Revaluation of the CA - PA system

[ham_aka_stam]

26 minutes ago, Cap'nRad said:

 

 

But people are talking about surpassing PA or increasing it if they move to a bigger club.  That's madness.

And yes, they might be talking about replacing it with a better system, but other than a further use of the minus PA, something already available, no-one has suggested a "better system".

[BruceyNTFC]

On 20/11/2017 at 18:18, Cap'nRad said:

 

That's my problem as well, that it might get a lot more complex for everyone. Although seeing as the researchers work for free, I would assume they enjoy what they do, and thus might like the added thought put into it. As @BruceyNTFC shows above.

Yeah for sure more work, but I do it so my club is as accurate as it can be in my favourite game. Can’t speak for anyone else, but any change for the better is welcome.

[Maaka]

8 minutes ago, ham_aka_stam said:

And yes, they might be talking about replacing it with a better system, but other than a further use of the minus PA, something already available, no-one has suggested a "better system".

This.

11 minutes ago, ham_aka_stam said:

Equally no-one has offered an alternative which is any better.

And this.

[Maaka]

-double post-

[Maaka]

-triple post-

[Dorin]

7 minutes ago, ham_aka_stam said:

But why? No-one has pointed to a reason why which isn't immediately disproved.  Equally no-one has offered an alternative which is any better.

My personal reason why has to do a bit with the youth rating set per nation that seems not to change throughout the game.  Since there are no abilities to work more with youth academies (or starting some academies like Ajax might have in South Africa for example) , the only way to get better players from underrated nations would be to try scouting a lot of young players of those nationalities and bringing them in a successful club setup where I'm hoping that they would improve given all the improved facilities, coaches, better opponents, etc. But this is pretty much impossible to do in the current FM game world where potential is fixed at the start (again, referring regens here) and nation youth ratings never change so very very few players of good PA are produced for some nationalities. I'd like to focus more on the 'nurture' part of growing players and my goal is in the long term save to develop nations football level and the level of their football competitions. 

[ham_aka_stam]

2 minutes ago, Dorin said:

My personal reason why has to do a bit with the youth rating set per nation that seems not to change throughout the game.  Since there are no abilities to work more with youth academies (or starting some academies like Ajax might have in South Africa for example) , the only way to get better players from underrated nations would be to try scouting a lot of young players of those nationalities and bringing them in a successful club setup where I'm hoping that they would improve given all the improved facilities, coaches, better opponents, etc. But this is pretty much impossible to do in the current FM game world where potential is fixed at the start (again, referring regens here) and nation youth ratings never change so very very few players of good PA are produced for some nationalities. I'd like to focus more on the 'nurture' part of growing players and my goal is in the long term save to develop nations football level and the level of their football competitions. 

You just need nation youth ratings to be fluid then... Maybe including the average CA of players from that nation into the equation would allow you to increase it by scouting more players and bringing more through your academy.  Or a way of investing your money into a given nation's youth setup.

But that's not a flaw in PA... just something the devs haven't got around to adding to the code yet.

[michaeltmurrayuk]

3 hours ago, Cap'nRad said:

I was talking more about the 3 variables in tandem, not just natural talent in isolation.

In real life, Jamie Vardy improved due to an improved attitude. He always had the physical talent (aka high physical rating), but lacked the professionalism and ambition. In the current system, once he hits the age of 24, he's done for. The best case scenario is he's a lower league striker. In the proposed system, if he improves his attitude (see below) and works hard in training (with the manager focusing on his physical attributes) then he might improve his physicals despite being over 24. A boost maybe high enough to make him overperform in his league, get picked up by a bigger team, and then improve his mentals faster due to playing in a higher league and being more dedicated to it. Sure, he might never reach Leicester levels but that would vary from save to save. And of course this would all be in the best case scenarios, which is fortunately what his real life was for him: simply one of the best case scenarios given his situation in the lower leagues. 

As you can see, in the proposal, there is the possibility for some upward variability, which even though it may not reflect real life exactly, is still more realistic and variable than the current system in which he would be doomed to retire at 26.

 

PS: This is a good example of another problem with the current 'age-limiting' system: Above a certain age, players attitudes are very hard to improve. It shouldn't be very easy, but possible; in game it's near impossible. Although I'm not sure if this is the thread for that.

This is already possible in the game - players have peaked at different ages for different positions for most if not all of FM's history (Strikers is I think 27-29, GK 30-31 for example), whilst SI have said that the player progression system has been tweaked over the last few years to make it more unpredictable increasing the chances of late bloomers and wonderkids peaking early. Whilst the role system makes it easier for players to outperform their CA if you play them to their strengths, though not sure how much the AI uses roles over CA (one of the features mentioned the AI looking at players on form but I haven't really played FM18 to check how it's working out).

[Dorin]

1 minute ago, ham_aka_stam said:

You just need nation youth ratings to be fluid then... Maybe including the average CA of players from that nation into the equation would allow you to increase it by scouting more players and bringing more through your academy.  Or a way of investing your money into a given nation's youth setup.

But that's not a flaw in PA... just something the devs haven't got around to adding to the code yet.

In that case I think it would be a great addition to the code and in my opinion would improve a lot the gaming experience as the seasons progress, even if it affects PA only indirectly,  maybe somehow to tie it in also with competition/nation rankings as they evolve.
Do you know if this is on the list of things being worked on by the devs for future editions? 

[michaeltmurrayuk]

18 minutes ago, Dorin said:

My personal reason why has to do a bit with the youth rating set per nation that seems not to change throughout the game.  Since there are no abilities to work more with youth academies (or starting some academies like Ajax might have in South Africa for example) , the only way to get better players from underrated nations would be to try scouting a lot of young players of those nationalities and bringing them in a successful club setup where I'm hoping that they would improve given all the improved facilities, coaches, better opponents, etc. But this is pretty much impossible to do in the current FM game world where potential is fixed at the start (again, referring regens here) and nation youth ratings never change so very very few players of good PA are produced for some nationalities. I'd like to focus more on the 'nurture' part of growing players and my goal is in the long term save to develop nations football level and the level of their football competitions. 

That's probably worth a separate thread as it's a different issue. Though I think some of this can be improved in game as users have had success improving the San Marino regens using the San Marino club side which plays in the Italian league, though not sure if that is because they are borrowing the 'Italian' rating.

[Cougar2010]

7 minutes ago, Dorin said:

In that case I think it would be a great addition to the code and in my opinion would improve a lot the gaming experience as the seasons progress, even if it affects PA only indirectly,  maybe somehow to tie it in also with competition/nation rankings as they evolve.
Do you know if this is on the list of things being worked on by the devs for future editions? 

Yeah, I agree that a fixed youth nation rating is an issue for those of us who like long term games.

It basically means that countries with low ratings will never achieve anything in International competition, they might get a world beater but you need more than that.  Taking an example of Poland who I managed in FM17.  In 2018-2022 they had a good core of a "golden" generation and performed well in International tournaments.  I took over in 2024 when the bulk of that generation were over 30yo with over half of them retired.  Aside from one world class ML newgen the rest of the squad just wasn't anywhere near the same level and dropped from knocking on the door of the top 10 in the World Rankings to being well outside the top 50.

 

[Cap'nRad]

53 minutes ago, ham_aka_stam said:

It does allow the future to happen.

