Promotion, Demotion & the AI

PSUColonel · 01-10-2018, 03:53 PM

What goes into the AI's analysis of how players are either promoted, demoted, or released? I'm not sure many people know for sure....but I am convinced it's 100 percent based on relative ratings.

That is all and well, but it also makes the system bit simplistic, and takes some of the player analysis out of it. If a player is a 45, he may be ready for the next level....if he's a 50 he's likely ready, and if he's more than a 50, he is definitely ready. These are relative ratings mind you.

I would like to think performance would have SOME part in the equation...especially at higher MiLB levels.

Any thoughts?

NoOne · 01-10-2018, 04:42 PM

it may be part of the equation already. If AI evaluation includes stats, that means it's using them to some extent.

When it comes to an "~18" year old in AA or AAA, it definitely matters. Anecdote: i currently have (3) 18-20year olds of high quality and seemingly ~aaa to ~aa developed. only 1 of them could put up good #'s at the level their ability deserved. i think i see better development if i avoid poor results (really bad .. ~70ops+ or worse = 30% less than average batting ability give or take). But, i've also seen someone hit that poorly and sky-rocket in development... too many moving parts to be sure.

The ai will promote some very young players quickly. i'd bet in those contexts, many factors are in tune and that's what separates other examples of similar ability/development. they don't get promoted for a reason... likely stats or age or some other factor the ai uses already.

i think "50", is middle of the road for that current level when on relative ratings. That's when i like to move them too, but not 100% sure if it's best either. i don't use relative ratings for minors, but the promotion timing is the ~same. the one drawback to relative is that as the talent ebbs and flows, it's difficult to see it. ie 50 in one year is inherently not the same as 50 in a different year, if it's realtive to 'current' talent in league.

regardless of what is known, there is definitely an optimal way to promote players. i'd wager a significant portion of the time you'll want to be systematic about it... reach X-ability(simplified, use common sense), then promote. i wouldn't be surprised that due to how things work you need to do things differently for certain contexts...

i'm curious is development is 'relative' too... e.g. if A-ball is ~50% developed, then 10current out of 20 potential is the same as 50current out of 100 potential (both 100pt scales) relative to optimal MiL level for development.

so if that is true and "eye" lags compared to % development of other ratings, is it better to keep them at the appropriate level for that current eye ability (as % of development - ie how far current has to go to reach potential) until it catches up?

or, maybe optimal placement is based on overall current ability and not the individual parts for this particular concept...

overall, i don't think there are huge negative effects pertaining to promotion/demotion timing. the AI is terrible at it... if there were the league would always have development problems. i often see ~mlb ready prospects stuck in A-ball. they eventually make it to the mlb at a ~normal age from what i can tell, but they'll have 2-3 years at same lower level just hammering the competition, stats-wise. i know there's a 'slight' impact for being too good for an mil level, but obviously it doesn't hurt 10-20+ of these high-quality prospects a year. future ASG type guys.

PSUColonel · 01-11-2018, 05:49 AM

One good thing about using relative ratings, is it illuminates the need for the awful green and red arrows. Too often the AI players see players as being between levels and can't make up it's mind about promotion or demotion. Other times, the AI just seems way off. When viewing ratings relative to a specific league, you can plainly see where a player should be....this combined with performance and maybe age is how players should be evaluated IMO. To me, arrows just muddy the waters with bad info.

Another thing to consider about the arrows, is they seem to be very generic, and are not actually based on any one (or group of) coaching staff member(s) I think it would be great to be able to interact with staff when it pertains to player development instead.

Situational_Lefty · 01-13-2018, 11:14 PM

[QUOTE= i often see ~mlb ready prospects stuck in A-ball. they eventually make it to the mlb at a ~normal age from what i can tell, but they'll have 2-3 years at same lower level just hammering the competition, stats-wise. i know there's a 'slight' impact for being too good for an mil level, but obviously it doesn't hurt 10-20+ of these high-quality prospects a year. future ASG type guys.[/QUOTE]

I see the same thing in my leagues and it has been one of the only major issues I have had with OOTP. I am seeing several top prospects spending several years in rookie or A-ball without getting promoted. Then they suddenly go from A-ball to the majors at 24,25,26, when they probably should have been ready for the bigs a couple of years earlier.

