. The later is what should be happening to rest players. The question is not does an inferior player play but how often.

I sim my games and still find my backups getting about the right amount of bats due to the manager subbing in the back ups to rest the starters or when there is a day to day. Sometimes if I click let AI to setup the depth chart. I only sometimes find the AI disagrees with the starters on the ML level. Usually if it does it is because it is trying to take advantage of splits. Backups can have a strong disagreement but is usually because I want the lighter hitting better SS or CF to be the backup there over a better hitting poor fielder.

Are the settings traditional or sabr for lineups? If it is SABR the overall numbers may not matter much. The AI will try to take advantage of splits. With SABR selected I find a lot more platooning with a bad fielder who hits right or lefties better. I find a lot less of this with traditional settings.

Also note that ratings are only part of AI evaluation. The lineup code is pretty black box but if it is paying attention to fielding stats a guy not used much might have great fielding stats just because he hasn't had the opportunity to make errors.

Using Traditional.

My frustration is the opposing teams' lineups when their AI manager sits their best player for someone not nearly as good...

[endgame]

Originally Posted by david limbaugh

Using Traditional.

My frustration is the opposing teams' lineups when their AI manager sits their best player for someone not nearly as good...

This is a conversation I know we'll be having prior to 17. It's a big, big, picture with fairly divided opinions and interpretations between users with experience in the game and at least one major component of the development team.

But to your comment, david. I've seen this, too, but don't always take the time to try to come to terms with the why behind it. Sometimes it is a matter of rest, but coupled with that I'm particularly taken back when that player who is sitting is suddenly used as a PH late in the game. IMHO, unless it's a playoff game (in which he'd likely be playing) or a tight race in the division, then if he's going to sit, let him sit. I see this with starting catchers occasionally, too.

[ScottM816]

I experienced this last night when Lorenzo Cain went down with an injury. Instead of Shifting Jarrod Dyson over to CF from LF and moving Whit Merrifield from 2B to LF and then putting an IF in at 2B. They decided to bring in Christian Colon, who has no experience in at center field.

[Biggio509]

Originally Posted by ScottM816

I experienced this last night when Lorenzo Cain went down with an injury. Instead of Shifting Jarrod Dyson over to CF from LF and moving Whit Merrifield from 2B to LF and then putting an IF in at 2B. They decided to bring in Christian Colon, who has no experience in at center field.

That raises the question if the AI is actually programmed to shift positions upon a substitution or if it just puts a guy in the injured guy's position.

[Biggio509]

One thing to note also is this is not unrealistic as you think. Craig Biggio was moved to CF when Jeff Kent didn't want to play 3B. Lance Berkman was also moved from 1B to RF. At one time the Astros had 3 IFs in the OF. A real life manager did this to keep the big hitters in the game despite the huge drop off in D. There is footage of Berkman going up Tal's Hill to catch a fly ball!

I think the position rating is also overrated. I think this is more evidence that it has no effect on the game. We have to remember no position rating does not mean a player can not play a position. What it means is he is lacking something managers generally want at that position. A slower CF might still be a good fielder in that he doesn't make many errors. Part of the reason for that is he doesn't get to as many balls. By the game metrics any one can get a rating at corner OF or 1B. All other positions no rating means he doesn't say have the arm for 3B or SS, the range for CF, SS, 2B, or the turn DP for SS, 2B, and 3B.

What we may or may not be seeing here is the AI is not looking at position ratings but how the 3 OF or 4 IF ratings stack up. The AI may be making a decision that a slower CF that has no rating due to lack of range is better than an error prone CF. It may be saying a 3B who can't turn a DP is better than one with little range and error prone. In both cases the player will not have a rating but still might perform better than a player with a rating. One guy has all the basic tools to play say CF but is barely. While one guy lacks one tool but makes up for it in the other tools.

I think the position ratings are only a quick guide for players and have no real impact on the game. It is a way to quickly see how guys stack up. As a manager sometimes I ignore the position ratings. If I actually have range in RF and LF I might go with a lower range guy but low error guy like Biggio. If my SS has plus range a higher rated 3B who has more range but less something else may be benched for a lower range 3B who has better error, arm, and TDP.

I don't think looking at fielding ratings or lack of alone is a good comparison. A lack of a rating means a guy is lacking one tool. That doesn't mean if say you are willing to give up some DPs a 3B with low TDP may not be better than one with a high TDP and a rating if his error rating, range, and arm are better. The only thing holding him back is TDP. I might in some cases go with the guy with no rating.

Some have accused the game of having a hidden fielding rating. I don't believe that is the case. I think the AI simply doesn't use the fielding rating and for each position the code somehow weights the component fielding ratings that help form a fielding rating. In some cases that means a guy with no rating might be better than a guy with a rating. A guy lacking one tool may be better than a guy who has all tools but is borderline in all his ratings.

[ScottM816]

One thing I'm also trying to figure out is why KC keeps using Christian Colon at SS instead of Escobar...Colon is batting .187 in 52 games and has a 55 rating at SS while Escobar is hitting .358 in 21 games with a 60/80 rating at SS. I'm using 0-67-22-11 AI settings with Sabermetric setting.

[bigmonk]

Originally Posted by Biggio509

One thing to note also is this is not unrealistic as you think. Craig Biggio was moved to CF when Jeff Kent didn't want to play 3B. Lance Berkman was also moved from 1B to RF. At one time the Astros had 3 IFs in the OF. A real life manager did this to keep the big hitters in the game despite the huge drop off in D. There is footage of Berkman going up Tal's Hill to catch a fly ball!