It clearly doesn't seeing as researchers have to adjust P.A values across fm's. If all paths were accounted for this would be unnecessary.

 

53 minutes ago, ham_aka_stam said:

There isn't only one path, there is only one maximum CA he can reach.  Firstly, CA is not the be all and end all.  This is the maximum the researcher thinks that player could reach.  It doesn't mean he's going to reach it, or follow a path that allows him to reach it.  If  he languishes in the lower leagues, he'll never reach it.  If he chooses to play for teams who play him out of position, he may perform badly and never reach it.  If he gets a bad injury, he may never reach it.

Nature is not the PA, nature is your assumption that they have a fixed path regardless of environmental factors.  That simply isn't the case in FM.

I think I see the problem. You're focusing on the variability of the game downward; instead you should focus on it's upward variability. Is it possible for Jamie Vardy in fm13 to become fm17's Vardy? That limiting factor is the problem.

 

53 minutes ago, ham_aka_stam said:

You mention a player getting passed what his P.A. would have been.  Can you not see how daft that is?  If players could be better than their maximum possible ability, then you'd just grab any high CA youngster from the lower leagues, play them, and they'd improve indefinitely.

I've repeated myself so many times now, I really don't want to continue saying the same thing. So I'll try and be as clear as possible: 

If there were two fm13's, one with the current system and the other with a system in which P.A is essentially limitless, then players from the limitless fm could have a  C.A that exceeds the P.A of the same players in the other fm. That is what I meant when I said a player getting passed what his P.A might have been. And if you read my first response, and a few of my later ones, you would understand that I immediately mentioned that the main issue to the proposal would be figuring out some sort of growth limiting factor so that players don't grow endlessly. That's something to figure out.

I don't know if I can make this any more clear than that. Apologies if I sound direct but I feel like I've been making the same arguments over and over.

[Cap'nRad]

29 minutes ago, michaeltmurrayuk said:

This is already possible in the game - players have peaked at different ages for different positions for most if not all of FM's history (Strikers is I think 27-29, GK 30-31 for example), whilst SI have said that the player progression system has been tweaked over the last few years to make it more unpredictable increasing the chances of late bloomers and wonderkids peaking early. Whilst the role system makes it easier for players to outperform their CA if you play them to their strengths, though not sure how much the AI uses roles over CA (one of the features mentioned the AI looking at players on form but I haven't really played FM18 to check how it's working out).

Yes, I think that tweak is a good step. However isn't the peak you talk about more in terms of performances and consistency than actual changes in ability?

[Cap'nRad]

55 minutes ago, ham_aka_stam said:

But people are talking about surpassing PA or increasing it if they move to a bigger club.  That's madness.

Again, no one's saying this that I can see.

 

56 minutes ago, ham_aka_stam said:

And yes, they might be talking about replacing it with a better system, but other than a further use of the minus PA, something already available, no-one has suggested a "better system".

What are the problems with the proposals of @Raptor Longe and myself (besides the one I already outlined, of course)? If you can identify them then we can make progress. 

[enigmatic]

54 minutes ago, Cap'nRad said:

Again, no one's saying this that I can see.

What are the problems with the proposals of @Raptor Longe and myself (besides the one I already outlined, of course)? If you can identify them then we can make progress. 

There are six pages of arguments, some of them very detailed, against these proposals already. Perhaps more progress would be made if you deigned to read them...

 

 

 

As for the nation's "youth rating", my experiments with FM17 suggested it had very little impact on national side player potential relative to other facts such as nation reputation (which is very dynamic) and clubs

 

[Cap'nRad]

2 minutes ago, enigmatic said:

There are six pages of arguments, some of them very detailed, against these proposals already. Perhaps more progress would be made if you deigned to read them...

Give me an example of an argument I haven't addressed, and I'll respond to it. I've already gone through the thread and I don't think I've missed anything.

[Cougar2010]

2 hours ago, Cap'nRad said:

I've repeated myself so many times now, I really don't want to continue saying the same thing. So I'll try and be as clear as possible: 

If there were two fm13's, one with the current system and the other with a system in which P.A is essentially limitless, then players from the limitless fm could have a  C.A that exceeds the P.A of the same players in the other fm. That is what I meant when I said a player getting passed what his P.A might have been. And if you read my first response, and a few of my later ones, you would understand that I immediately mentioned that the main issue to the proposal would be figuring out some sort of growth limiting factor so that players don't grow endlessly. That's something to figure out.

I don't know if I can make this any more clear than that. Apologies if I sound direct but I feel like I've been making the same arguments over and over.

Kinda feel like the point is being repeated here but:

If you are going to limit the growth so that a player doesn't grow endlessly how is that different to PA?

You are effectively capping the player but in a different way.

[Mikado911]

19 minutes ago, Cap'nRad said:

Give me an example of an argument I haven't addressed, and I'll respond to it. I've already gone through the thread and I don't think I've missed anything.

I do not understand how you addressed the argument presented by @ham_aka_stam with his post on Vardy and various others before. I will try to make the point once again in a different way. Let us consider different scenarii:

We are in 2013 (so we have no information on the future) and we are the researcher that has to evaluate him. Here is what we know at the time:

 - Quick and physical player with no good technique skills.

 - Deplorable attitude and full of bad habits.

 - No visible ambition.

 

Now, what will probably happen and how should we rate him as researchers?

 

In the current system:

Low ambition and determination, low PA.

Result: Vardy will never develop as in real life.

 

In your proposed system:

Low ambition and determination, low natural talent.

Result: Vardy will never develop as in real life (only marginally improvement if you are really lucky).

 

In both cases, the researcher could not predict the actual outcome. In fact nobody could otherwise he would have been signed much earlier by a very good club to be built into an excellent player if that was all it took.

 

What can we do if we change things?

If we want to add more variability to the game, we need some variable attributes, random attributes or random events. But if you can have them affecting players positively, we also need to have them affect players negatively the SAME way. Conclusion, the more you add this randomness, the more unpredictable the game becomes and that is why it works well with regens and not with real life players. Assume that you play a game and one player you brought has a PA>180 (or natural talent>18), professionalism is over 15 and suddenly due to some random event (he found that girl in a club, starts doing drugs for fun, drinks in excess) he lost all that (see George Best or Diego Maradona but before Maradona became a god or more recently Samir Nasri, Hatem Ben Arfa who completely wasted their greatness potential). You suddenly got a failed product that goes nowhere. What reaction would you have? Probably that the system is rigged against you and that this is unrealistic. The situation with regens is better. Because even if the path is set as you say, it will be different for every regen.

Now, some source of improvement might be that you randomize more the PA or natural ability at the beginning of your save. Hence you get more variability from one save from another. Try not to forget however that this makes things more problematic on a lot of gaming aspect. No more, look how this player became good in my save with your friends and then your friend tries with the same player and fails (because the player is just bad in his save). Lots of "look that player has incredible potential in the real world but he is a piece of crap in my game -> so unrealistic". People tends to enjoy random events that go in their favour not the other way around.

The truth is, there is  way to have more fun with more randomness (if that is what you want) and actually enjoy the game this way: randomizes all players at the beginning of the save. I am not sure how many do actually that.

