Anyone done any testing on prospect development in v25.2.52?
 

I know the patch just came out today but wondering if anyone has done any testing on this.

Buehler?

So far it's still a disaster.


I'm in game-year 2028, and literally no generated prospect has made it to a rating over 40/65 -- and that guy was a highly developed college SP.

Preface
I've been doing some extensive testing on the league ratings the last few days. It seems that development has certainly became better, but I suspect that it could be improved further or something about the development in the minors could be tweaked (EDIT: Since I have split into multiple posts with the others coming later, this tweak does not seem relevant given just the age analysis). Although a disclaimer I would need to run more leagues and also try to spend more time understanding the data.

Testing Method
For the data I have collected, I ran a 20 year simulation on a default settings league (but remove league evolution). For each year in the 20 years, I generate a league roster export Opening Day. I then have some Python scripting that can analyze those exports, generate statistical analysis data on them per league and per position group, and can generate some basic graphs.

Notice to Emphasize
A general preface: More testing is needed to confirm these general trends. I have only ran a 20 year sim on a league to test this so far, and multiple sims of more years would be nice. Having a yearly export on Opening Day function would immensely help in collecting the data as I could just sim X years without having to stop each year to export data.

Age
* Note - All Ages are calculated from July 1st of that year as that is the method MLB uses.

MLB
Average Age
The first thing I was looking into is the league age over time. For MLB, the average age for position players seemed relatively flat, but with a slight bump. The average pitcher tend to get older the first 6 years and then get a few years younger. I suspect the reason they get older at first is the lack of quality pitching prospects in the draft pools based off real players. This could partially be due to bad real life draft classes, but it also could be some undervaluing of pitchers in the few draft classes. Once the generated players start maturing, the average pitcher age drops down a good amount (my hypothesis for this will be explained later in rating analysis).

Outlier Ages
The game seems to have a very slight tone down on the outlier ages. The youngest player in the league tends to be 1 to 1.5 years older than in real life. Real life trends have the league getting younger and younger and superstar potential players debuting earlier on average, but I don't think the game does a good job of modelling those outlier development cases.On the inverse side of old players, the same general trend occurs where the oldest player in the league is a little younger. Again this is likely the game not handling outliers like Justin Verlander as well and letting players age so well.

The outlier ages I suspect can be fixed by modifying the player development and aging speeds. I realize the developers have to make defaults that generally fit everywhere. My ultimate goal is to get more testing in with various setting changes to get a league that matches reality of the modern MLB.

Minors
Note
I first want to state that I am unsure how OOTP assigns players to minor league rosters as the average age of the initial rosters for a given level are much older than reality. Here is a good article from Baseball America showing the minor league ages per level.

League Age Analysis
As noted above, the minor league level ages are simply much older than they should be when getting to Opening Day 2024. A sneak peak for a future post/reply, but the average talent in the minors is higher than probably expected at first, which is probably due to the players being older. However, OOTP's simulation does appear to do a great job of converging over time to the real numbers, so kudos to the devs on getting real life data and making the simulation engine match for the ages.

See image below for table of ages.

The above average age can likely be fixed very easily by setting limits for pro experience per minor league level like the real minors have. I always set these when playing a league, so I'll do testing later to try it out and see the effects on minors aging.

Coming Soon
I planned on putting more data about my analysis of the rating development over time and the potentials into this post, but this post is taking a while to write out. There should be more overly detailed content coming later.

Bummer, appreciate the reply!

I'm playing in fictional leagues. In saves started since the patch, results seem a touch better than during beta and initial launch, where league wide almost no drafted players exceeded 50 overall during their peaks, but some of the original concerns persists. Prospects still seem to plateau and flatten both in overall ratings and performance compared to past experiences--I've been playing for 15+ years. I've seen some posters and developers focus on whether or not extremely young players are developing quickly and succeeding, but I really think the issues are about long-term and peak performance.

Position Player Development
Preface
Continuing my previous post, I have set up a league and ran 20 years of sims and done a yearly export on Opening Day of each year. I then parse these exports with Python to generate some basic data analysis and charts.

For position player development, I will only show a graph of vL or vR randomly as I have a 5 image limit, and they are essentially the same for vL and vR anyway. I will be excluding Contact as it is a combination of BABIP and Avoid K's. I also will only be focusing on the primary offensive stats, although I do have a wide range of data locally on defensive stats per position group breakdown and even personality stats.

A note from one of the OOTP 25 videos, but the general idea is that MLB is going to be a 400 average, with AAA being 350, AA being 300, and so on in 50 decrements. However, please note that is just a general statement, and there is nothing that means that has to be the letter of the law.

BABIP and Avoid K's
BABIP and Avoid K's show a similar general trend over time. For both, the MLB average player sticks right around where it started in the base game with a quite flat line. This means that each has the league steadily keeping up with the talent level of the default game. However, there are some interesting things to note.

