[Suggestion]"Floating" potential ability

[HUNT3R]

Just now, doucong said:

Yes I know that, the CA is a combination of attribute, calculate by some secret algorithm.  Different by position etc. Determination/Aggressive/etc is not calculated in. What I am trying to say is, no matter how you arrange the attributes within CA50 limit, you cannot design a player that can do 15 goals and 10 assists. If you could, there must in history such a player and  in the database his CA is 50. And we all know that in database there is no real player how did that before been marked PA50(his peak). So for the same reason if a player in the game did that, he should not be mark as 50 as well 

So you've answered your own question. A 50 CA/PA player cannot do 15/10, so the entire debate is pointless.

If he could, then great. He's in an absolutely perfect system with the perfect teammates. Raise his CA because of that doesn't make sense and will make him improve, which improves his performances even more. Again, pointless as when will it end?

[doucong]

9 minutes ago, HUNT3R said:

Then you don't know what consistency is. It doesn't change from season to season, but game to game.

Yeah then I may misunderstand the consistency. So how to present a player whom could do very good in one season but pretty poor for the other seasons in the database? 

[HUNT3R]

Just now, doucong said:

Yeah then I may misunderstand the consistency. So how to present a player whom could do very good in one season but pretty poor for the other seasons in the database? 

There's a lot more to it than just CA and attributes. Everyone keeps saying it in this thread? Tactical system, role and teammates matter too.

Vardy isn't poorer player than he was a year ago. Things change.

[herne79]

FYI, a player with 20 Consistency (hidden attribute) doesn't mean he's always going to play at his maximum ability.  He'll just do it a bit more often than someone with lower Consistency.

[doucong]

18 minutes ago, HUNT3R said:

So you've answered your own question. A 50 CA/PA player cannot do 15/10, so the entire debate is pointless.

If he could, then great. He's in an absolutely perfect system with the perfect teammates. Raise his CA because of that doesn't make sense and will make him improve, which improves his performances even more. Again, pointless as when will it end?

Make it simple, is it possible a CAPA 50 player do 15/10 in a PL season in game?

[herne79]

25 minutes ago, doucong said:

Make it simple, is it possible a CAPA 50 player do 15/10 in a PL season in game?

ok I misread, I thought you'd actually got an example of a 50CAPA player doing this in your game.

Is it possible?  Who knows, you'd have to try it - you'd have to edit a player accordingly though to give him the relevant attributes within the CA limit and then set up a tactical system that suits him.  It's pretty extreme though but that's the whole point: focus on attributes and the needs for your tactical system, not CA/PA.

[Cougar2010]

1 hour ago, doucong said:

Make it simple, is it possible a CAPA 50 player do 15/10 in a PL season in game?

Probably not which makes the discussion even more bizarre.

Unless you can provide an example then its all hyperthetical, the chance of it ever happening in the game is so rare that its not worth discussing.

[doucong]

6 minutes ago, herne79 said:

ok I misread, I thought you'd actually got an example of a 50CAPA player doing this in your game.

Is it possible?  Who knows, you'd have to try it - you'd have to edit a player accordingly though to give him the relevant attributes within the CA limit and then set up a tactical system that suits him.  It's pretty extreme though but that's the whole point: focus on attributes and the needs for your tactical system, not CA/PA.

Thank you for that.

When I ask that question I assume the answer is possible. And my next question will be have ever a PL player with 15/10 marked as CAPA50? I assume the answer is no, no first team PL player have CA50. That is where my discussion come from.

Well I never have a CAPA50 player did 15/10, but I do have a CAPA~70 player whom did very well in PL and I believe no real player with similar stats ever been marked even close as low as CAPA70 in FM database  (I remember is Hurst in my Hednesford save). My initial thought is if a player in real life perform as such will be mark CA120 in the game for example, he should have something similar(like PA for veteran indicate his full ability at his best time, my understanding is CA means his full ability today) and of course with some game mechanism prevent spiral effect(like consistency or other mechanism). Increase CA will lead to spiral effect is not the excuse for make such a player stay with CAPA50. And yes I know CAPA is only one of many part deciding performance, I just use the term CAPA to represent full ability here, we can change it the conversation to stars: if a player did 15/10 should he have at least 3 star instead of half?

Anyway no matter we agree each other or not, I feel we have both shown our opinion, that should be enough.

Btw I am quite surprise some moderator's discourage tongue here, and even comment such discussion pointless.

