Generated Player Development - Page 2

Hrycaj · 03-18-2024, 05:45 PM

I do not know if my comments here will muddy the waters but this sounds like an issue I have been having with fictional leagues. I'm on the beta team so I have been messing with this thing for months. here is my overall issue: When trying to create a fictional history universe I noticed that my leaderboards were always low when it came to things like career hits and wins. I realized quickly that the problem was that players were developing too late and then had significant drop off as they reached 40. In short players were not getting the opportunity to challenge for 3000 hits because they simply never got enough plate appearances in their career. Pitchers were also not logging enough innings to challenge 300 wins either. A player needs at minimum 17 solid seasons to challenge 10,000 Plate appearances. To get to where the historical leaderboards are players really need 20 solid seasons. That would get you into the 12,000 plate appearance range and give you a chance to put up numbers that fall in line historically. I'm pretty sure I would never get a Pete Rose outliers that played 24 seasons and logged 15,890 PA. Still, I wanted to get close so I started messing with the development modifiers. My latest test in the last build was:

Batting and pitching aging speed: .200
Batting and pitching development speed: 3.500
Dev target age: much younger
Aging target age: Much older

I feel this should have broke the game. It didn't. In a 50 year sim one player logged 12,016 PA and had 3500 hits. Another player had just over 10,000 PA and didn't challenge 3000 hits. I had 7 pitchers break 300 wins. So in conclusion, I too believe something is off.

Le_Roke · 03-18-2024, 08:02 PM

I think I've noticed GameGen prospects aren't good enough to perform at rookieball. My top prospect shortlist is just a sea of negative WAR.

This Reddit comment points out the gap between high school prospects and the average rookie baller is greater than the gap between the average rookie baller and average MLBer so I wonder if there:s something to that

https://www.reddit.com/r/OOTP/s/HUP7qiv5bJ

But I haven't read about the rating scale changes for this version.

Lukas Berger · 03-19-2024, 04:34 AM

Quote:

Originally Posted by Le_Roke

I think I've noticed GameGen prospects aren't good enough to perform at rookieball. My top prospect shortlist is just a sea of negative WAR.

This Reddit comment points out the gap between high school prospects and the average rookie baller is greater than the gap between the average rookie baller and average MLBer so I wonder if there:s something to that

https://www.reddit.com/r/OOTP/s/HUP7qiv5bJ

But I haven't read about the rating scale changes for this version.

I do think we might need to look at pulling up the current ratings for HS guys a touch.

Cartopis84 · 03-19-2024, 07:34 AM

Quote:

Originally Posted by Lukas Berger

I do think we might need to look at pulling up the current ratings for HS guys a touch.

So if the save I sent you has the same problem do you advise waiting on a fix to continue playing long term games?
Thanks

jg2022 · 03-19-2024, 08:50 AM

Quote:

Originally Posted by Cartopis84

So if the save I sent you has the same problem do you advise waiting on a fix to continue playing long term games?
Thanks

Would a temporary fix be to go into commisioner mode and manually change current ratings across the board for these prospects?

greenOak · 03-21-2024, 01:51 AM

I definitely think the low current ratings for high school prospects is a big culprit in slow development.

In my standard 2024 MLB start, I've simmed to 2048 in my game to get rid of all (or at least almost all) the initial database. After the day of the 2048 draft, the average BABIP/Avoid K/Gap/Power/Eye ratings for high school hitters taken in the first round was 124 / 97 / 115 / 106 / 94. My rookie ball league averages are 217/180/202/187/196 or 213/183/195/177/184 depending on the league. Not a single player has any rating over the rookie ball league averages, and these averages are with an age limit of 22 and 3 year maximum pro service imposed on the league.

At first I thought maybe its not so bad - maybe many of these prospects would be ready for rookie ball by the time their first full season would start. However, looking at the 2047 1st round high school hitters, the average ratings were 155/128/138/116/116 - still well below the rookie ball average. In this group only two players had a single rating above league average - a 1B with 235 BABIP and a 2B with 188 Power. The result is that it is common for good high school prospects to take 3+ years before they are finally ready to handle rookie ball. In real life, while a number of first rounders will struggle in rookie ball, many dominate from day one. The progression system would make a lot more sense if the top HS prospects already had ratings around 200. With the same amount of rating improvement, in three years top HS prospects could be challenging for AA/AAA spots instead of finally figuring out how to hit over .200 in rookie ball.

I suspect there also isn't enough of a gap in current ratings between the top high school prospects and the rest of the pack. The top prospects (going by potential rating with 100% accurate scouting) are only a bit better than their peers and that gap doesn't seem to widen much in the first 3-4 years of development.

Also, the gap in current ratings is small enough such that in the same 2048 draft, the majority of the top hitters (by current talent) have middling potential. While you occasionally see the best hitters get moved down draft boards for lacking raw tools, more often than not the 18 year olds thought to have the most potential are the 18 year olds who are currently the best at baseball.