The league actually starts out with the top levels being quite clustered together and HIGH-A to MLB only having 60ish difference on the average players.In general, the talent levels seem to be higher in the initial levels, although this may be an overestimation of the base skills of minor league players initially. After the first year, the talent level starts dropping off with more noticeable drops in the lower levels first and then propagating up each level year by year (due to promotions). This partially could be poor draft classes, but once the generated players kick in with their sky high potentials, the trend stays the same before flattening out for each rating.

Once the ratings do flatten off after a few years on the sims, there are interesting observations to be made from the data. The gap between MLB and AAA is only about 25, but the gap between AAA and AA is over 100. The ratings do end up closer to the theoretical 50 gap difference once they stabilize beyond AA (roughly). The average Rookie player ends up actually becoming a little better over time and I expect this is specifically related to some Patch notes this patch about giving HS players better ratings.

I noted in the Age analysis that the average age of players in each minor level was a good amount higher on a default save than the real minors. I expect that this higher age is the explanation for why the LOW A and above minors seem to be higher than the theoretical 50 diff per level. Testing an experience limit will be helpful to see if it addresses this.

Eye and GAP
In both of these cases, MLB gets a little better as the real life prospects mature, and then actually decreases a good amount below the starting level after the draft classes start to mature. I think this initial increase is explained by the over talent classification of many upper level minors player similar BABIP and Avoid K's.

Once the generated players start graduating, the average MLB rating goes down. I thought this may be due to lower potentials on these stats, but the league potential actually goes up for both Eye and GAP. This seems to indicate the problem is actually in the development of these statistics and they do not fully develop over time on generated players. This likely needs a slight tweak to bump up the development rates on these stats. Maybe a bump on the overall Position Player Development could help this, but potentially at a cost of over growing other ratings.

Another thing to note that is seen on 4 of the 5 major ratings, the DSL players have a higher overall rating than the Rookie league players in the initial player pool. Maybe age limits help here, but this is a little odd as DSL players may spend time in Rookie to up the challenge.

Power
Power seems to show the best behavior on offensive stats on keeping a stabilized league and is closer to the 50 diff per level. However, there is about a league diff (50) hit on each of the mid minors levels once the generated players start moving up which may indicated a slower growth rate on power than the starting game in a player's career.

General Remark
I would be curious to look more into the AA to AAA gap. It seems players may get stuck in AA for a while simply due to not growing fast enough until they get to AA and having to make such a jump to AAA.

Offensive Potentials
 

Potentials Preface
The main reason I started all of my testing. Potential. I start losing touch with the game 5+ years in as the generated players start coming in via the draft along with a few years of IAFA. The generated players simply all make it look like some development in player teaching has occurred and you can take any regular joe and they could potentially turn into an MLB player. My ultimate goal of my scripting is to arrive at Player Creation Modifiers that reduce the potential of players so I no longer has Top 100 prospect lists with all being 65+ POT (100% accuracy even) while retaining a consistent MLB level.


Observations
The overall potential of the league for each major offensive stat increases over time as the generated players hit. As seen in the previous post, the actual MLB talent level doesn't really see much of an increase though. This seems to indicate that despite all players having potentials, most only live up to being league average. I think this may explain some people's complaints that there prospect never fully develops, although it is a good thing in my estimation as you don't want the talent level rising too high.

The minors show an interesting trend with some ratings showing it more than others, but the homogenization of potentials among the levels. I think this is due to the game simply having small draft pools and every prospect being having the potential to be a AAA level player at some point. However, that isn't the case in reality. While every minors player wants to have a shot at the bigs, there are plenty of them that don't even have the potential be much more than minors depth a the A or AA ball levels and even lower for the IAFA's.

As stated in the preface, I think most of this can be addressed via Player Creation Modifiers, so I will do more testing over time to find the settings that lead to a more stable league.

Pitchers Control and Movement
 

Mostly going to post images with some quick notes as I could spend all day talking about this :)

- Control drops over time a good amount. Probably a combo of lower generated levels along with slower than desired development.
- Movement is VERY clustered toward the top in an the opening minors
- Movement drops over time in the majors. Given that potential actually increases, this is probably due to slower than desired development, although it does stabilize closer to 400 which is likely good
- Super sharp decline in Movement over the years, although part is due to way too high initial ratings.

Pitcher Stuff
 

- Stuff level drops early in which may be due to to low of Stuff on real life draft classes (although they are actually weaker)
- Stuff ultimately ends up a little below 400
- Minors stuff looks quite good actually for diffs between levels and would allow a nice development pattern
- Stuff potential drops initially (bad real life draft class pools) and then jumps up a LOT when generated players start hitting MLB. This means most never live up to it

Included a couple of random graphs as well to bring up and share
- Arm slot tends to go up in the league. Could be generation of these is higher than it should be or a higher chance of a higher arm slot leading to success
- Wild pitches really drop off on the generated players. Likely a too low creation rate
- Work ethic generally rises. Likely numerous explanations like too few IRL high WE, too many generated WE, or higher success rate of higher WE

The game has been in beta testing since October, so....