[HUNT3R]

12 minutes ago, doucong said:

Btw I am quite surprise some moderator's discourage tongue here, and even comment such discussion pointless.

Again, you have the wrong end of the stick. I never said the discussion is pointless. YOU said you can't make a 20/10 player with 50 CA/PA. If that's the case, why discuss a hypothetical player that cannot exist?

The fact that we're on page 2 already proves that the discussion is a discussion.

[Weston]

2 hours ago, herne79 said:

FYI, a player with 20 Consistency (hidden attribute) doesn't mean he's always going to play at his maximum ability.  He'll just do it a bit more often than someone with lower Consistency.

So consistency means only how often a player will perform to his CA? Consistent players cannot consistently over or underperform, they simply will play closer to the CA more often? I wondered if inconsistent players would sometimes overperform as well as underperform, but after reading this it seems that just means he will rarely reach as high as his CA? And CA, then, is not the average for the player, but the max?

[Weston]

7 hours ago, Cougar2010 said:

Whilst I disagree with PA changing once a save has started I have many times over the years floated the idea of an option for totally random PAs.

FM already has a system mostly in place for doing this.  Instead of using -1 to -9 for the PA in the database using 0 will give you a totally random PA which changes from save to save so a player could be a 200PA in one save and 120PA in the next.

The way I see it working is that you have it as a tick option on the setup screen and then the real players would be given a random PA above their current CA level when a save is setup. 

Its something that has/is used by other management games in the genre - Pro Cycling Manager by Cyanide for instance is one that has had it in for a number of years as has some of the Grey Dog Software titles (Called a dynasty rating).

So every single save I play will have the real life players with the exact same PA? Obviously different in-game experiences will alter the outcomes, but there is not currently the random fluctuation you talk about?

[Cougar2010]

49 minutes ago, Weston said:

So every single save I play will have the real life players with the exact same PA? Obviously different in-game experiences will alter the outcomes, but there is not currently the random fluctuation you talk about?

Currently all players are either given a fixed PA or a random PA within a small 30 point range.

The random PAs are normally given to younger real players but they are still given a PA within a range so the random isn't as much as it could be.  A player with a high range will always be a top class player, just in some saves he will be Messi and in others he'll just be Harry Kane there is no chance of him totally failing unless he picks up a career ending injury.

 

[santy001]

I think this is an idea we've all had at one point or another, and as we then come to understand more about the systems behind it a lot move on to the developmental, or means through which players acquire CA and the apparent ease through which this happens.

There's an awful lot more to this whole element of the game than perhaps people realise in the hidden mental attributes, and even some of the visible ones. There's so many elements that come into play on this topic.

I don't expect the average player to be as able to precisely hone in on a great prospect like some of the more seasoned users of this forum or researchers, but it falls into a predictable pattern.

Find a young talented player, look at his profile and see a few key attributes high - lets say determination and work rate. 

Determination doesn't strictly impact upon his development, but it works to your benefit more in matches. Now if this player also has reasonably high professionalism and consistency then he's going to be improving rapidly and performing well with that improvement. Particularly if most of his CA acquisition is coming through technical development. If he has high ambition this further amplifies the effect, when the "storm" of right attributes are thrown together even at around 11/12's or above players become rapid developers. 

I don't think FM quite has any modelling for those who peak early and then decline, those who get a lot of money and lose their ambition or become less professional (in the worst case Michael Johnson who lost his desire for football entirely), nor is there any modelling of the opposite, for those who start off quite wasteful but become far more refined as their career progresses (I wouldn't necessarily say Vardy applies here as I don't know what he was like in his younger days, but Jon Walters has said several times during his younger days he just sort of assumed success would come and before he knew it was down in League 2).

Players PA's in a fashion do sort of "float" though, not in the sense intended in this topic but a 190PA player is almost certainly unlikely to fulfil it if he's playing in League 2 with a club that has poor training facilities. By moving from the lesser equipped clubs in the game, players are removing the shackles that impose an invisible, unknowable limit on the level their CA can grow into their PA. A lot of it is realising these mechanics are in the game, and they aren't meant to be clearly defined, known entities, because they aren't such in real life. Coming back to the case of the 190PA player, if you could forcibly keep him shackled to a league 2 club that never goes anywhere and has very poor training facilities, I doubt he'd get much above 100CA. 

This also applies in reverse, the actual PA a player has is always fixed. The situation a player finds himself in within the game provides the float. The only thing is, the float mechanic is brilliant for the games purpose because you cannot in any way determine what it is or how to game it. Better coaches, better facilities and a better level of football just help development - that is all you can know. 

