We know how the growth is distributed, but what factors into how much growth? Someone posted a machine learning video on Reddit.

by Han106, Nov 12, 2024

This video right here.

Key Term(Margin of growth/Progression): The difference between PA and CA. A larger Margin means more growth.

Using 30000 data points he came up with these conclusions by 4 age ranges and overall conclusions.
Under 20 Years:

Margin of progression has a huge impact at this stage
Determination needs to be at least 6, after which the impact increase is smaller
Professionalism stops mattering at 15
Even 200-300 minutes of playing time can have big impact, though minutes played matter less compared to other age groups
Training Facilities stop mattering past 17
League reputation is not a huge factor
Injuries don't have much impact

20-25 Years:

While margin is very important, playtime and division reputation are big factors
Professionalism up to 11 doesn't change growth much, but there's a big increase for each point after that
Ambition has fairly small impact
Injuries have relatively small impact
Training facilities appear to stop mattering past 16

25-30 Years:

Development naturally slows down or halts
Playing a lot and at a high level is still key
Injuries start gaining relevance
Fast increase in growth impact when determination is up to 10, slows after
Professionalism should be at least 6, after which it has relatively small influence
Training Facilities do matter all the way up to 20

Over 30 Years:

Margin has no effect on player decay, but Current Ability does (higher CA means faster decline)
Professionalism is the "undisputed king" of Hidden Attributes in stopping decay
Ambition only needs to be at least 7, after which it doesn't matter
Training Facilities do not affect player decay at all
League reputation doesn't have a big impact
Injuries have a very large impact at this stage
Players should play at least 3000 minutes to help stop decay

This breakdown shows how the importance of different factors shifts dramatically across a player's career, from development-focused in younger years to decay-prevention in later years.

I think watching the video and looking at the data even more conclusions could be found.

I do think the age ranges could have noise to them but I'm sure he has the tools to gauge that in the future.

I find this to be a great second part to the puzzle that @harvestgreen22 found. If I were to use an analogy of taxes to explain the difference between the two studies, the video talks about what effects the amount in taxes(growth) the government(player) takes in, while the harvestgreen22 study talks about how the taxes are distributed to the people(attributes).

4

Here's a link to the person's reddit post they made on the video: https://www.reddit.com/r/footballmanagergames/comments/1govdf6/using_machine_learning_to_research_player/

Seems like it does say that between 20-25, minutes played and league rep matter a lot.

Does the margin of growth thing not seem very obvious? Higher gap between CA and PA = more growth.

Still can't get over the fundamental flaw in the system where concepts like CA and PA seem to be irrelevant when a few attributes are disproportionally more important than all the others combined

0

Flashedmind said: Here's a link to the person's reddit post they made on the video: https://www.reddit.com/r/footballmanagergames/comments/1govdf6/using_machine_learning_to_research_player/

Seems like it does say that between 20-25, minutes played and league rep matter a lot.

Does the margin of growth thing not seem very obvious? Higher gap between CA and PA = more growth.

Still can't get over the fundamental flaw in the system where concepts like CA and PA seem to be irrelevant when a few attributes are disproportionally more important than all the others combined


A point I would like to make is that if it is your goal to min-max your Pace and Acceleration, I would think anything that effects amount of growth would increase the amount of Pace and Acceleration that a player gets per year. Like for example low professionalism means less Pace and Acceleration YoY and high professionalism means more. These are the other factors which can juice up the amount of whatever attribute you are looking to min-max. So this is why I believe this 2nd part to be important.

0

Han106 said: A point I would like to make is that if it is your goal to min-max your Pace and Acceleration, I would think anything that effects amount of growth would increase the amount of Pace and Acceleration that a player gets per year. Like for example low professionalism means less Pace and Acceleration YoY and high professionalism means more. These are the other factors which can juice up the amount of whatever attribute you are looking to min-max. So this is why I believe this 2nd part to be important.

Yes absolutely agree, which is why I would argue that it is the combination of professionalism and CA/PA margin that is key for younger players. But that is something we already knew, in addition to match time and to a lesser extent training facilities

0

Flashedmind said: Yes absolutely agree, which is why I would argue that it is the combination of professionalism and CA/PA margin that is key for younger players. But that is something we already knew, in addition to match time and to a lesser extent training facilities

Yes but there is nuance in the data like how Professionalism early might have diminishing returns past 15. How good Training Facilities need to be. One thing I learned was the extent Reputation of Division matters.

0

Han106 said: This video right here.

Key Term(Margin of growth/Progression): The difference between PA and CA. A larger Margin means more growth.

Using 30000 data points he came up with these conclusions by 4 age ranges and overall conclusions.
Under 20 Years:

Margin of progression has a huge impact at this stage
Determination needs to be at least 6, after which the impact increase is smaller
Professionalism stops mattering at 15
Even 200-300 minutes of playing time can have big impact, though minutes played matter less compared to other age groups
Training Facilities stop mattering past 17
League reputation is not a huge factor
Injuries don't have much impact

20-25 Years:

While margin is very important, playtime and division reputation are big factors
Professionalism up to 11 doesn't change growth much, but there's a big increase for each point after that
Ambition has fairly small impact
Injuries have relatively small impact
Training facilities appear to stop mattering past 16

25-30 Years:

Development naturally slows down or halts
Playing a lot and at a high level is still key
Injuries start gaining relevance
Fast increase in growth impact when determination is up to 10, slows after
Professionalism should be at least 6, after which it has relatively small influence
Training Facilities do matter all the way up to 20

Over 30 Years:

Margin has no effect on player decay, but Current Ability does (higher CA means faster decline)
Professionalism is the "undisputed king" of Hidden Attributes in stopping decay
Ambition only needs to be at least 7, after which it doesn't matter
Training Facilities do not affect player decay at all
League reputation doesn't have a big impact
Injuries have a very large impact at this stage
Players should play at least 3000 minutes to help stop decay

This breakdown shows how the importance of different factors shifts dramatically across a player's career, from development-focused in younger years to decay-prevention in later years.

I think watching the video and looking at the data even more conclusions could be found.

I do think the age ranges could have noise to them but I'm sure he has the tools to gauge that in the future.

I find this to be a great second part to the puzzle that @harvestgreen22 found. If I were to use an analogy of taxes to explain the difference between the two studies, the video talks about what effects the amount in taxes(growth) the government(player) takes in, while the harvestgreen22 study talks about how the taxes are distributed to the people(attributes).


thank you for this, keep up the great work

0

It's a shame the age divisions seem a bit off - presumably 15, 16 and 17 year olds have different requirements, then you have a mid range where the factors should be the same but growth gets less as you get older and then ~30 you go into the decline where a 3rd lot of factors kick in

0
Create an account or log in to leave a comment