AI releasing pitching prospects

[downwithootp]

I’ve tested this in normal starts and in live starts, teams will release a few of their 50 potential pitching prospects randomly. It’s even worse after the draft because then you can come away with 4-5 pitchers that were just drafted in 2nd and 3rd round.

Is this a bug other people are experiencing? Is this being addressed in new patch?

[jpeters1734]

Online leagues will always be king due to the never-ending game of AI whackamole.

[1991Cubs]

Originally Posted by downwithootp

I’ve tested this in normal starts and in live starts, teams will release a few of their 50 potential pitching prospects randomly. It’s even worse after the draft because then you can come away with 4-5 pitchers that were just drafted in 2nd and 3rd round.

Is this a bug other people are experiencing? Is this being addressed in new patch?

Yes. Seeing some decent pitching prospects drafted then released almost immediately

[phenom]

Tinstaapp

[willboat75]

This is something that has distorted immersion in my most recent MLB save. Two pitchers who were both in the top 10 overall prospects at the time of their releases signed with different teams and stayed in the top ten. I signed a guy who my scout rated as a 45 potential but was released after being a relatively high pick, but now he has inflated my farm ranking due to his newfound top 20 pedigree.

The only thing I can think of that would explain this is roster limits in the minors. I use the default service time requirements and the real-life roster limits of 28 for AAA/AA, 30 for A+/A and I even bump it from 35 to 40 for the rookie leagues. Maybe I only cap the high minors?

[Paul Reuschel's Mustache]

Originally Posted by willboat75

This is something that has distorted immersion in my most recent MLB save. Two pitchers who were both in the top 10 overall prospects at the time of their releases signed with different teams and stayed in the top ten. I signed a guy who my scout rated as a 45 potential but was released after being a relatively high pick, but now he has inflated my farm ranking due to his newfound top 20 pedigree.

The only thing I can think of that would explain this is roster limits in the minors. I use the default service time requirements and the real-life roster limits of 28 for AAA/AA, 30 for A+/A and I even bump it from 35 to 40 for the rookie leagues. Maybe I only cap the high minors?

Hopefully, this won't be an issue anymore with the new patch.

Originally Posted by Matt Arnold

This patch, 26.6.73, focuses heavily on player development and roster AI, cleaning up some items and giving a more balanced long term sim.

Changelist 26.6.73

• Player Development
  **Adjusted BABIP development
  **Adjusted aging speed ratings
  **Adjusted SP development (slower to start, develops later)
  **Pitchers slower to develop, but will continue to develop later (avoids peaking at 25 and then falling off)
  **Improved College player development, increased starting ratings
  **General balance of development/aging
  **Balance pass through general player creation and update
  **Adjusted new standard game development
• Roster Updates
  **All players from the 2025 draft added and rated
  **Most 2025 UDFA signings added and rated
  **Lots of new DSL players added and rated
  **Many complex level players ratings adjusted
  **Majority of players at the DSL and complex level who are actually prospects have been rated to properly reflect their prospect status
  **Lots of other prospect ratings adjusted
  **Quite a few MLB player ratings adjusted based on 2025 performance
• Improved minor league roster management
  
• Updated minor league roster logic to release decent prospects less
  
• Hide coach retire/delete if hide advanced options selected
• Fixed international amateur free agent period handling for custom dates
• Fixed coloring in lists with overall ratings disabled
• Live Start: Fixed some roster issues
• Live Start: Improved injury handling, injured players on rosters
• Adjusted historical stamina calculations for swing SP
• Adjusted RP contract demands
• PT: Hide the clock in tournaments/drafts when score is hidden
• PT: Fixed pack opening settings not updating
• PT: Fixed clubhouse card list variant relations showing wrong points
• PT: Fixed slow sort/filter release timestamp column
• PT: Fixed Fastspring key redemption not updating packs
• PT: Changed universe screen to start in the middle
• PT: Fixed variants display in choice packs and drafts
• Misc roster updated, misc bug fixes, UI fixes

[BarneyRubble]

I also believe it's roster limits. I don't believe the ' AI' does a good job discerning between filler players and prospects. The minors is filled with filler players. They are crucial to success, but they get released before a real prospect.