 

Finally remember, no mathematical model can accurately predict the future. We build system that can be correct in most cases. In the Vardy case, you probably have hundred if not thousands of players in the same situation (low level play with no visible future) and one made it that we missed. Conclusion, the accuracy of the system is over 99%, that is pretty damn good.

 

 

[Mikado911]

2 hours ago, Dorin said:

In that case I think it would be a great addition to the code and in my opinion would improve a lot the gaming experience as the seasons progress, even if it affects PA only indirectly,  maybe somehow to tie it in also with competition/nation rankings as they evolve.
Do you know if this is on the list of things being worked on by the devs for future editions? 

That sounds like an awesome idea. You should put it in another thread.

[Dorin]

Since there have been many posts so far and no real specific suggestions of an improved system, I'll give it a try. It will be a long one and maybe more hard to follow, but hope you appreciate the effort at least. I've added some game logic to keep it as specific as possible to a system improvement suggestion, so hope the proposed game logic about PA increase will make sense:
 

Trigger event for re-evaluating PA:
Trigger event: If at a later date within the game the CA = PA for a player

THEN:

Step 1. 

Take into consideration key mental attributes: ambition, determination, consistency (i think these are key mental attributes to overachieving but used for example only here, so others can be added)  - this is paX (see below)

Variables used for step 1:
X2= result of a math formula for creating an average score out of the key ratings (such as the 3 I mentioned above), where some ratings may weight more than others. 
X2 range can be between 0.1 and 2
eg: X2= 1.7  for a player with very high key rating attributes
    X2= 0.3  for a player with low key rating attributes

X1= standard benchmark, eg determined when calculating ratings with a medium value of 10 for determination, ambition, consistency for example.
    For simplicity of calculation in this example this can be set to 1 here, but would normally be a more specific result when applying the formula for scoring the key ratings.

paX=  X2/X1  
eg.   paX=1.5

AND 

Step 2. 
Take into consideration: training quality, club facilities, coach star ratings AND match experience in previous year + player ratings in those matches: paY

Variables for step 2:
Y2= result of a math formula for creating an average score out of the ratings mentioned at Step 2. Each rating there can have a different weight on Y2 result (so that match experience and player ratings can be more important than training facilities for example)

Y1= Similar to Y2, but this is set at the date of youth intake ( to allow to measure environment change and how drastic this is, so if Y2 is much higher than Y1 it means the player is now in a much better environment to reach his potential set at game start and perhaps overachieve, where overachieving means increasing allowed PA)

paY=Y2/Y1
eg: paY = 1.3 if the current setup provides afor example 30% better conditions than those that were present when the player was generated in the game world.
    paY= 0.8 if the current setup is worse

If paX & paY are both higher than 1, meaning all conditions exist for further improvement both from an individual perspective and environment/circumstances perspective then trigger PA increase event.

PA increase event:
Percentage of PA increase per event should be tied directly with paX & paY scores. The higher the score the higher PA increase percentage should be allowed, but I'd suggest not more than 5%-15% on a single PA increase event, since that's already a good jump in potential.
So in the most unlikely max scenario a player with max paX and paY score would increase 15% of his potential. If he is rated at 100 PA, his new PA will be 115, but since this is with perfect key attributes & all perfect conditions, it's not a realistic possibility. If the PA could increase by 3 to 7 points it would still bring a considerable boost to keep things dynamic and more realistic, taking into account many other factors and not doing it randomly as it was so often mentioned in this thread. 

Attributes allowed to be increased should also be different per category: mental, technical and physical, with physical increase being either none or very limited, since the players have obvious physical limits that cannot be improved on after a certain point (such as how fast you can run or how high you can reach). 

Further reevaluation of player potential after the first PA increase:
Trigger PA increase event again after X time passes from previous reevaluation and if CA=new PA   (for example only after 1 year min from previous PA increase event). 

Obviously, if the player gets injured, doesn't play regularly, doesn't train well, etc then the CA will just decrease and as long as the player never reaches his potential there's nothing to reevaluate.

[Cap'nRad]

29 minutes ago, Cougar2010 said:

Kinda feel like the point is being repeated here but:

If you are going to limit the growth so that a player doesn't grow endlessly how is that different to PA?

You are effectively capping the player but in a different way.

This is a valid point. I feel the difference is in it being a limiter rather than a hard cap, however, we haven't even come up with one, I don't know the nature of it so don't know for sure. 

[Cougar2010]

5 minutes ago, Cap'nRad said:

This is a valid point. I feel the difference is in it being a limiter rather than a hard cap, however, we haven't even come up with one, I don't know the nature of it so don't know for sure. 

Well the idea of a limiter was Raptor's original suggestion in the OP - "Natural Talent"

 

[Cap'nRad]

57 minutes ago, Mikado911 said:

In your proposed system:

Low ambition and determination, low natural talent.

Result: Vardy will never develop as in real life (only marginally improvement if you are really lucky).

The difference would be that he would be rated highly physically due to his good physicals. And with the potential to improve further despite being over 24, as well as a change in attitude despite being over 24, as well as the lack of a limit on his P.A, he might reach higher levels. He may not reach his current level (which is a very good case scenario), but he would probably grow beyond his hard limit in your first scenario. Do you understand?

57 minutes ago, Mikado911 said:

If we want to add more variability to the game, we need some variable attributes, random attributes or random events. But if you can have them affecting players positively, we also need to have them affect players negatively the SAME way.

I feel this part is not needed because, as has been stated earlier, the problem we have with P.A is the upper limit. Values above a certain P.A are inaccessible, while in game it is possible for a player to fall short of his P.A so this part is already covered naturally by the game.

 

57 minutes ago, Mikado911 said:

Conclusion, the more you add this randomness, the more unpredictable the game becomes and that is why it works well with regens and not with real life players.

True. However, the introduction of these values (physical, mental, and technical potential) as well as some sort of growth monitor, is what will be used to control the randomness. Also, life is quite unpredictable so it would be cool if the game could mirror that aspect without being totally inaccurate.

 

57 minutes ago, Mikado911 said:

Assume that you play a game and one player you brought has a PA>180 (or natural talent>18), professionalism is over 15 and suddenly due to some random event (he found that girl in a club, starts doing drugs for fun, drinks in excess) he lost all that (see George Best or Diego Maradona but before Maradona became a god or more recently Samir Nasri, Hatem Ben Arfa who completely wasted their greatness potential). You suddenly got a failed product that goes nowhere. What reaction would you have? Probably that the system is rigged against you and that this is unrealistic. The situation with regens is better. Because even if the path is set as you say, it will be different for every regen.

Firstly, this isn't the exact system I proposed although this sprinkle of randomness could be interesting if it is controllable. Secondly, doesn't the game somewhat mirror this scenario with youngsters who get tutored by less than professional players mirroring the bad attitude? It may not be the exact scenario but it's similar, and has similar effects.

 

57 minutes ago, Mikado911 said:

Now, some source of improvement might be that you randomize more the PA or natural ability at the beginning of your save. Hence you get more variability from one save from another. Try not to forget however that this makes things more problematic on a lot of gaming aspect. No more, look how this player became good in my save with your friends and then your friend tries with the same player and fails (because the player is just bad in his save). Lots of "look that player has incredible potential in the real world but he is a piece of crap in my game -> so unrealistic". People tends to enjoy random events that go in their favour not the other way around.