[noikeee]

14 hours ago, doucong said:

Well I never have a CAPA50 player did 15/10, but I do have a CAPA~70 player whom did very well in PL and I believe no real player with similar stats ever been marked even close as low as CAPA70 in FM database  (I remember is Hurst in my Hednesford save). My initial thought is if a player in real life perform as such will be mark CA120 in the game for example, he should have something similar(like PA for veteran indicate his full ability at his best time, my understanding is CA means his full ability today) and of course with some game mechanism prevent spiral effect(like consistency or other mechanism). Increase CA will lead to spiral effect is not the excuse for make such a player stay with CAPA50. And yes I know CAPA is only one of many part deciding performance, I just use the term CAPA to represent full ability here, we can change it the conversation to stars: if a player did 15/10 should he have at least 3 star instead of half?

A player with a CA of 70 shouldn't be performing well in the Premier League. That is really something quite extreme. But this isn't a problem with the development of CA/PA, it's a problem somewhere else, likely of CA not being linked to his attributes correctly - clearly his attributes make him able to perform at a level that we perceive as a higher level than a CA of 70.

I suppose if you have a very extreme system tailored to get the best out of very specific traits of a player, you could get good performances out of very average players, so these things can happen, but maybe shouldn't to that extent.

As for staff rating stars, they already take into account current form (ie, current average ratings) to a large degree. I see them fluctuating wildly depending on player form in my FM15 save, I recall the same happening when I was playing FM16. If you make it even more dependent on form than actual CA, then your staff opinion stops having any relevance, as you could just look at average ratings then.

[Weston]

7 hours ago, noikeee said:

As for staff rating stars, they already take into account current form (ie, current average ratings) to a large degree. I see them fluctuating wildly depending on player form in my FM15 save, I recall the same happening when I was playing FM16. If you make it even more dependent on form than actual CA, then your staff opinion stops having any relevance, as you could just look at average ratings then.

Really? I didn't know this... to what extent is "a large degree"? I've also always been bothered by stars dipping in off season as players are out of form, because I always viewed stars as ability, not temporary form which I used ratings and fitness to determine. That dip just royally confuses my transfer activity when deciding who to upgrade in between seasons.

[Cougar2010]

13 minutes ago, Weston said:

Really? I didn't know this... to what extent is "a large degree"? I've also always been bothered by stars dipping in off season as players are out of form, because I always viewed stars as ability, not temporary form which I used ratings and fitness to determine. That dip just royally confuses my transfer activity when deciding who to upgrade in between seasons.

0.5* maybe tops.

[noikeee]

50 minutes ago, Cougar2010 said:

0.5* maybe tops.

I'm playing FM15 but on this version it's definitely more than that. I had a player that had a superb season last year, so he was rated 4* CA at the beginning of this season. After 3 months, he's been average so has been dropped to 2.5* CA. So for my staff he's lost 1.5* in 3 months. Notice meanwhile I've barely made any signings at all that could change the CA my players are being compared to; and this player hasn't been injured or declining due to age or anything.

1 hour ago, Weston said:

Really? I didn't know this... to what extent is "a large degree"? I've also always been bothered by stars dipping in off season as players are out of form, because I always viewed stars as ability, not temporary form which I used ratings and fitness to determine. That dip just royally confuses my transfer activity when deciding who to upgrade in between seasons.

It's meant to royally confuse. Players in real life are wildly overrated or underrated depending if they've played well lately or not. If staff just told you their correct CA all the time it'd be too easy. People are wrong in football all the time, this is what this is meant to simulate.

[Weston]

7 minutes ago, noikeee said:

I'm playing FM15 but on this version it's definitely more than that. I had a player that had a superb season last year, so he was rated 4* CA at the beginning of this season. After 3 months, he's been average so has been dropped to 2.5* CA. So for my staff he's lost 1.5* in 3 months. Notice meanwhile I've barely made any signings at all that could change the CA my players are being compared to; and this player hasn't been injured or declining due to age or anything.

It's meant to royally confuse. Players in real life are wildly overrated or underrated depending if they've played well lately or not. If staff just told you their correct CA all the time it'd be too easy. People are wrong in football all the time, this is what this is meant to simulate.