If I have a filler player 45/45 in AAA, he is there for backup and the general health of the team. When a prospect 40/60 comes along, the filler player goes first if I HAVE to get rid of one of them. Usually, it's another bench guy I get rid of. I don't think the AI gets it sometimes. If I see it coming, I'll try to trade the filler player for younger Milb players. The AI seems to understand his value during trades, but not so much with roster management. It's tough for OOTP since they have allowed us to configure the game anyway we see fit.

Maybe the patch fixes it.

[NoOne]

Absolutely could be part of the problem. if what you see doesn't happen without the rules in place, obviously that is the cause.

the defaults are no rules for age or svc time or roster limits in MiL, so you added or changed those.

Strongly recommend removing roster limits from mil teams. if you've done svc time rules i'd revert those too and rethink them. age limits can be used to avoid pooling in a reactive way. roster limits in rookie leagues are absolutely a horrible idea. adjust new players added each year to fix that type of problem.

Anytime you do something that changes number of new players made per year (draft, iafa etc) or rules regarding how they traverse the mil system, you better do some due dilligence. Use a throwaway league to test the new rules to make sure pooling or weird ai transactions don't occur. you'll need to zoom out ~10 years to see the full effect of changes to roster rules in mil top-to-bottom. if onlylooking at rookie league, obviously only need enough years to cover what your rules cause in regard to turnover.

When it comes to mil roster rules, less is more. Don't try to force a square peg through a round hole. There are some things you can influence, but if you try to force it to do something it simply can't, bad things will happen.

I've played around with mil service time and age limits in minors a lot. The only things that work are the very basic control ovr them.

e.g. i tried to make one rookie league the 'younger' one with age rules while a coinciding rookie league should have been receptive to the rest of the younger/newer players added to the system, but it simply doesn't work... so i reverted it. no need trying to force it to do something it cannot do. There were more than enough of each type of player that it should have worked, but it didn't because of the way the AI works.

That's not a criticism of the ai. Very possible that if it was configured in a way that would allow such a thing to work it could have 2x as many problems otherwise, so i'm not assuming anything in that regard.

What i found does work is some minimal upper-age restrictions and don't mess with AAA much.. maybe a liberal max roster limit, but otherwise it wills screw things up.

Don't mess with rookie league roster limits. just gonna screw things up, lol.

There's too much of a gap between hs and college ages that age limits early on screws things up, lol... common theme.

if you are getting too many new players, slim down rounds in draft or iafa or discoveries etc...

if you do use rules, make soem common sense extrapolations based on how many enter the league per year, how many leagues at each level you have andwhether that can actually properly fill it up or not.

a rule to mitigate pooling can be useful. this is a reactive thing and not a proactive behaviour.. if you see a ton pooling in one area, look at why it's happening... is it age-related? is it because too many are pumped into the system to start? choose how to react based on why/how of it happening.

Less is more when it comes to MIL age/svc time and roster limits.

i'll do somethings to ensure older players don't linger but trying to tightly control it or shape it in a specific way will not work. get that out of your head to avoid the frustration.

there are so many levers and dials as to how you can set up a league, it shouldn't take anything away from the game. somethings you simply cannot makework the way you want to. you can do that in so many other ways, it shouldn't matter too much.

There is what you want and what the game can do... you'll have a lott less consternation stickign to what is possible.

Originally Posted by willboat75

This is something that has distorted immersion in my most recent MLB save. Two pitchers who were both in the top 10 overall prospects at the time of their releases signed with different teams and stayed in the top ten. I signed a guy who my scout rated as a 45 potential but was released after being a relatively high pick, but now he has inflated my farm ranking due to his newfound top 20 pedigree.

