@harvestgreen22

It looks like doing just 'Rest' for 2 years and then 'Full' training is best when comparing your data. How many test repeats did you do?

Also since it is clear that combining two extreme training is best (max quickness and then max CA), is it worth trying a different Max CA training program which has less of a focus on quickness?

@Middleweight165

U18:  [Quickness]+[Match Practice]+[Attacking]+[Recovery]x7+[Double Intensity] +[Additional Focus Quickness]
B Team:  [Rest]+[Double Intensity]+[Addtional Focus Quickness]

Then move players into first team once Acc and Pace >16

First Team: [Quickness]+[Match Practice]+[Attacking]+[Recovery]x7+[Double Intensity]

Additional focus is removed so that CA growth becomes less specific to quickness and creates a more rounded player once my 16 threshold is met.

rossbalch said: So here's a question, it seems like league reputation doesn't matter. But what about first team, vs say a "B" team. If you set it up so both the First Team and the B team compete in the same league, do the first team players develop better than the First Team players? Expand

They develop the same

Robbo84FM said: Ok and then what schedule do you use for the senior squad? Expand

Plan 3 but have no additional focus (or you can choose one if you want)

Thank you @harvestgreen22 !

Based on the data the best training strategies are either Strategy 17 (1 in the previous version) or Strategy 21 (2 in the previous version).
Atm I prefer Strategy 21.

To apply this to the game do as follows:
1. Use Plan 1 or 3 for the u18s (2 years, age 15/16-18)
2. when they age out move to B Team where you should use the opposite of what you used for u18s
3. Then when either CA high enough or Pace/Acc high enough (depends which way round you did it), move to senior squad. I stop actively training Pace/Acc when both are >16
4. When in senior squad remove Quickness additional focus

HOWEVER, I have some questions:

In the image I compare the growth patterns for the 2 different strategies (Note: Strat 1=Strat 17, and Strat 2=Strat 21, they just have different starting parameters).
1. When comparing Strat 1 vs Strat 2 something is wrong with your data @harvestgreen22 . Strat 2 has a much higher final CA, but overall has lower attributes, even when taking attribute weighting into account.
2. Additionally, when comparing Strat 17 vs Strat 21, Strat 17 appears to be slightly better. Which makes me question the CA values in the initial Strat 1 vs Strat 2 test even more.

Thanks again, @harvestgreen22 . Hopefully this is useful for everyone.

Interesting research! Despite using a slightly different methodology, your results were very similar to those of the chinese tests (see image) done recently. This is good because it shows that the results are replicable. (Second image shows comparison)

HOWEVER, the testing methodology is not necessarily the most robust, for a number of reasons but is probably as robust as you can get without given the time and processing constraints. (Am a data scientist)

Having scoured the FM forums, the most robust (by far) is the testing done by ykykyk0525 (some basketball developer) who employ a multi-layered AI neural network. Their explanation is translated here: https://fm-arena.com/thread/2182-important-attributes-for-blue-dm/ . Admittedly this was done for FM22, so there might have been some change in the game engine and attribute effect, but knowing SI unlikely.

An additional benefit is that it provides different weightings for different positions (for a specified tactic)

@Zippo I was wondering what your thoughts were?

harvestgreen22 said: What are the initial conditions for the CA program? You need to give a specific value

Initial age,
Initial attribute ( such as ' all-attribute 11 ' ) or initial CA (such as 130CA)
Initial PA
I'll try it when I have time Expand

Initial age 16, initial CA = 90, initial PA=160, professionalism= 12

Try different combintations of the following training sessions:
Plan1. [Rest]+[Double Intensity]+[Addtional Focus Quickness]
Plan2. [Quickness]+[Attacking Direct]+[Recovery]x7+[Addtional Focus Quickness]
Plan3. [Quickness]+[Match Practice]+[Attacking]+[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]

Try the them in groups of 2 years from age 16 to 20.
E.g. 16-18: Plan1, 18-20: Plan3

Plan1, Plan1
Plan1, Plan2
Plan1, Plan3
Plan2, Plan1
Plan2, Plan2
Plan2, Plan3
Plan3, Plan1
Plan3, Plan2
Plan3, Plan3

@harvestgreen22 are you going to try different combinations with the high CA program?:
[Quickness]+[Match Practice]+[Attacking]+[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]

Just looking at the raw results in your most recent post.

By my calculations C12, J11, K10, Q10 are the best options based on how much useful CA they generate.

(@harvestgreen22 I'm not quite sure why you are focussing on E12 and U11?)

Based on my analysis you can pick one of those 4 I listed based on how extreme you want to go with focussing on only developing the 'impactful' attributes.

Most Extreme -> C12 -> J11 -> K10 -> Q10 -> Least Extreme

With C12 and J11 you will lose attributes in other places.

I would say the most 'balanced' option is definitely K10.

(also @harvestgreen22 why did you at some point you switch from 'attacking wings' to 'attacking shadow play'?)

@harvestgreen22 This is the original thread, which you might find interesting: https://playgm.cc/thread-943500-5-1.html

@harvestgreen22 Why do, for some attributes, your results differ a lot from the ykykyk05251 (a chinese game developer) results?