I've also seen some pretty wild fluctuations that could be at least in part due to form. And yeah, I mean, the thing that bothers me about stars is not that they're confusing and fallible but the fact that the game is so unclear about in which ways they are. I wouldn't be bothered by them adjusting due to a realistic fog of war of certain factors if that was explained going into things.

[Cougar2010]

12 minutes ago, noikeee said:

I'm playing FM15 but on this version it's definitely more than that. I had a player that had a superb season last year, so he was rated 4* CA at the beginning of this season. After 3 months, he's been average so has been dropped to 2.5* CA. So for my staff he's lost 1.5* in 3 months. Notice meanwhile I've barely made any signings at all that could change the CA my players are being compared to; and this player hasn't been injured or declining due to age or anything.

I haven't seen anything that wild although I'll say I haven't been looking either.

I keep a spreadsheet of my players which I update about 3 times a season and I've never seen CA stars change that much, its only ever been 0.5* at a time maybe 1* on a new signing as the staff get to know him.

I'm not saying your wrong either noikeee as you come across as someone who has good attention to detail so maybe something for us all to look out for to get a better idea of how it works.

[Northernpilgrim]

A player's PA is always fixed, but not always known. That's the key point. Vardy in hindsight, always had the potential to be a world class striker (as I think most would agree he now is). Why he didn't make it earlier is down to a number of factors. Yes, he developed later than some, so was dismissed by league clubs, and doubtless became disheartened by that. So he ends up dropping into non-league, playing and training part-time. He therefore doesn't develop as quickly as he might have, but whilst playing for Stocksbridge, he performs better than anyone else at that level, so teams from higher up start to notice. In 2010, still only 23, he joins Halifax, in the Northern Premier League. He again outperforms his peers, so teams higher up take notice, and he goes to Conference club Fleetwood. Whilst at Fleetwood he scores 31 in 36 games in the league, which is going to get the attention of League 2 and League 1 clubs, but also scores the goals that knock League 1 club Wycombe out of the FA Cup. That's going to attract attention from bigger clubs too, and sure enough, in 2012, now 25, Leicester take a punt on him. He isn't an instant success, but still manages 4 in 26 in his first season. His next season is where we see him start to show what he can do. In his second season in the Championship, he manages 16 in 37. His first PL season isn't fantastic, with 5 in 34, but 24 in 36 last season was brilliant. I know that many are familiar with the Vardy story, but putting it in these terms shows that the potential was always there, but it took a combination of the right club, right manager and right tactics to get the best out of him. FM, being a stats and numbers based entity is never going to replicate that accurately, although I guess it could be done to an extent by using a combination of professionalism, p.a and c.a, and determination. The best way to replicate it would be that his PA is very high, his determination and professionalism were probably not that high until the last few seasons. Maybe increasing the amount these attributes can change by would allow or the 'Vardy effect'?

[Firecracker048]

Bec

On 10/7/2016 at 01:03, HUNT3R said:

If he IS already over-performing, why does he need more CA? He's already playing better than everyone thought he would, so he is already playing as if he's had a CA increase.

Because you can't sell a player who over performs with such a low CA

[Firecracker048]

5 hours ago, Northernpilgrim said:

A player's PA is always fixed, but not always known. That's the key point. Vardy in hindsight, always had the potential to be a world class striker (as I think most would agree he now is). Why he didn't make it earlier is down to a number of factors. Yes, he developed later than some, so was dismissed by league clubs, and doubtless became disheartened by that. So he ends up dropping into non-league, playing and training part-time. He therefore doesn't develop as quickly as he might have, but whilst playing for Stocksbridge, he performs better than anyone else at that level, so teams from higher up start to notice. In 2010, still only 23, he joins Halifax, in the Northern Premier League. He again outperforms his peers, so teams higher up take notice, and he goes to Conference club Fleetwood. Whilst at Fleetwood he scores 31 in 36 games in the league, which is going to get the attention of League 2 and League 1 clubs, but also scores the goals that knock League 1 club Wycombe out of the FA Cup. That's going to attract attention from bigger clubs too, and sure enough, in 2012, now 25, Leicester take a punt on him. He isn't an instant success, but still manages 4 in 26 in his first season. His next season is where we see him start to show what he can do. In his second season in the Championship, he manages 16 in 37. His first PL season isn't fantastic, with 5 in 34, but 24 in 36 last season was brilliant. I know that many are familiar with the Vardy story, but putting it in these terms shows that the potential was always there, but it took a combination of the right club, right manager and right tactics to get the best out of him. FM, being a stats and numbers based entity is never going to replicate that accurately, although I guess it could be done to an extent by using a combination of professionalism, p.a and c.a, and determination. The best way to replicate it would be that his PA is very high, his determination and professionalism were probably not that high until the last few seasons. Maybe increasing the amount these attributes can change by would allow or the 'Vardy effect'?