Yes, which is why I didn't think more on this method. It involves a type of game-generated randomness as opposed to naturally created through gameplay.

 

57 minutes ago, Mikado911 said:

The truth is, there is  way to have more fun with more randomness (if that is what you want) and actually enjoy the game this way: randomizes all players at the beginning of the save. I am not sure how many do actually that.

Oh yeah that's true but firstly I doubt many use that option. Tbh I even forgot it exists! And secondly, it still has a set limit on players P.A so that kind of randomization doesn't solve the problem presented by O.P.

 

57 minutes ago, Mikado911 said:

Finally remember, no mathematical model can accurately predict the future. We build system that can be correct in most cases. In the Vardy case, you probably have hundred if not thousands of players in the same situation (low level play with no visible future) and one made it that we missed.

Exactly why I think trying to figure out a maximum number for a player's potential ability is faulty. Pretending that the game is real life, we should say that there are many ways a player's career could go, but they are mostly governed by his genetics, his environment, and his attitude. I don't think P.A fits into any of these categories, but rather is as a result of these categories. So basically P.A should be the number you become rather than the highest number you could ever become.

 

57 minutes ago, Mikado911 said:

Conclusion, the accuracy of the system is over 99%, that is pretty damn good.

Really, 99%? I assumed the accuracy would be below 99% if we're not including revisions.

[Cap'nRad]

21 minutes ago, Cougar2010 said:

Well the idea of a limiter was Raptor's original suggestion in the OP - "Natural Talent"

 

Yes this is what I mean. A limiter as in a variable that slows down progression, without capping it as P.A would. This would be so that late progression is still possible.

[Mikado911]

Thanks @Dorin. This is a new idea with variable PA.

My take on it is that I disagree with its interest. In your proposition, a player with excellent mental attributes and training in a good club will automatically have an unlimited growth of its PA until it reaches the maximum you can allow. Unless your PA boost is a one time event but I do not when you will trigger it then. In some sense, you make the PA irrelevant and base all your development on mental attributes and training. Therefore, you just replace PA by metal attributes (I caricature a bit, I admit).

My problem with it is that to improve, you need both a perfect dedication and the room to improve. I have my limitations and despite training hard and being dedicated I could never go beyond a certain level (at least in sports).

[Cougar2010]

Just now, Cap'nRad said:

Yes this is what I mean. A limiter as in a variable that slows down progression, without capping it as P.A would. This would be so that late progression is still possible.

But as discussed earlier in the thread thats all about CA progression and not PA.

If CA progression was sorted so that it worked in the way we want it to PA would become irrelevant and could be done away with.  The issue is no-one has come up with a workable idea that does that effectively.

Having fixed control over that progression also doesn't work because that leads to linear growth rather than the variable growth we already have under the PA system.

[Cap'nRad]

4 minutes ago, Cougar2010 said:

But as discussed earlier in the thread thats all about CA progression and not PA.

If CA progression was sorted so that it worked in the way we want it to PA would become irrelevant and could be done away with.  The issue is no-one has come up with a workable idea that does that effectively.

Yeah, I agree that's the problem. If age isn't the limit and P.A isn't the limit, then what is it? That might be tough to figure out.

4 minutes ago, Cougar2010 said:

Having fixed control over that progression also doesn't work because that leads to linear growth rather than the variable growth we already have under the PA system.

What do you mean by fixed control? You mean the limiting variable?

[Mikado911]

9 minutes ago, Cap'nRad said:

The difference would be that he would be rated highly physically due to his good physicals. And with the potential to improve further despite being over 24, as well as a change in attitude despite being over 24, as well as the lack of a limit on his P.A, he might reach higher levels. He may not reach his current level (which is a very good case scenario), but he would probably grow beyond his hard limit in your first scenario. Do you understand?

Very well. How to fix it in the current system? Just have its PA 5 points higher and its done. You will never go beyond that with your proposed system (what I mean by marginally improving).

 

About the 99% accuracy, here is how I measure it: How many outliers with progress is completely different from what the game predicted out of the maybe 60000 players that are modelled? Very few have completely be beyond expectations if you look five years down the road (both positively or negatively). I dare say less than 600 so that means that the game got it more or less correctly in 99% of the cases. Hence the 99% accuracy.

Now, the updates you make every year serves as recalibrating the game and taking these outliers into account and this is necessary.

[Cap'nRad]

31 minutes ago, Mikado911 said:

Very well. How to fix it in the current system? Just have its PA 5 points higher and its done. You will never go beyond that with your proposed system (what I mean by marginally improving).

Firstly, in the current system a player's growth is limited after the age of 24, so 5 extra P.A points won't do much  (although I hear that may have changed in fm18, although if so I don't know the extent of the change).

Secondly, if you go back to my first post on this topic, you'll see me explain how mental attributes would improve through time (given the player is dedicated to improving) without any cap or reduction in other areas to compensate. The only limit to his mental attributes would be his mentality, which means with a changed mentality and improved physical and mental attributes, he can potentially reach a lot more than 5 P.A, (with the added bonus of it flowing naturally with the game).

The issue would then be making sure his mental attributes' growth is regulated somehow, which is what I'm currently trying to figure out. 

 

31 minutes ago, Mikado911 said:

About the 99% accuracy, here is how I measure it: How many outliers with progress is completely different from what the game predicted out of the maybe 60000 players that are modelled? Very few have completely be beyond expectations if you look five years down the road (both positively or negatively). I dare say less than 600 so that means that the game got it more or less correctly in 99% of the cases. Hence the 99% accuracy.

Now, the updates you make every year serves as recalibrating the game and taking these outliers into account and this is necessary.

I would assume if you go back to the very first time a player appeared in an fm game and compare their P.A with that of their latest self, it would be quite different. This is why I was surprised with your estimate.

 

[Cougar2010]

19 minutes ago, Cap'nRad said:

Yeah, I agree that's the problem. If age isn't the limit and P.A isn't the limit, then what is it? That might be tough to figure out.

What do you mean by fixed control? You mean the limiting variable?

By fixed control I mean a variable that creates fixed predictable growth so the graph would look like this:

 

Rather than variable growth which could look like any of the following:

[enigmatic]

1 hour ago, Cap'nRad said:

The difference would be that he would be rated highly physically due to his good physicals. And with the potential to improve further despite being over 24, as well as a change in attitude despite being over 24, as well as the lack of a limit on his P.A, he might reach higher levels. He may not reach his current level (which is a very good case scenario), but he would probably grow beyond his hard limit in your first scenario. Do you understand?

I certainly don't understand why SI would calibrate a new system where players probably grow beyond the limits of the PA set by researchers at present (which most players don't even reach in game). They haven't spent decades collecting data points on how good researchers think older players actually are and balancing how good researchers think younger players might become to permit any new development model to ignore all that and assume that the younger players are mostly going to be a lot better than the last generation and if they've got the right attitude they'll continue improving indefinitely.

If you changed the hard limit into a combination of growth rate parameters, it's going to be balanced so on average players end up about the same level before. Some players might be capable of doing a bit better than their current PAs and some players would do a bit worse. Not necessarily the ones researchers want. 

But there's no logical development model that's going to allow a full grown adult noted for his unprofessionalism and not noted for particularly stellar talent to develop like this and work across a database full of unheralded, quick players  scoring plenty in lower divisions.