The only thing I can think of that would explain this is roster limits in the minors. I use the default service time requirements and the real-life roster limits of 28 for AAA/AA, 30 for A+/A and I even bump it from 35 to 40 for the rookie leagues. Maybe I only cap the high minors?

[OutS|der]

Originally Posted by NoOne

Absolutely could be part of the problem. if what you see doesn't happen without the rules in place, obviously that is the cause.

the defaults are no rules for age or svc time or roster limits in MiL, so you added or changed those.

Strongly recommend removing roster limits from mil teams. if you've done svc time rules i'd revert those too and rethink them. age limits can be used to avoid pooling in a reactive way. roster limits in rookie leagues are absolutely a horrible idea. adjust new players added each year to fix that type of problem.

Anytime you do something that changes number of new players made per year (draft, iafa etc) or rules regarding how they traverse the mil system, you better do some due dilligence. Use a throwaway league to test the new rules to make sure pooling or weird ai transactions don't occur. you'll need to zoom out ~10 years to see the full effect of changes to roster rules in mil top-to-bottom. if onlylooking at rookie league, obviously only need enough years to cover what your rules cause in regard to turnover.

When it comes to mil roster rules, less is more. Don't try to force a square peg through a round hole. There are some things you can influence, but if you try to force it to do something it simply can't, bad things will happen.

I've played around with mil service time and age limits in minors a lot. The only things that work are the very basic control ovr them.

e.g. i tried to make one rookie league the 'younger' one with age rules while a coinciding rookie league should have been receptive to the rest of the younger/newer players added to the system, but it simply doesn't work... so i reverted it. no need trying to force it to do something it cannot do. There were more than enough of each type of player that it should have worked, but it didn't because of the way the AI works.

That's not a criticism of the ai. Very possible that if it was configured in a way that would allow such a thing to work it could have 2x as many problems otherwise, so i'm not assuming anything in that regard.

What i found does work is some minimal upper-age restrictions and don't mess with AAA much.. maybe a liberal max roster limit, but otherwise it wills screw things up.

Don't mess with rookie league roster limits. just gonna screw things up, lol.

There's too much of a gap between hs and college ages that age limits early on screws things up, lol... common theme.

if you are getting too many new players, slim down rounds in draft or iafa or discoveries etc...

if you do use rules, make soem common sense extrapolations based on how many enter the league per year, how many leagues at each level you have andwhether that can actually properly fill it up or not.

a rule to mitigate pooling can be useful. this is a reactive thing and not a proactive behaviour.. if you see a ton pooling in one area, look at why it's happening... is it age-related? is it because too many are pumped into the system to start? choose how to react based on why/how of it happening.

Less is more when it comes to MIL age/svc time and roster limits.

i'll do somethings to ensure older players don't linger but trying to tightly control it or shape it in a specific way will not work. get that out of your head to avoid the frustration.

there are so many levers and dials as to how you can set up a league, it shouldn't take anything away from the game. somethings you simply cannot makework the way you want to. you can do that in so many other ways, it shouldn't matter too much.

There is what you want and what the game can do... you'll have a lott less consternation stickign to what is possible.

So basically give up on ever recreating the rules of the MILB cause the AI can't handle it, never has and likely never will.

[jpeters1734]

Originally Posted by NoOne

the defaults are no rules for age or svc time or roster limits in MiL, so you added or changed those.

tl;dr, but I did notice this is incorrect. The standard MLB and fictional games have pro years limits below AA and standard fictional games have roster size limits for all levels, except Rookie.

You are correct that the AI sucks at meeting human expectations on who is releasable and it's best to never set limits for rosters

[BarneyRubble]

Keep in mind, there is a long list of star players that were released once. Ortiz, Bautista. etc... It does happen. I once picked up a prospect that turned out bust.

[PSUColonel]

I’d say you can likely do this:

No age limits.

Service limits: 28 AAA/AA
30 A+/A
44: R
35: DSL