Season 1 v Season 20 - Results of a Regen Test

So over the holidays, I ran a holiday game to have a look at the regens in FM09.

I set the game up with English, Spanish, German, Italian, Dutch and French leagues active, with a large database, and ran through to season 20. At season 1 and season 20, I used the latest version of Genie Scout to export every player in the database to Excel, with hidden and non-hidden attributes selected. At season 1, there was a total of 25,857 players; at season 20, there was a total of 28,739 players.

The Excel file with all the data in can be found below (player names have been removed). Feel free to do your own analysis!

MAIN FINDINGS:

Current Ability

Season 1

1-50: 22.0%

51-100: 50.4%

101-150: 26.0%

151-200: 1.5%

Total: 100.0%

Season 20

1-50: 18.2%

51-100: 48.3%

101-150: 31.2%

151-200: 2.3%

Total: 100.0%

So the difference in the spread of Current Ability isn't too bad, but the proportion of players with Current Ability over 100 definitely goes up by season 20.

Current Ability Of Players Aged 18 Or Under

Season 1

1-50: 50.0%

51-75: 36.9%

76-100: 10.2%

101-150: 2.8%

Total: 100.0%

Season 20

1-50: 41.1%

51-75: 36.6%

76-100: 17.9%

101-150: 4.3%

Total: 100.0%

I think the spread of Current Ability among younger players is slightly more of a problem -- by season 20, there's a much higher percentage of youngsters with Current Ability over 75, and the same with youngsters with Current Ability over 100. My feeling is that regens are starting with their Current Ability slightly too high, which is why some of them have such impressive attributes and many of them are representing their nation at Under-21 level at a very young age.

Weaker Foot Ability

Each player has a rating out of 20 for their 'right foot' and 'left foot' ability. Every player has 20 for one foot, but scores vary for the second, 'weaker' foot.

Season 1

0-5: 47.1%

6-10: 34.4%

11-15: 16.9%

16-20: 1.6%

Season 20

0-5: 72.6%

6-10: 14.5%

11-15: 11.4%

16-20: 1.6%

This is the main problem, for me, and it's one that's been talked about on the forum quite a bit. You can see that by season 20, the vast majority of players have a weaker foot ability of 5 or less, compared to less than half of players in season 1.

Number Of Positions

Each player has a rating out of 20 for each position. When players have 15 or over for a position, they're rated as 'accomplished' for that position. Below is a comparison of how many 'accomplished' or 'natural' positions players have in season 1 and season 20:

Season 1

1: 50.7%

2: 28.3%

3: 10.3%

4+: 10.7%

Season 20

1: 52.8%

2: 29.5%

3: 11.3%

4+: 6.3%

Not too much of a problem here - players in season 20 were slightly more likely to have just one or two accomplished/natural positions, but the difference isn't too big.

Attributes

I don’t really have time to post up an in-depth analysis of every attribute, but here’s a summary of the attributes where there’s a particularly big difference. In general, most attributes were higher in season 20. A few stats did go down, on average - flair, off the ball, loyalty and decisions, for instance, but the differences weren't as big as those attributes that went up.

Outfield players

In order of difference, the attributes that are noticeably higher in season 20:

Important Matches

Versatility

Determination

Strength

Acceleration

First Touch

Long Throws

Agility

Goalkeepers

In order of difference, the attributes that are noticeably higher in season 20:

Kicking

Reflexes

Throwing

--------

I did most of the tables and analysis in SPSS, but the csv file (can be opened in Excel) with the original data is here: YouSendIt link

Excellent work. The weaker foot issue really jumps out.

Good post.

I'm gonna play devils advocate and try and use a real life explanation to maybe explain your findings. If you look back 20 years ago at the talent in Footballers as well as talent in most major sports, players have typically gotten bigger, stronger, faster, and generally develop a greater level of talent, at an earlier and earlier age. New training techniques, National development, improved diet and supplements, higher youth involvement, etc. all could potentially explain your findings, such as more players and a higher skill level. A more in depth analysis of the superstars, might show if some of the superstars have improved at an equal ratio to the more average talent. If the percent of players in a 150-175, as well as 176-200 bracket go up at an equal ratio, then the world class players are improving at close to an equal rate as some of the lesser talent. One may still find this a balanced talent pool.

I can't explain a weaker foot theory, maybe youngsters now a days and the future superstars are just lazy and stick to what works the best.