Most strikers who score nearly 29 goals in the Northern League don't have the talent to make it in League 2, never mind the England team, and researchers are going to continue to be expected to acknowledge that reality with whatever growth parameters they're allowed to set.

It's especially hard when I can't think of a single growth parameter (age, attitude, talent, club trained at) which, say, Tom Cleverley wouldn't have been logically given much, much better values than Vardy, yet the system needs to be designed to ensure Cleverley didn't improve much over similar periods. Of course the much easier way is to listen to all those people saying "yeah, Cleverley's a good player at a big club with a good attitude but he's never going to be any better than say, James Milner", and set a hard limit on his potential ability around the level of James Milner.

[Cap'nRad]

1 hour ago, Dorin said:

Since there have been many posts so far and no real specific suggestions of an improved system, I'll give it a try. It will be a long one and maybe more hard to follow, but hope you appreciate the effort at least. I've added some game logic to keep it as specific as possible to a system improvement suggestion, so hope the proposed game logic about PA increase will make sense:
 

Trigger event for re-evaluating PA:
Trigger event: If at a later date within the game the CA = PA for a player

THEN:

Step 1. 

Take into consideration key mental attributes: ambition, determination, consistency (i think these are key mental attributes to overachieving but used for example only here, so others can be added)  - this is paX (see below)

Variables used for step 1:
X2= result of a math formula for creating an average score out of the key ratings (such as the 3 I mentioned above), where some ratings may weight more than others. 
X2 range can be between 0.1 and 2
eg: X2= 1.7  for a player with very high key rating attributes
    X2= 0.3  for a player with low key rating attributes

X1= standard benchmark, eg determined when calculating ratings with a medium value of 10 for determination, ambition, consistency for example.
    For simplicity of calculation in this example this can be set to 1 here, but would normally be a more specific result when applying the formula for scoring the key ratings.

paX=  X2/X1  
eg.   paX=1.5

AND 

Step 2. 
Take into consideration: training quality, club facilities, coach star ratings AND match experience in previous year + player ratings in those matches: paY

Variables for step 2:
Y2= result of a math formula for creating an average score out of the ratings mentioned at Step 2. Each rating there can have a different weight on Y2 result (so that match experience and player ratings can be more important than training facilities for example)

Y1= Similar to Y2, but this is set at the date of youth intake ( to allow to measure environment change and how drastic this is, so if Y2 is much higher than Y1 it means the player is now in a much better environment to reach his potential set at game start and perhaps overachieve, where overachieving means increasing allowed PA)

paY=Y2/Y1
eg: paY = 1.3 if the current setup provides afor example 30% better conditions than those that were present when the player was generated in the game world.
    paY= 0.8 if the current setup is worse

If paX & paY are both higher than 1, meaning all conditions exist for further improvement both from an individual perspective and environment/circumstances perspective then trigger PA increase event.

PA increase event:
Percentage of PA increase per event should be tied directly with paX & paY scores. The higher the score the higher PA increase percentage should be allowed, but I'd suggest not more than 5%-15% on a single PA increase event, since that's already a good jump in potential.
So in the most unlikely max scenario a player with max paX and paY score would increase 15% of his potential. If he is rated at 100 PA, his new PA will be 115, but since this is with perfect key attributes & all perfect conditions, it's not a realistic possibility. If the PA could increase by 3 to 7 points it would still bring a considerable boost to keep things dynamic and more realistic, taking into account many other factors and not doing it randomly as it was so often mentioned in this thread. 

Attributes allowed to be increased should also be different per category: mental, technical and physical, with physical increase being either none or very limited, since the players have obvious physical limits that cannot be improved on after a certain point (such as how fast you can run or how high you can reach). 

Further reevaluation of player potential after the first PA increase:
Trigger PA increase event again after X time passes from previous reevaluation and if CA=new PA   (for example only after 1 year min from previous PA increase event). 

Obviously, if the player gets injured, doesn't play regularly, doesn't train well, etc then the CA will just decrease and as long as the player never reaches his potential there's nothing to reevaluate.

This seems interesting. The main thing I would be worried about is that in a very good case scenario, a club with maximum facilities, top coaches, and excellent tutors can offer game time to random youngsters in such a way that their ratings are good enough to trigger the PA increase event every year or two and they grow exponentially. 

[HUNT3R]

IMO, the focus should be less on PA and more on CA development. PA is fine as it is. If it was even harder (especially talking about us rather than the AI) to develop CA, then researchers can assign slightly higher PAs. This would allow for the player to reach, say, 80-90% of the PA assigned in most saves, but also allow for cases where he's both better and worse than they would expect.

Natural talent is just another PA. Someone will need to set it  what it is like they do PA and CA. In the end, it's basically CA development and it needs to have limiters in place so as not to get every second 50CA player becoming another Messi. The limiter in the current system is PA. Whether we're talking real or regen players, it works as it is - very well, in fact, if you don't look at the PA number.

[Dorin]

5 hours ago, Cap'nRad said:

This seems interesting. The main thing I would be worried about is that in a very good case scenario, a club with maximum facilities, top coaches, and excellent tutors can offer game time to random youngsters in such a way that their ratings are good enough to trigger the PA increase event every year or two and they grow exponentially. 

Yes, there is some playing around with the formulas to not offer exponential growth, but rather a curved growth that slows down if more PA increase events occur for the same player.  This could simply mean adding a percentage to Y2 variable that calculates paY for example.

so if paY=Y2/Y1, after the first PA increase the new Y2=0.75% of initial Y2  .My reasoning is that the environment still plays an important part in player development, but from a year to year in the new environment progress isn't that drastic since there isn't much of a circumstances shift, and maybe match ratings, experience, etc should weight more after. 
Of course, this can be calculated in a new way also based on the improvement of the current clubs' facilities, training and so on.. so if there isn't much improvement constantly this limit becomes lower and lower each time after the first PA increase simply due to the formula used.

Or something like that


PS: edit, my maths was wrong on that one earlier, adjusted it now  ( I mixed up Y2 with Y1)
 

[ham_aka_stam]

10 hours ago, Cap'nRad said:

It clearly doesn't seeing as researchers have to adjust P.A values across fm's. If all paths were accounted for this would be unnecessary.

I think I see the problem. You're focusing on the variability of the game downward; instead you should focus on it's upward variability. Is it possible for Jamie Vardy in fm13 to become fm17's Vardy? That limiting factor is the problem.

I've repeated myself so many times now, I really don't want to continue saying the same thing. So I'll try and be as clear as possible: 

If there were two fm13's, one with the current system and the other with a system in which P.A is essentially limitless, then players from the limitless fm could have a  C.A that exceeds the P.A of the same players in the other fm. That is what I meant when I said a player getting passed what his P.A might have been. And if you read my first response, and a few of my later ones, you would understand that I immediately mentioned that the main issue to the proposal would be figuring out some sort of growth limiting factor so that players don't grow endlessly. That's something to figure out.

I don't know if I can make this any more clear than that. Apologies if I sound direct but I feel like I've been making the same arguments over and over.

Researchers have to adjust everything each game.  Each game is a snapshot of the data at that point in time.  The issue here is in researchers being human, not in the way PA works.