I had a "Vardy effect", as you call it in my game with Tony Lee. The only problem was he did not have the "potential" to play in higher leagues, even though he was out-performing those who were well above his CA over the course of 3 seasons. The problem I had was that someone like that SHOULD get the attention of higher league clubs, but never did, so I never got to cash in his success(until I fired up the editor and made him good enough to sell for 2 mil). Maybe what I'm looking for is a temp "boost" in perceived CA, even if they never actually make it to there. I feel like someone like that would get attention from higher clubs, even If they would burn out quickly

[HUNT3R]

4 hours ago, Firecracker048 said:

Bec

Because you can't sell a player who over performs with such a low CA

You can, but we've already established that such a low CA can't perform to that level.

 

And the reason you can, is because scouts take form into consideration as well, so this player will be rated higher because of the good form.

[santy001]

9 hours ago, Firecracker048 said:

I had a "Vardy effect", as you call it in my game with Tony Lee. The only problem was he did not have the "potential" to play in higher leagues, even though he was out-performing those who were well above his CA over the course of 3 seasons. The problem I had was that someone like that SHOULD get the attention of higher league clubs, but never did, so I never got to cash in his success(until I fired up the editor and made him good enough to sell for 2 mil). Maybe what I'm looking for is a temp "boost" in perceived CA, even if they never actually make it to there. I feel like someone like that would get attention from higher clubs, even If they would burn out quickly

This is in the game, I'll recount an example from my own game experience on FM16. 

Marc Muniesa is a defender rated around 135CA, now due to injuries on FM16 at Stoke he became the main centreback pairing and performed excellently all season long. Now he has quite high injury proneness, but this wasn't an issue he had a favourable season in terms of injuries and performed excellently, arguably much better than you might have expected a 135CA defender to play. Chelsea paid £13.5m at the end of the first season to sign him from me. Sustained good form leads to that and ultimately by the end of the next season Chelsea were selling him for a substantial discount owing to the fact he wasn't as good as the scouting had earlier identified. 

Lower down the leagues its often the case that the clubs either don't have the money to take the gamble, or your database size can have a tremendous influence here. If the database is too large you can have a saturation of players at the top that filter down a bit and so there is no where near as much upwards movement of players. On the flip side, if your database is too small then there tends to be too much upwards movement. It's an element most may not consider when looking at this side of the game, but again drawing on my experiences at playing with Stoke on FM16. If I start with a small database, I can make close to £100m in sales in the first season. If I start with a large database unless I'm selling a key first team player who is performing really well I'd be lucky to be able to get around £20m. 

[Weston]

4 hours ago, santy001 said:

Lower down the leagues its often the case that the clubs either don't have the money to take the gamble, or your database size can have a tremendous influence here. If the database is too large you can have a saturation of players at the top that filter down a bit and so there is no where near as much upwards movement of players. On the flip side, if your database is too small then there tends to be too much upwards movement. It's an element most may not consider when looking at this side of the game, but again drawing on my experiences at playing with Stoke on FM16. If I start with a small database, I can make close to £100m in sales in the first season. If I start with a large database unless I'm selling a key first team player who is performing really well I'd be lucky to be able to get around £20m. 

I'd never really considered that, that's really interesting. Is it not over-saturation though, per se, to have a large database if that's more realistic?

[Yankee112DK]

The main issue is that so many FM'ers are WAY too obsessed with CA/PA thinking that high CA = instant win. It's not like that.

Leicester didn't have the highest CA value of the Premier League, nor should they. They managed to set together just the right players, the right coach and hit just the right season. They hit a streak and were very clever tactically and strategically and simply rode the wave.

A player like Jamie Vardy is not a top Premier League striker per se. He just had a VERY good pre-season, was on a very hard-working and focused team and hit a streak. And most importantly, his coach was very clever about playing to the strengths of both Mahrez and Vardy in conjunction with other teams' shortcomings.

[santy001]

On 15/10/2016 at 16:26, Weston said:

I'd never really considered that, that's really interesting. Is it not over-saturation though, per se, to have a large database if that's more realistic?

I don't fully remember all the details of what is the appropriate sized database vs leagues active, but I think its just leaving it as standard on the medium size which is mostly in that ballpark. 