I have tried multiple adjustments such as service time limits, age limits, etc, but I am still seeing it throughout the leagues I have played. I am really hoping for the next version of OOTP to changes on how the AI handles their minor league systems. It just does not feel realistic at all. Other than that, fantastic game.

NoOne · 01-14-2018, 12:23 PM

when i tradea a player like this and see t hem placed in a-ball, i often give them a 1 year mlb contract and promise a role (i make sure they replace junky player too).

i guess i could turn 100% scouting on, make a shortlist, export to external browser, end process ootp18.exe, reload with old scouting and force those players "up" in similar fashion, but i am not one to stick to that type of maintenance. it would take a longn time to find them all with inaccuracy at play.

@puscolonol, didn't mean that as an argument against them, but for this context it's important not to have it relative to learn -- not necessarily how you want to see it as you play for fun (the real reasons we play, whatever they may be).

based on delegation options the ai uses the manager's opinion. so numerous sliders could come into play... or maybe the scouting section? doubt taht last part because the coach's scouting ratings are very bad accross the board. almost 100% sure that the ratings/stats portion is importnat.

and those green/red arrows are correlated .. just not 1:1. it's not unlike how scouting inaccuracy is applied. when it's accurate it's gold, and hopefully enough callaborative information can make it an easier decision. this is a check for scouting accuracy, and ratings are a check for these arrows, and stats are a check for both etc etc.. 2/3 point in one direction you can assume the 3rd was bad info and feel safe about playing the correct odds.

PSUColonel · 05-31-2018, 02:49 PM

I'd really like to see these arrows removed. ....the entire organization has red arrows it's just not realistic to think you are going to demote so many players. Maybe just have green ones? or make this and option if you must....but all of the red arrows are bit annoying.

01-10-2018, 03:53 PM	#1
PSUColonel Hall Of Famer Join Date: Feb 2002 Posts: 13,106	Promotion, Demotion & the AI What goes into the AI's analysis of how players are either promoted, demoted, or released? I'm not sure many people know for sure....but I am convinced it's 100 percent based on relative ratings. That is all and well, but it also makes the system bit simplistic, and takes some of the player analysis out of it. If a player is a 45, he may be ready for the next level....if he's a 50 he's likely ready, and if he's more than a 50, he is definitely ready. These are relative ratings mind you. I would like to think performance would have SOME part in the equation...especially at higher MiLB levels. Any thoughts?

01-10-2018, 04:42 PM	#2
NoOne Banned Join Date: Apr 2015 Posts: 7,273 Infractions: 0/1 (3)	it may be part of the equation already. If AI evaluation includes stats, that means it's using them to some extent. When it comes to an "~18" year old in AA or AAA, it definitely matters. Anecdote: i currently have (3) 18-20year olds of high quality and seemingly ~aaa to ~aa developed. only 1 of them could put up good #'s at the level their ability deserved. i think i see better development if i avoid poor results (really bad .. ~70ops+ or worse = 30% less than average batting ability give or take). But, i've also seen someone hit that poorly and sky-rocket in development... too many moving parts to be sure. The ai will promote some very young players quickly. i'd bet in those contexts, many factors are in tune and that's what separates other examples of similar ability/development. they don't get promoted for a reason... likely stats or age or some other factor the ai uses already. i think "50", is middle of the road for that current level when on relative ratings. That's when i like to move them too, but not 100% sure if it's best either. i don't use relative ratings for minors, but the promotion timing is the ~same. the one drawback to relative is that as the talent ebbs and flows, it's difficult to see it. ie 50 in one year is inherently not the same as 50 in a different year, if it's realtive to 'current' talent in league. regardless of what is known, there is definitely an optimal way to promote players. i'd wager a significant portion of the time you'll want to be systematic about it... reach X-ability(simplified, use common sense), then promote. i wouldn't be surprised that due to how things work you need to do things differently for certain contexts... i'm curious is development is 'relative' too... e.g. if A-ball is ~50% developed, then 10current out of 20 potential is the same as 50current out of 100 potential (both 100pt scales) relative to optimal MiL level for development. so if that is true and "eye" lags compared to % development of other ratings, is it better to keep them at the appropriate level for that current eye ability (as % of development - ie how far current has to go to reach potential) until it catches up? or, maybe optimal placement is based on overall current ability and not the individual parts for this particular concept... overall, i don't think there are huge negative effects pertaining to promotion/demotion timing. the AI is terrible at it... if there were the league would always have development problems. i often see ~mlb ready prospects stuck in A-ball. they eventually make it to the mlb at a ~normal age from what i can tell, but they'll have 2-3 years at same lower level just hammering the competition, stats-wise. i know there's a 'slight' impact for being too good for an mil level, but obviously it doesn't hurt 10-20+ of these high-quality prospects a year. future ASG type guys. Last edited by NoOne; 01-10-2018 at 05:03 PM.