I think the position rating is also overrated. I think this is more evidence that it has no effect on the game. We have to remember no position rating does not mean a player can not play a position. What it means is he is lacking something managers generally want at that position. A slower CF might still be a good fielder in that he doesn't make many errors. Part of the reason for that is he doesn't get to as many balls. By the game metrics any one can get a rating at corner OF or 1B. All other positions no rating means he doesn't say have the arm for 3B or SS, the range for CF, SS, 2B, or the turn DP for SS, 2B, and 3B.

What we may or may not be seeing here is the AI is not looking at position ratings but how the 3 OF or 4 IF ratings stack up. The AI may be making a decision that a slower CF that has no rating due to lack of range is better than an error prone CF. It may be saying a 3B who can't turn a DP is better than one with little range and error prone. In both cases the player will not have a rating but still might perform better than a player with a rating. One guy has all the basic tools to play say CF but is barely. While one guy lacks one tool but makes up for it in the other tools.

I think the position ratings are only a quick guide for players and have no real impact on the game. It is a way to quickly see how guys stack up. As a manager sometimes I ignore the position ratings. If I actually have range in RF and LF I might go with a lower range guy but low error guy like Biggio. If my SS has plus range a higher rated 3B who has more range but less something else may be benched for a lower range 3B who has better error, arm, and TDP.

I don't think looking at fielding ratings or lack of alone is a good comparison. A lack of a rating means a guy is lacking one tool. That doesn't mean if say you are willing to give up some DPs a 3B with low TDP may not be better than one with a high TDP and a rating if his error rating, range, and arm are better. The only thing holding him back is TDP. I might in some cases go with the guy with no rating.

Some have accused the game of having a hidden fielding rating. I don't believe that is the case. I think the AI simply doesn't use the fielding rating and for each position the code somehow weights the component fielding ratings that help form a fielding rating. In some cases that means a guy with no rating might be better than a guy with a rating. A guy lacking one tool may be better than a guy who has all tools but is borderline in all his ratings.

Well, this is a frightening concept! Are you claiming the computer manager knows about and uses managing techniques of which the game player is ignorant?
Does the computer manager know how to juggle the abilities of unrated players to properly use them in the game and I have no way of knowing how or why that works? Crackers in a barrel!
When we started playing APBA or Strat, the cool part was looking at the cards and seeing how the company's ratings of the player compared with our idea of how the player should be rated. Then you played with the card and the ratings directed the course of play. That's what made the game fair. I knew what was on your cards and you knew what was on mine. Then the dice decided the game.
But if the computer uses ratings we don't know about ...

[The Wolf]

Originally Posted by ScottM816

I experienced this last night when Lorenzo Cain went down with an injury. Instead of Shifting Jarrod Dyson over to CF from LF and moving Whit Merrifield from 2B to LF and then putting an IF in at 2B. They decided to bring in Christian Colon, who has no experience in at center field.

That should definitely never have happened. This needs fixing.

[david limbaugh]

Originally Posted by Biggio509

One thing to note also is this is not unrealistic as you think. Craig Biggio was moved to CF when Jeff Kent didn't want to play 3B. Lance Berkman was also moved from 1B to RF. At one time the Astros had 3 IFs in the OF. A real life manager did this to keep the big hitters in the game despite the huge drop off in D. There is footage of Berkman going up Tal's Hill to catch a fly ball!

I think the position rating is also overrated. I think this is more evidence that it has no effect on the game. We have to remember no position rating does not mean a player can not play a position. What it means is he is lacking something managers generally want at that position. A slower CF might still be a good fielder in that he doesn't make many errors. Part of the reason for that is he doesn't get to as many balls. By the game metrics any one can get a rating at corner OF or 1B. All other positions no rating means he doesn't say have the arm for 3B or SS, the range for CF, SS, 2B, or the turn DP for SS, 2B, and 3B.

What we may or may not be seeing here is the AI is not looking at position ratings but how the 3 OF or 4 IF ratings stack up. The AI may be making a decision that a slower CF that has no rating due to lack of range is better than an error prone CF. It may be saying a 3B who can't turn a DP is better than one with little range and error prone. In both cases the player will not have a rating but still might perform better than a player with a rating. One guy has all the basic tools to play say CF but is barely. While one guy lacks one tool but makes up for it in the other tools.

I think the position ratings are only a quick guide for players and have no real impact on the game. It is a way to quickly see how guys stack up. As a manager sometimes I ignore the position ratings. If I actually have range in RF and LF I might go with a lower range guy but low error guy like Biggio. If my SS has plus range a higher rated 3B who has more range but less something else may be benched for a lower range 3B who has better error, arm, and TDP.

I don't think looking at fielding ratings or lack of alone is a good comparison. A lack of a rating means a guy is lacking one tool. That doesn't mean if say you are willing to give up some DPs a 3B with low TDP may not be better than one with a high TDP and a rating if his error rating, range, and arm are better. The only thing holding him back is TDP. I might in some cases go with the guy with no rating.

Some have accused the game of having a hidden fielding rating. I don't believe that is the case. I think the AI simply doesn't use the fielding rating and for each position the code somehow weights the component fielding ratings that help form a fielding rating. In some cases that means a guy with no rating might be better than a guy with a rating. A guy lacking one tool may be better than a guy who has all tools but is borderline in all his ratings.

I don't mind the game 'mimicking' what happened in real life (a lot of players have moved from position to position over the years) but I don't want the game to do things that have never happened yet or never will.

All I want is an option that lets me tell the game a player can not play a position he does not have a defensive rating for...