Taking a step back from it all and looking at the bigger picture it always has to be remembered FM is going to have technical limitations on what it can simulate. Most don't have all the additional leagues actively being processed even if they load them, because that does make the game far more resource hungry. If they aren't run in full detail then the league development & player development within those leagues is a little less likely to go as intended. 

While not strictly speaking the main jist of this thread, its one of the big factors in how the transfer market plays out in a persons save. It does tie in with player development though. 

Coming back to the topic at hand though, I think this is something that will be resolved organically as the years come to pass. I think the hidden mental attributes will be what resolves it though, they need to be a bit more fluid over time and influenced by factors in the game. However, it would begin to get incredibly complicated and would certainly be quite buggy or exploitable by the playerbase in early years. 

In reality some players owe their careers - or at least the peak of their careers - to a certain manager and the way he treated them, or to being the star player at a club. The managers personality gets them to change their mindset towards the game and get more from them. Or the fans adulation brings them around to get them more focused. 

Tony Pulis is a manager who excelled at this at Stoke for quite a while. He gave Ricardo Fuller the freedom to be a mercurial player, when he needed some time off it was given, when he got a bit of out of hand exceptions to the rule were made. The fans embraced him despite this almost as a lead from Pulis doing the same and he thrived in it despite looking like at the point he signed for Stoke he was probably on his way to being written off as a league 1 journeyman. The biggest thing would be to allow manager styles to both have positive and negative impacts upon players careers. As much as Pulis was great for the likes of Fuller, Liam Lawrence, Matthew Etherington he was the nail in the coffin for some, Dave Kitson wasn't a one time working class lad and Pulis and him just never particularly got on. Even when Kitson did eventually get on a good streak Pulis just dropped him and seemingly crushed his confidence as a footballer. 

I think this area of the game is where we will eventually come to see players develop as late bloomers or have brief periods in their career in which they're phenomenal and then drop back down a bit. The biggest risk with something like this though is the cookie-cutter, theorycrafting type element it could bring from the playerbase. I also play WoW on the side and I've seen how a determined playerbase analysing can brutalise the game to get definitive "this is the right approach for maximum rewards" and once a playerbase begins doing that no game currently can withstand that and not become a very transparent issue. The problem if something like that happened to FM is it would kill all the fun in the game, except for those interested in the race to the top of the score leaderboards. 

[Urotsukidoji]

Most of the players who make their way from the lower leagues to the top, are usually young players who at first at premiership league academies, and told they have no future at the club, and so they have dropped down the divisions.

In FM terms, as a lower league manager you can plough the top teams released or unwanted players, they become stars at your small team, and you either sell them, or they are with you as you move up the league.

 

[Maaka]

As stated before, the PA should be fixed, as the potential cannot be higher or lower than what it actually is. The idea of having a change of the PA at the start of each game as mentioned, is interesting (if implemented the right way, and with certain limits).

However, I think that the main issue here, is the fact that it's "too easy" to identify high PA players, which makes those late bloomers too rare. As it seems now, all high PA players will be picked up by major teams pretty early on in their careers, and then either be good, or fall right through. If anything, the PA should be even more "obscure" in-game, to ascertain that neither human- or computer-controlled teams can find them all too easy. Then more Vardys will probably arise.

[XaW]

Now, I've read throughout this post, and I've earlier read a lot of post regarding CA/PA, late bloomers, changing PA, etc, but I find the system mostly ok, though with a few lacks here and there.

Still, one point I want to make regarding this is the following. Make it much much harder for players to reach their potential, and up the potential of most players by quite a bit. This is possible solve a couple of the issues at hand here. This would enable a lot of players to never reach their potential due to one or more issues with playing time, off pitch events, facilities, lack of professionalism, etc, etc. But it would also open up for the "late bloomer" effect. Players who find their confidence in the lower leagues and get a CA boost. Some might have success in the higher leagues like Vardy at Leicester, while others could struggle to handle the higher level of opposition, like Beckford at Everton for example.

This could mean potential world beaters would not be great in one game and falter away like Adu or Samba, while in other games they could be the new Messi if the circumstances would allow it. This will also give managers some new things to cope with, in that the potential of the players will be there, but the skill might not follow. As someone who play a lot of games that focus on developing youth players, this would give me an extra challenge, as I would not know if the player would cut it, even though he has the potential. This will lead to managers, both player and AI, failing to develop youngsters who then might get a "new spring" at another club who are willing to take the chance.