I'm not "focusing" on anything.  It is possible for A Jamie Vardy in FM13 to become an FM17 Vardy, but not THE Vardy, because the researcher gave him a low PA.  You could argue this is the researcher's fault for not spotting the potential that the player had.  But you're effectively asking them, or the game, to read the future and allow that to happen in FM. I expect you also want that Leicester team to win the PL too?  And for all the transfers to happen the same?

Please do stop repeating yourself.

A player, cannot, out-perform, their potential.  That is a logical impossibility.  I cannot become better than the best I could possibly be. Does that make sense?  Your issue is not with PA, it is with researchers rating PA too low for players who have potential, but are unlikely to reach it.  I am not blaming any researcher for this because it is impossible to know.  However I expect the best solution here is for SI to continue to develop the player progression paths such that players can be given higher PAs with less and less chance of players with certain poor attributes reaching them.

10 hours ago, Cap'nRad said:

Again, no one's saying this that I can see.

What are the problems with the proposals of @Raptor Longe and myself (besides the one I already outlined, of course)? If you can identify them then we can make progress. 

Yes they are. Read the thread.

The issues are highlighted throughout this thread. Read the thread.

[santy001]

There's no justification for ever making a players potential just subject to random increases @Dorin

No matter the rationale, it just makes no sense to say "Well Usain Bolt was doing well for the last 12 months, he's bought some new equipment this year, now he can potentially do 100m in 8 seconds"

The game used to have what was its 'regen' day where you knew the date regens were coming and when to begin scouring clubs youngsters. 

All this would do is create "PA day" the day when thousands of players around the world magically are better than they were the day before. Think of how it would be in game, that on June 30th, little striker A is at the peak of his powers, he's a brilliant mid-table striker and maybe one of the top teams will look at bringing him in. However on July 1st, he's just become a champions league winning level striker and now clubs need to fawn over him.

 

 

[Dorin]

3 minutes ago, santy001 said:

There's no justification for ever making a players potential just subject to random increases @Dorin

No matter the rationale, it just makes no sense to say "Well Usain Bolt was doing well for the last 12 months, he's bought some new equipment this year, now he can potentially do 100m in 8 seconds"

The game used to have what was its 'regen' day where you knew the date regens were coming and when to begin scouring clubs youngsters. 

All this would do is create "PA day" the day when thousands of players around the world magically are better than they were the day before. Think of how it would be in game, that on June 30th, little striker A is at the peak of his powers, he's a brilliant mid-table striker and maybe one of the top teams will look at bringing him in. However on July 1st, he's just become a champions league winning level striker and now clubs need to fawn over him.

 

 

Nothing I suggest even gets to this scenario... first of all, there is no random increase, it's all based on measuring a lot of different factors that would point at the possibility of improving the CA past an existing limitation within the game.
Then there is no physical increase I suggest at.
Then the PA increase event is not triggered on some magical day, I started the post with the specific trigger and the conditions that have to be in place.

Then, tell me if I am wrong, but can you say someone can never get better (past a hardcapped limit) at the ability to:
1. ability to play in a team (teamwork)
2. ability to move without the ball 
3. leadership
4. anticipation
5. concentration
6. composure
7.positioning
8. decisions
9.. work rate
10. ability to pass the ball more accurately
11. ability to shoot better
12. ability to take better free kicks?
These are just some examples.
Are these learned or are these genetic (talent only)?


And nobody just becomes a CL striker suddenly, nothing I suggested even pointed at that (if you consider the formulas and what their results mean)... it just makes the system more dynamic in my opinion  

[HUNT3R]

21 minutes ago, Dorin said:

Nothing I suggest even gets to this scenario... first of all, there is no random increase, it's all based on measuring a lot of different factors that would point at the possibility of improving the CA past an existing limitation within the game.
Then there is no physical increase I suggest at.
Then the PA increase event is not triggered on some magical day, I started the post with the specific trigger and the conditions that have to be in place.

Then, tell me if I am wrong, but can you say someone can never get better (past a hardcapped limit) at the ability to:
1. ability to play in a team (teamwork)
2. ability to move without the ball 
3. leadership
4. anticipation
5. concentration
6. composure
7.positioning
8. decisions
9.. work rate
10. ability to pass the ball more accurately
11. ability to shoot better
12. ability to take better free kicks?
These are just some examples.
Are these learned or are these genetic (talent only)?


And nobody just becomes a CL striker suddenly, nothing I suggested even pointed at that (if you consider the formulas and what their results mean)... it just makes the system more dynamic in my opinion  

If someone gets better in any of those listed categories, it's their CA improving. All of us, in football or elsewhere, will get to a point where you cannot get better, no matter how much you practice. I don't know Usain Bolt, but he's not the #1 hardest training person in the world. There will be someone who trains harder - why aren't they faster? Why was someone like Darren Fletcher never as good as Messi? He's very professional, ambitious and determined. He was at Man Utd, who aren't lacking in facilities and quality coaches.

In your proposal, what's to stop an Ambitious, Professional and Determined player from not becoming world class? The player will just always get better and better.

It also doesn't allow for average players just being used perfectly in the right system, because they will just get better and better to eventually just end up being great players.

 

Apologies if I missed this in your proposal.

[santy001]

It's hard to respond in a meaningful way, because you're cutting and carving it up in a way that does indeed suit the argument you're trying to make but misrepresenting the whole system in place. 

No individual attribute has a hardcap, beyond the 20 limit. That would be a total catastrophe to remove because then you start to get these players who define a new "high standard" in something like passing, only to then in turn be eclipsed, and before you know it 30 becomes the new 20 and the whole thing is on its arse.

The spread of attributes is something that is used to create a player profile, a representation of that player. Jon Walters in the game, with some system that lets him start getting up to 15+ finishing, passing, dribbling, first touch just isn't Jon Walters. It's a fantasy player with the name Jon Walters. 

Using PA as a genetic limitation is also inclusive of their capacity to learn how to do these things better as well. The biggest thing in this thread is mis-representing PA, CA and attributes in order to advocate for change, but its still coming from a very flawed position. 

Encrypting PA and making it absolutely invisible to players is by far the best solution, no matter what editor is used etc. In the pre-game editor, give it a blank field for you to enter a new PA if you want, but removing the ability to observe PA kills dead all the predictable transfers from game to game, kills dead the predictable superstars, it means players no longer feel dissatisfied because their great performing player doesn't have a number as big as they'd like. 

As others have said, the acquisition of CA, the journey from start point to potential does need, and is getting, continual overhauls, tweaks, changes and nuances added to make that journey more bespoke from game to game, to create the pivots and fluctuations a players career can experience and allow for the influence of the right manager at the right time, or the wrong manager at the wrong time. Adding random power-up mechanics to the game, adding conditional power-ups to the game which basically boils down to "Score 10 goals get +1 finishing" is just not what a management game should be about. 

I'd even go as far as to say CA should be encrypted, it gets so frustrating to come into these discussions over a period of time and seeing it being so misrepresented or misunderstood. I don't blame people for not having that knowledge, but its a headline people focus on that is so irrelevant to the things they're seeing in game. 

[Cougar2010]

35 minutes ago, Dorin said:

Nothing I suggest even gets to this scenario... first of all, there is no random increase, it's all based on measuring a lot of different factors that would point at the possibility of improving the CA past an existing limitation within the game.
Then there is no physical increase I suggest at.
Then the PA increase event is not triggered on some magical day, I started the post with the specific trigger and the conditions that have to be in place.