01-14-2018, 12:23 PM	#5
NoOne Banned Join Date: Apr 2015 Posts: 7,273 Infractions: 0/1 (3)	when i tradea a player like this and see t hem placed in a-ball, i often give them a 1 year mlb contract and promise a role (i make sure they replace junky player too). i guess i could turn 100% scouting on, make a shortlist, export to external browser, end process ootp18.exe, reload with old scouting and force those players "up" in similar fashion, but i am not one to stick to that type of maintenance. it would take a longn time to find them all with inaccuracy at play. @puscolonol, didn't mean that as an argument against them, but for this context it's important not to have it relative to learn -- not necessarily how you want to see it as you play for fun (the real reasons we play, whatever they may be). based on delegation options the ai uses the manager's opinion. so numerous sliders could come into play... or maybe the scouting section? doubt taht last part because the coach's scouting ratings are very bad accross the board. almost 100% sure that the ratings/stats portion is importnat. and those green/red arrows are correlated .. just not 1:1. it's not unlike how scouting inaccuracy is applied. when it's accurate it's gold, and hopefully enough callaborative information can make it an easier decision. this is a check for scouting accuracy, and ratings are a check for these arrows, and stats are a check for both etc etc.. 2/3 point in one direction you can assume the 3rd was bad info and feel safe about playing the correct odds. Last edited by NoOne; 01-14-2018 at 12:28 PM.

01-11-2018, 05:49 AM	#3
PSUColonel Hall Of Famer Join Date: Feb 2002 Posts: 13,106	One good thing about using relative ratings, is it illuminates the need for the awful green and red arrows. Too often the AI players see players as being between levels and can't make up it's mind about promotion or demotion. Other times, the AI just seems way off. When viewing ratings relative to a specific league, you can plainly see where a player should be....this combined with performance and maybe age is how players should be evaluated IMO. To me, arrows just muddy the waters with bad info. Another thing to consider about the arrows, is they seem to be very generic, and are not actually based on any one (or group of) coaching staff member(s) I think it would be great to be able to interact with staff when it pertains to player development instead.

01-13-2018, 11:14 PM	#4
Situational_Lefty Minors (Triple A) Join Date: Jul 2009 Posts: 265	[QUOTE= i often see ~mlb ready prospects stuck in A-ball. they eventually make it to the mlb at a ~normal age from what i can tell, but they'll have 2-3 years at same lower level just hammering the competition, stats-wise. i know there's a 'slight' impact for being too good for an mil level, but obviously it doesn't hurt 10-20+ of these high-quality prospects a year. future ASG type guys.[/QUOTE] I see the same thing in my leagues and it has been one of the only major issues I have had with OOTP. I am seeing several top prospects spending several years in rookie or A-ball without getting promoted. Then they suddenly go from A-ball to the majors at 24,25,26, when they probably should have been ready for the bigs a couple of years earlier. I have tried multiple adjustments such as service time limits, age limits, etc, but I am still seeing it throughout the leagues I have played. I am really hoping for the next version of OOTP to changes on how the AI handles their minor league systems. It just does not feel realistic at all. Other than that, fantastic game.

05-31-2018, 02:49 PM	#6
PSUColonel Hall Of Famer Join Date: Feb 2002 Posts: 13,106	I'd really like to see these arrows removed. ....the entire organization has red arrows it's just not realistic to think you are going to demote so many players. Maybe just have green ones? or make this and option if you must....but all of the red arrows are bit annoying.