One strange thing is that 'versatility' is higher, but #positions is lower, as is 'weaker foot'. So that versatility isn't being capitalized on by the AI managers. Not tooo surprising, but still...

Excellent work. Maybe something to look at for 9.3 by the SI guys.

This would be a more solid experiement if you re-did it several times.

Good post.

I'm gonna play devils advocate and try and use a real life explanation to maybe explain your findings. If you look back 20 years ago at the talent in Footballers as well as talent in most major sports, players have typically gotten bigger, stronger, faster, and generally develop a greater level of talent, at an earlier and earlier age. New training techniques, National development, improved diet and supplements, higher youth involvement, etc. all could potentially explain your findings, such as more players and a higher skill level. A more in depth analysis of the superstars, might show if some of the superstars have improved at an equal ratio to the more average talent. If the percent of players in a 150-175, as well as 176-200 bracket go up at an equal ratio, then the world class players are improving at close to an equal rate as some of the lesser talent. One may still find this a balanced talent pool.

I can't explain a weaker foot theory, maybe youngsters now a days and the future superstars are just lazy and stick to what works the best.

It would seem pretty unlikely that this would be deliberately programmed in. As I understand it, the idea is to have the game world overall in line with how it is now. You might be right, but it's unlikely to be an intended consequence of the regen system.

To me it looks more like the issue noted in FM08 - where regens developed too slowly - has been fixed, but slightly over compensated for, and as such there are more talented youngsters than at the start.

There was a thread in the CSE forum doing a similar experiment on FM08, and as I recall the decline in standard after a number of seasons was far sharper than the increase this time - looks to me like a very decent job has been done of fixing it.

Overall it looks okay to me, or at least a lot better than regens in previous versions but the weaker foot issue could do with a tweak as we all know. Mind you, Morten Gamst Pedersen was very right footed until aged 18, until his dad forced him to practice with his left foot, and now his left foot is awesome. That should be an option in the game to make players work on their weaker foot.

This would be a more solid experiement if you re-did it several times.

Yeah, you're right. It's just that I don't really have the time to keep re-doing it, unfortunately!

If other forum users were able to do something similar and put their results in Excel, I'd be happy to do the analysis for them. It's just a case of saving your game at the beginning of season 1 and season 20, opening the two files with Genie Scout and exporting it to Excel.

One possible explanation of the increased abiliity as time goes on is the active leagues selected. I had the top divisions from the major European leagues selected, so it may be that fewer players from low reputation teams are loaded, as time goes by.

For me, the main problem is the weaker foot abillity, which is something I suspected before I ran the test. The big increases in 'important matches', 'versatility' and 'determination' (which are attributes that I don't think are linked to Current Ability) are worth looking at, as well.

EXCELLENT. now SI please fix the weakfoot error thing!

Nice work RT.

I think Bermy has covered my thoughts nicely. It seems like there has been a slight rise in the number of high PA players, but compared to how much the quality of regens improved in FM08 it's a vast improvement. I'd rather see slightly too many good players in a few years than no decent players under the age of 23 as it used to be.

The either footed issue seems to be the most obvious, but do players not improve their weaker foot when being trained to play on the opposite side of the pitch than they're used to? If this is the case then this may be linked to the fact that the AI managers do not seem to be re-training players to play in multiple positions very often.

Nice work - very interesting!

Yeah, this is interesting. Did you manage to do any checking of stats between years 1 and 20? So, would you have any idea of whether these changes occur gradually over 20 years or whether they fluctuate all over the place in between?

Yeah, this is interesting. Did you manage to do any checking of stats between years 1 and 20? So, would you have any idea of whether these changes occur gradually over 20 years or whether they fluctuate all over the place in between?

That is certainly a good idea - I do not know if you can save the games at the end of every season automatically but it would allow you to plot more accurate progressions of these statistics.

Original post was excellent - I like it.

out of curiosity, how long did it take to do twenty seasons?

There definitly seems to be a big improvement on the amount/quality of regens though from previous versions. Big step in the right direction SI

Yeah, this is interesting. Did you manage to do any checking of stats between years 1 and 20? So, would you have any idea of whether these changes occur gradually over 20 years or whether they fluctuate all over the place in between?

No, I didn't do this - again, it was just a matter of time constraints, but if anyone else has the time, it would be a great idea! You can auto-save 'every year' but I think you could only do it in batches of three, before it started over-writing older saves.

out of curiosity, how long did it take to do twenty seasons?

I think I left it running overnight, for 3 or 4 nights in a row!

Not bad for regens