Then, tell me if I am wrong, but can you say someone can never get better (past a hardcapped limit) at the ability to:
1. ability to play in a team (teamwork)
2. ability to move without the ball 
3. leadership
4. anticipation
5. concentration
6. composure
7.positioning
8. decisions
9.. work rate
10. ability to pass the ball more accurately
11. ability to shoot better
12. ability to take better free kicks?
These are just some examples.
Are these learned or are these genetic (talent only)?


And nobody just becomes a CL striker suddenly, nothing I suggested even pointed at that (if you consider the formulas and what their results mean)... it just makes the system more dynamic in my opinion  

 

As Hunt3r has said above these are all areas that are to do with CA progression, not PA.

I want to cover a couple of points with them though:

 

A) If a player works hard in one area he can improve it.

Yes in theory but the closer he gets to his limit the more work he has to put in to get a smaller % improvement eg its harder to go from 95% to 96% of your talent in an area than it is to go from say 10% to 50%.  On top of that if he doesn't put the work in on an area it will naturally decline/degrade.  So a player could work really hard and spend a lot of time improving his free kicks but because he does that for a marginal improvement his defending, corner taking, passing etc all degrade because he isn't putting the work in to keep them at the same level.  The PA system allows all of this to happen currently, the further away from his PA a player is the faster he improves and if you make his training unbalanced he will improve in the trained areas while degrading in the untrained areas.

 

B) The second point I want to mention although I think Santy already did earlier was that FM is representing professional footballers.  The difference between 1 & 20 in an attribute isn't the difference between your granny & Messi, its the difference between a well trained player/athlete who borders on being professional and Messi.  The average man in the street is going to score less than 1 in a lot of attributes.

[Dorin]

1 hour ago, HUNT3R said:

If someone gets better in any of those listed categories, it's their CA improving. All of us, in football or elsewhere, will get to a point where you cannot get better, no matter how much you practice. I don't know Usain Bolt, but he's not the #1 hardest training person in the world. There will be someone who trains harder - why aren't they faster? Why was someone like Darren Fletcher never as good as Messi? He's very professional, ambitious and determined. He was at Man Utd, who aren't lacking in facilities and quality coaches.

In your proposal, what's to stop an Ambitious, Professional and Determined player from not becoming world class? The player will just always get better and better.

It also doesn't allow for average players just being used perfectly in the right system, because they will just get better and better to eventually just end up being great players.

 

Apologies if I missed this in your proposal.

To answer your question and use your example of Darren Fletcher if he was a regen, in my proposed system what stops this is the paY formula. Here's how, maybe I didn't clarify more how I thought it could work because it's also confusing with all the variables:
Derren Fletcher - was produced my Man U youth, meaning that:

Y1 variable, which is set at youth intake date based on the club, training facilities, coaches, etc. let's say is 1.7 since it's a club with great setup.
Y2 variable, which is set as at the present date in the game, if this is simulated to say 3 years after the youth intake and is based on the current club facilities, training, coaching level and also his match experience. His current club setup is no different than that when he was created as a regen, and even with match experience and good results in matches it would result the current Y2 rating would be let's say 1.8  (if he had great performances and great match experience since the time his CA has already reached his PA, which is the first trigger event)
paY=Y2/Y1 means paY=1.8/1.7= 1.05 

Let's take the example of his PA rating of 140:
Current PA=140
paY=1.05
1.05x140=148
New potential= (148-140)x0.1 (this is his 10% increase limit)=0.8 PA increase.

The result in potential ability increase would be insignificant, really next to nothing.


Let's take the example of a player created at a low profile club who was then transferred to a bigger one and has reached his max CA at the bigger club. Same PA of 140:
Current PA=140
Y1=0.8 (keeping in mind the lower profile club environment and his potential ability set while he was generated in this environment that's not suitable to overachieve)
Y2=1.8 (current club environment of a successful club with top training/coaching, opposition, etc. Man Utd for example) 
paY=1.8/0.8=2.25
2.25x140=315
New potential=(315-140)x0.1 (this is his 10% increase limit of PA, I suggested between 5 and15)
So New Ppotential= 157.5 (increase of 17.5).
This might seem a bit much at first, but keep in mind that past this point, the environment and circumstances don't change anymore so there's no more significant growth, because Y2 cannot get much higher (if the max limit is 2) and Y1 then becomes 1.8 (his current club environment & circumstances), so paY would be at most if everything is perfect 2/1.8=1.11

Also, it doesn't mean that the player has to increase his CA later to reach the new PA, this would be a possibility, but subject to the same factors of increasing CA as exist now in the game.


The Ambitious, Professional and Determined part is a precondition to trigger the PA increase event and may add just a modifier to the calculations above, based on the paX variable.

PS: the exact math formulas can obviously be tweaked to keep the extreme scenarios in limit (if paY=2 / 0.1 for example), and the values reasonable, but hope the logic behind is starting to be more clear. 
PS2: also keep in mind that players at lower profile clubs would very rarely have an initial PA of 140, usually will be much lower.

[Maaka]

But the main issue here is that the POTENTIAL ability should not change, his potential ability hasn't been changed just because he's suddenly at a bigger club, the potential was always there.
What the transfer to the bigger club means, is that he's more likely to reach his potential (or getting closer to it) than he was at the smaller club, but his potential will always be the same. He can of course become a better player than what he'd become had he stayed at the smaller club, but his potential was the same back then.

[HUNT3R]

59 minutes ago, Dorin said:

To answer your question and use your example of Darren Fletcher if he was a regen, in my proposed system what stops this is the paY formula. Here's how, maybe I didn't clarify more how I thought it could work because it's also confusing with all the variables:
Derren Fletcher - was produced my Man U youth, meaning that:

Y1 variable, which is set at youth intake date based on the club, training facilities, coaches, etc. let's say is 1.7 since it's a club with great setup.
Y2 variable, which is set as at the present date in the game, if this is simulated to say 3 years after the youth intake and is based on the current club facilities, training, coaching level and also his match experience. His current club setup is no different than that when he was created as a regen, and even with match experience and good results in matches it would result the current Y2 rating would be let's say 1.8  (if he had great performances and great match experience since the time his CA has already reached his PA, which is the first trigger event)
paY=Y2/Y1 means paY=1.8/1.7= 1.05 

Let's take the example of his PA rating of 140:
Current PA=140
paY=1.05
1.05x140=148
New potential= (148-140)x0.1 (this is his 10% increase limit)=0.8 PA increase.

The result in potential ability increase would be insignificant, really next to nothing.

So Darren Fletcher is handicapped because he started at a great club with great facilities? If the same Darren Fletcher started in League 2 at a club with poor coaches and facilities, he'd get a bigger bonus to PA if Man Utd snapped him up? This doesn't make much sense. A player doesn't suddenly get more potential because a bigger team buys you.