As a general rule, the base talent levels for each skillset really should remain pretty constant year-over-year. So what these charts seem to be telling is mostly a good news story--though (as it seems has always been the case) the first decade of a new game can see wild variances as initially generated players get over-taken by the game's generated players.

Of more interest, I think, in the world of player development as I think about it is a chain of player ratings over their ages. Where they start, when they peak, and how often they fall off. Then as an effect of that, how the players ratings convert into statistical output...which is really more about pitcher/hitter interactions and league totals.

Guys, I'm tired

I'm tired of having supported this series for the last 8 years. I'm staring at my steam inventory....I've got OOTP 25 thru 18 in it. I'm real sick and tired of the annual major issues that plague this series. Whether it's the long term development issue of this year, or financials of OOTP 24, or the fact we had problems with OSA scouting like some 5 years back. I make fun of people who only have positive things to say, not because I hate positivity, but I hate liars. I have a problem with people who want to ball-wash OOTP and act like its perfect. It's not. I don't expect the world, I just expect that as a middle-aged man with a couple kids trying to pay the bills to make his $50 game work properly. I'm tired of people who act like it should be expected that the $50 game I just bought can't be played for a few months while they work the bugs out.

Just...trying to be real with you. I'm not some young punk with no patience....I'm genuinely distrurbed both by how this company is run and by how compliant many of you are with the ineptitude. You ruin it for everyone else.

Football Manager 2024 is put out by Sports Interactive, a UK-based studio with a team of 279 and around 1,300 researchers (according to their website). They put out the world's # 1 football (soccer) management simulation game.

Take a look at the FM24 forum and you'll quickly find comments such as this one:

"I already opened a bug report for this months ago. Have two saves that are affected (one a 23 port, another a fresh 24 save). Still tagged as "under review". I won't say its a deal breaker for playing but it does kind of ruin the immersion a bit when you're just going 1,2,3 through the league along with everyone else."

Or take a look at the Bugs subforum and realize that FM24 was put out on November 6, 2023!

I'm not sure how many people are actively working on OOTP25 compared to FM24, but I'd bet that the OOTP team is a fraction of what Sports Interactive allocates toward Football Manager. Despite having a lot more people working on the game, FM seems to have the same annual major issues that plague this series, as you complained about for OOTP.

Just sayin'. ;)

It's easy to claim ineptitude when you have no understanding of how these games are created. I get being upset with the game's state, but if you're upset with it then stop purchasing it yearly as it comes out. No one's claiming the game is perfect, they're just being more realistic than you in understanding that new systems in a game this complex aren't going to magically be implemented with no issues.



People like you ruin it for everyone else because your wanting to drag down other people's belief in the game when you should probably just move along.


Maybe start with keeping a discussion like this more on topic rather than using it as your personal opinion column.

Just because FM is broken doesn't mean it's okay for your game to be broken too. That's not a good argument.

Game isn't broke. The development issue is mostly people not being happy the displayed ratings are the right colors.

Saying all games have bugs is a valid argument and that if FM which is 5x if not 10x the size can have bugs for this long then do you really expect this team to have no bugs?

I think the main issue here now is that people are expecting development to work in an unrealistic way that it's not designed to do.

Most prospects don't develop to their peak potential. Very few prospects will ever develop to be over 50 ratings no matter what their peak potentials are. Many real-life draft classes don't have more than 15 or so above average MLB players once all is said and done, whatever their potential was viewed at the draft. Just because a player has an 80 potential for example, it does not mean he's expected to turn into an 80 overall rated player at his peak. Some may. Most will not. It's going to be a rare player to develop to a 55 or 60 or above rating. If you have an 80 rated prospect and he turns into a 55 or 60 player at his peak, this is a pretty good result.

This is even more the case this year than in previous years, because we've adjusted the overalls to incorporate standard deviations a bit more and thus they are a little tighter and they end up tracking a little closer to the real-life scale now (they are still not identical, so you will have more 70 and 80 players in OOTP than there would be in real-life f.e.). So more players will end up fairly tightly clustered in the 40-60 range once fully developed and it will be somewhat rarer to see a player to hit 60-80 than in previous version of the game. Thus a player whose ratings would have given him a 65 overall rating a couple versions back might have more of a 55 or a 60 now, even if the underlying ratings are essentially the same.

The original concern with development was more just that players were taking too long to develop and there were very few young players well developed. The patch does greatly improve that. What the patch does not do is to make all young players develop to max potential. That's not how it works in reality, and it's not how the game is supposed to work. The league talent levels will stay roughly equal over time (though they drop a bit from the initial roster set ratings, which is to be expected as these are slightly higher than the long term targets), which pretty clearly show that enough players are developing overall.

There are some minor things with development we're still planning to work on, including creating more outliers on the default settings. So basically creating more players who are very good very young (19-21), more players who stay very good when old (late 30's to early 40's) etc.