For me this would be a more realistic way of handling the CA/PA way, and solve quite a bit of the issues that have been discussed here. Of course this would require a whole lot of testing in order to get the balance right, so it might not be enough return of investment.

Anyway, my thoughts regarding the whole CA/PA discussion

[enigmatic]

On 10/18/2016 at 08:00, XaW said:

Still, one point I want to make regarding this is the following. Make it much much harder for players to reach their potential, and up the potential of most players by quite a bit. This is possible solve a couple of the issues at hand here. This would enable a lot of players to never reach their potential due to one or more issues with playing time, off pitch events, facilities, lack of professionalism, etc, etc. But it would also open up for the "late bloomer" effect. Players who find their confidence in the lower leagues and get a CA boost. Some might have success in the higher leagues like Vardy at Leicester, while others could struggle to handle the higher level of opposition, like Beckford at Everton for example.

I think to an extent SI do that with regens: I haven't systematically studied it but I'm pretty sure there are a lot more players at the 180-200PA range after several seasons of a save than there are in a fresh save from the occasional messing around with scout tools, and I'm equally certain their personalities are pretty random and therefore unlikely to be as dedicated as Messi and Ronaldo. Having more than one candidate Messi for the next generation and having a bit more chance of getting regens that'll be good for your level if you want to work really hard at bringing them through (assuming the same distributional effect applies lower down the scale) is probably a good thing, even at the cost of a little imbalance. After all, it's what makes Youth Challenges possible.

But when it comes to real players, it's a simulation trying to make the best guess at what their actual career might look like. So upping all the  the real player PAs and making it so that players usually fall significantly short of them unless they have the setup and make the right decisions and selections will tends to result in far more unrealistic player careers than the current system. Sure, earlier versions of the game didn't allow for even the remotest possibility that Jamie Vardy would turn out to be an England international, and well, doesn't SI and all everyone else in the football world that wasn't competing to buy him from Fleetwood look a bit silly now! But it also wasn't designed so the likes of Lukaku, Sturridge and de Bruyne failed to improve at all because Chelsea didn't love them enough and were comfortably surpassed by the entire squad of young high-end Championship players from a manager that knew how to treat them right, because every single one of them had their potential set high enough to let them become high-end Premiership players with the right treatment. That would have been a much bigger realism issue.

For every Vardy, there's many more Andy Kings who despite great setups and regular appearances for rapidly-improving clubs they love, haven't changed that much since FM11. 

 

 

Edit: I do quite like the idea of the OPTION to very slightly randomise the potential of real players for the sake of people getting bored of players they tend to shortlist turning out the same on every save, but I think the overall career trajectories of existing players has tended to be a pretty solid aspect of FM.

[Firecracker048]

On 10/14/2016 at 15:12, Northernpilgrim said:

A player's PA is always fixed, but not always known. That's the key point. Vardy in hindsight, always had the potential to be a world class striker (as I think most would agree he now is). Why he didn't make it earlier is down to a number of factors. Yes, he developed later than some, so was dismissed by league clubs, and doubtless became disheartened by that. So he ends up dropping into non-league, playing and training part-time. He therefore doesn't develop as quickly as he might have, but whilst playing for Stocksbridge, he performs better than anyone else at that level, so teams from higher up start to notice. In 2010, still only 23, he joins Halifax, in the Northern Premier League. He again outperforms his peers, so teams higher up take notice, and he goes to Conference club Fleetwood. Whilst at Fleetwood he scores 31 in 36 games in the league, which is going to get the attention of League 2 and League 1 clubs, but also scores the goals that knock League 1 club Wycombe out of the FA Cup. That's going to attract attention from bigger clubs too, and sure enough, in 2012, now 25, Leicester take a punt on him. He isn't an instant success, but still manages 4 in 26 in his first season. His next season is where we see him start to show what he can do. In his second season in the Championship, he manages 16 in 37. His first PL season isn't fantastic, with 5 in 34, but 24 in 36 last season was brilliant. I know that many are familiar with the Vardy story, but putting it in these terms shows that the potential was always there, but it took a combination of the right club, right manager and right tactics to get the best out of him. FM, being a stats and numbers based entity is never going to replicate that accurately, although I guess it could be done to an extent by using a combination of professionalism, p.a and c.a, and determination. The best way to replicate it would be that his PA is very high, his determination and professionalism were probably not that high until the last few seasons. Maybe increasing the amount these attributes can change by would allow or the 'Vardy effect'?