 

Quote

Let's take the example of a player created at a low profile club who was then transferred to a bigger one and has reached his max CA at the bigger club. Same PA of 140:
Current PA=140
Y1=0.8 (keeping in mind the lower profile club environment and his potential ability set while he was generated in this environment that's not suitable to overachieve)
Y2=1.8 (current club environment of a successful club with top training/coaching, opposition, etc. Man Utd for example) 
paY=1.8/0.8=2.25
2.25x140=315
New potential=(315-140)x0.1 (this is his 10% increase limit of PA, I suggested between 5 and15)
So New Ppotential= 157.5 (increase of 17.5).
This might seem a bit much at first, but keep in mind that past this point, the environment and circumstances don't change anymore so there's no more significant growth, because Y2 cannot get much higher (if the max limit is 2) and Y1 then becomes 1.8 (his current club environment & circumstances), so paY would be at most if everything is perfect 2/1.8=1.11

Also, it doesn't mean that the player has to increase his CA later to reach the new PA, this would be a possibility, but subject to the same factors of increasing CA as exist now in the game.

So in just a season, you can bump PA by almost 20 points. In another season, it could be as high as 175 PA.

MASSIVELY open to exploits, for one thing. I would be lower league prospects like crazy.

[Dorin]

1 hour ago, Cougar2010 said:

 

As Hunt3r has said above these are all areas that are to do with CA progression, not PA.

I want to cover a couple of points with them though:

 

A) If a player works hard in one area he can improve it.

Yes in theory but the closer he gets to his limit the more work he has to put in to get a smaller % improvement eg its harder to go from 95% to 96% of your talent in an area than it is to go from say 10% to 50%.  On top of that if he doesn't put the work in on an area it will naturally decline/degrade.  So a player could work really hard and spend a lot of time improving his free kicks but because he does that for a marginal improvement his defending, corner taking, passing etc all degrade because he isn't putting the work in to keep them at the same level.  The PA system allows all of this to happen currently, the further away from his PA a player is the faster he improves and if you make his training unbalanced he will improve in the trained areas while degrading in the untrained areas.

 

B) The second point I want to mention although I think Santy already did earlier was that FM is representing professional footballers.  The difference between 1 & 20 in an attribute isn't the difference between your granny & Messi, its the difference between a well trained player/athlete who borders on being professional and Messi.  The average man in the street is going to score less than 1 in a lot of attributes.

I perfectly agree with all these points. CA progression is definitely a huge part of all this. What is really frustrating for me now is when I have a player who is 23 for example and he is playing well, doing well, training well, but his CA stays the same and nothing will really change in the next 5-7 years if I keep him in the team, starting against better opponents and doing well, going from championship team to CL games, getting better training & coaching, etc. I think that's a bit unrealistic. He will only decline as a result of injury or overtraining, and then just came back to the same limit. There should be a way to avoid this kind if rigidity in development I think. 

[Dorin]

17 minutes ago, HUNT3R said:

So Darren Fletcher is handicapped because he started at a great club with great facilities? If the same Darren Fletcher started in League 2 at a club with poor coaches and facilities, he'd get a bigger bonus to PA if Man Utd snapped him up?

No, Darren Fletcher who started at Man U got his potential set higher in the start because of his club. If Darren Fletcher would have started at a league 2 club I believe his initial potential would have been set lower because of the club ratings. 

"MASSIVELY open to exploits, for one thing. I would be lower league prospects like crazy."

I don't like exploits either, but I believe with a bit of playing around with the equations used for different scenarios these can be avoided in the majority of instances, taking into account all the different possible scenarios. If I had the time I would look at different scenarios, but that's not the point here. The idea is to suggest an alternative which I did, with flaws I admit, but it can be taken as a suggestion and worked with to get to a better working model than what FM currently uses for player development once CA is reached. Whether it's rethinking CA progression only, or something else. I'm all for removing some of the current limitations.

[santy001]

8 minutes ago, Dorin said:

No, Darren Fletcher who started at Man U got his potential set higher in the start because of his club. If Darren Fletcher would have started at a league 2 club I believe his initial potential would have been set lower because of the club ratings. 

Here's the fundamental issue though. 

Being at Man Utd plays no part in assigning the PA of Darren Fletcher. 

You're working under the assumption that every player who comes through at Man Utd, by virtue of being at Man Utd, is higher than a player coming through at Wycombe etc. Which isn't how it works, and therefore is why what you're suggesting is in no way appropriate.

[Dorin]

1 minute ago, santy001 said:

Here's the fundamental issue though. 

Being at Man Utd plays no part in assigning the PA of Darren Fletcher. 

You're working under the assumption that every player who comes through at Man Utd, by virtue of being at Man Utd, is higher than a player coming through at Wycombe etc. Which isn't how it works, and therefore is why what you're suggesting is in no way appropriate.

I think that's pretty much how the game world develops once you simulate 5-10 years further.

[Cougar2010]

8 minutes ago, Dorin said:

I perfectly agree with all these points. CA progression is definitely a huge part of all this. What is really frustrating for me now is when I have a player who is 23 for example and he is playing well, doing well, training well, but his CA stays the same and nothing will really change in the next 5-7 years if I keep him in the team, starting against better opponents and doing well, going from championship team to CL games, getting better training & coaching, etc. I think that's a bit unrealistic. He will only decline as a result of injury or overtraining, and then just came back to the same limit. There should be a way to avoid this kind if rigidity in development I think. 

But there are thousands of examples of this IRL.

There are players at every level who for no obvious reason are very similar to how they were 3/5/7 years ago.

You just have to look at the players who get released from the big clubs to see examples.  Fletcher & Cleverley have been mentioned but there are so many others every season.

If the big clubs can't produce a number of Messi's every year despite the staff, facilities and good attitudes of the players then clearly something else is in play which is what PA represents.

[Dorin]

Here is the summary of my suggestion, put visually (Static PA= current system being used in FM). Keep in mind I am not suggesting exponential growth, since the big jump in better circumstances cannot happen over and over again.

Edit: to be accurate, the CA should equal the PA at the date of slight increase in PA (forgot to set this correctly in the graph).

If the ratings don't change (club facilities, coaching level, training level, match exp increase) there is nothing to increase in PA as well. This also keeps players from getting their PAs higher and higher if no external circumstances change (as is the case of Darren Fletcher not increasing his PA... it shouldn't, nothing else has changed for him, just match experience increase alone wouldn't be enough).

[ham_aka_stam]

1 hour ago, Dorin said:

I perfectly agree with all these points. CA progression is definitely a huge part of all this. What is really frustrating for me now is when I have a player who is 23 for example and he is playing well, doing well, training well, but his CA stays the same and nothing will really change in the next 5-7 years if I keep him in the team, starting against better opponents and doing well, going from championship team to CL games, getting better training & coaching, etc. I think that's a bit unrealistic. He will only decline as a result of injury or overtraining, and then just came back to the same limit. There should be a way to avoid this kind if rigidity in development I think. 

You're wrong.

• Firstly at 23 you'd have to be pretty lucky to have got him to his CA.
• Secondly things should change in 7 years, you'll be moulding his attributes
• Additionally his CA will increase when he plays well,
• And it'll take hits if he gets injured.
• He may not even be able to recover that lost CA

So the problem you're trying to fix simply doesn't exist

1 hour ago, Dorin said:

No, Darren Fletcher who started at Man U got his potential set higher in the start because of his club. If Darren Fletcher would have started at a league 2 club I believe his initial potential would have been set lower because of the club ratings. 

I'm not a researcher, but that's certainly not the way I would expect them to be working.