Lukas Berger · 03-21-2024, 05:46 AM

Yeah, the low initial ratings for generated players was definitely a bug and a big part of the problem here. We've fixed this up for the next patch.

sprague · 03-21-2024, 02:12 PM

Quote:

Originally Posted by Lukas Berger

Yeah, the low initial ratings for generated players was definitely a bug and a big part of the problem here. We've fixed this up for the next patch.

Hi Lukas,
The question is the slow developement from low initial values when generated, or if the developement engine moves very slow.
Example, I used the names of players in bold as tests, gave them a rather good boost when they came in as 19 year olds.
At age 22 none of them are ready for MLB, the one who is there, is just because all the team's centerfielders were lost to free agency and none yet replaced
It is the same slow progression for pitchers
I had development up to 1.080 for batters, 1.100 for Pitchers

03-18-2024, 05:45 PM	#21
Hrycaj All Star Starter Join Date: Oct 2010 Posts: 1,618	I do not know if my comments here will muddy the waters but this sounds like an issue I have been having with fictional leagues. I'm on the beta team so I have been messing with this thing for months. here is my overall issue: When trying to create a fictional history universe I noticed that my leaderboards were always low when it came to things like career hits and wins. I realized quickly that the problem was that players were developing too late and then had significant drop off as they reached 40. In short players were not getting the opportunity to challenge for 3000 hits because they simply never got enough plate appearances in their career. Pitchers were also not logging enough innings to challenge 300 wins either. A player needs at minimum 17 solid seasons to challenge 10,000 Plate appearances. To get to where the historical leaderboards are players really need 20 solid seasons. That would get you into the 12,000 plate appearance range and give you a chance to put up numbers that fall in line historically. I'm pretty sure I would never get a Pete Rose outliers that played 24 seasons and logged 15,890 PA. Still, I wanted to get close so I started messing with the development modifiers. My latest test in the last build was: Batting and pitching aging speed: .200 Batting and pitching development speed: 3.500 Dev target age: much younger Aging target age: Much older I feel this should have broke the game. It didn't. In a 50 year sim one player logged 12,016 PA and had 3500 hits. Another player had just over 10,000 PA and didn't challenge 3000 hits. I had 7 pitchers break 300 wins. So in conclusion, I too believe something is off. __________________ Click on my signature to read about the great game of baseball in Normington.

03-21-2024, 01:51 AM	#26
greenOak Minors (Triple A) Join Date: Aug 2014 Posts: 203	I definitely think the low current ratings for high school prospects is a big culprit in slow development. In my standard 2024 MLB start, I've simmed to 2048 in my game to get rid of all (or at least almost all) the initial database. After the day of the 2048 draft, the average BABIP/Avoid K/Gap/Power/Eye ratings for high school hitters taken in the first round was 124 / 97 / 115 / 106 / 94. My rookie ball league averages are 217/180/202/187/196 or 213/183/195/177/184 depending on the league. Not a single player has any rating over the rookie ball league averages, and these averages are with an age limit of 22 and 3 year maximum pro service imposed on the league. At first I thought maybe its not so bad - maybe many of these prospects would be ready for rookie ball by the time their first full season would start. However, looking at the 2047 1st round high school hitters, the average ratings were 155/128/138/116/116 - still well below the rookie ball average. In this group only two players had a single rating above league average - a 1B with 235 BABIP and a 2B with 188 Power. The result is that it is common for good high school prospects to take 3+ years before they are finally ready to handle rookie ball. In real life, while a number of first rounders will struggle in rookie ball, many dominate from day one. The progression system would make a lot more sense if the top HS prospects already had ratings around 200. With the same amount of rating improvement, in three years top HS prospects could be challenging for AA/AAA spots instead of finally figuring out how to hit over .200 in rookie ball. I suspect there also isn't enough of a gap in current ratings between the top high school prospects and the rest of the pack. The top prospects (going by potential rating with 100% accurate scouting) are only a bit better than their peers and that gap doesn't seem to widen much in the first 3-4 years of development. Also, the gap in current ratings is small enough such that in the same 2048 draft, the majority of the top hitters (by current talent) have middling potential. While you occasionally see the best hitters get moved down draft boards for lacking raw tools, more often than not the 18 year olds thought to have the most potential are the 18 year olds who are currently the best at baseball. Last edited by greenOak; 03-21-2024 at 02:13 AM.

03-21-2024, 05:46 AM	#27
Lukas Berger OOTP Developments Join Date: Aug 2007 Location: Nice, Côte d'Azur, France Posts: 21,352	Yeah, the low initial ratings for generated players was definitely a bug and a big part of the problem here. We've fixed this up for the next patch. __________________ lukas@ootpdevelopments.com PreOrder Out of the Park Baseball 26! Need to upload files for us to check out? Instructions can be found here

03-18-2024, 08:02 PM	#22
Le_Roke Bat Boy Join Date: Mar 2024 Posts: 3	I think I've noticed GameGen prospects aren't good enough to perform at rookieball. My top prospect shortlist is just a sea of negative WAR. This Reddit comment points out the gap between high school prospects and the average rookie baller is greater than the gap between the average rookie baller and average MLBer so I wonder if there:s something to that https://www.reddit.com/r/OOTP/s/HUP7qiv5bJ But I haven't read about the rating scale changes for this version.