I had a "Vardy effect", as you call it in my game with Tony Lee. The only problem was he did not have the "potential" to play in higher leagues, even though he was out-performing those who were well above his CA over the course of 3 seasons. The problem I had was that someone like that SHOULD get the attention of higher league clubs, but never did, so I never got to cash in his success(until I fired up the editor and made him good enough to sell for 2 mil). Maybe what I'm looking for is a temp "boost" in perceived CA, even if they never actually make it to there. I feel like someone like that would get attention from higher clubs, even If they would burn out quickly

 

On 10/19/2016 at 14:39, enigmatic said:

I think to an extent SI do that with regens: I haven't systematically studied it but I'm pretty sure there are a lot more players at the 180-200PA range after several seasons of a save than there are in a fresh save from the occasional messing around with scout tools, and I'm equally certain their personalities are pretty random and therefore unlikely to be as dedicated as Messi and Ronaldo. Having more than one candidate Messi for the next generation and having a bit more chance of getting regens that'll be good for your level if you want to work really hard at bringing them through (assuming the same distributional effect applies lower down the scale) is probably a good thing, even at the cost of a little imbalance. After all, it's what makes Youth Challenges possible.

But when it comes to real players, it's a simulation trying to make the best guess at what their actual career might look like. So upping all the  the real player PAs and making it so that players usually fall significantly short of them unless they have the setup and make the right decisions and selections will tends to result in far more unrealistic player careers than the current system. Sure, earlier versions of the game didn't allow for even the remotest possibility that Jamie Vardy would turn out to be an England international, and well, doesn't SI and all everyone else in the football world that wasn't competing to buy him from Fleetwood look a bit silly now! But it also wasn't designed so the likes of Lukaku, Sturridge and de Bruyne failed to improve at all because Chelsea didn't love them enough and were comfortably surpassed by the entire squad of young high-end Championship players from a manager that knew how to treat them right, because every single one of them had their potential set high enough to let them become high-end Premiership players with the right treatment. That would have been a much bigger realism issue.

For every Vardy, there's many more Andy Kings who despite great setups and regular appearances for rapidly-improving clubs they love, haven't changed that much since FM11. 

 

 

Edit: I do quite like the idea of the OPTION to very slightly randomise the potential of real players for the sake of people getting bored of players they tend to shortlist turning out the same on every save, but I think the overall career trajectories of existing players has tended to be a pretty solid aspect of FM.

I guess what I've been going for is making it so that players who "over perform" their CA/PA over multiple seasons will have a massively inflated value. because it seems like AI teams scouts just seem to always know who will and will not pan out, despite form and results. like form and results should sway larger teams opinions of players 

[qaz]

This has been an annoyance of mine for many years now.

Now I completely understand why in earlier editions of Football Manager, when processing power was less, the potential ability field made complete sense as a shortcut. 

But it's not real. Human beings don't have fixed potential. We have points where gains becomes increasingly incremental, we have injuries, mindsets, environments, etc which facilitate (or not) development. But we don't have fixed potential. Potential itself isn't a real thing, rather an invented concept.

In real life, when Rooney burst onto the scene as an exciting 16 year old with Everton, it wasn't that he had necessarily more potential than you or I. Rather he had outstanding ability for someone of this age and this meant he was afforded more attention in coaching, better facilities, more and higher level experience.

Author Malcolm Gladwell in his book "Outliers" goes into detail about this, using the example of the NHL (National (ice)Hockey League. Where the vast majority of players were born early in the year rather than later. This is because say in an under 14's competition, the player whose born in January is nearly a year older (and more developed) than the player whose born in December). The best players of that age group (statistically usually the oldest), are often then spotted by scouts and given more development opportunities. Something the players who were unlucky enough not to be stand outs are youth levels don't receive. This carries forward for pretty much their entire career and leads to some pretty crazy statistics if you can bothered to look it up. Here's a small link to explain briefly what I mean (http://www.arcticicehockey.com/2008/12/malcolm-gladwells-outliers-and-making.html)

I'm not a great writer so I doubt what i've written itself will be very convincing, but I hope those that don't understand will be curious enough to investigate further.

My point is that potential ability is an outdated idea and isn't real. Football Manager now has enough depth to drop the statistic entirely and instead double down on things like experience, personality, environment, coaching, facilities, injuries etc to calculate how a player will develop. Just taking into account the number of footballers who could have become top-class players but didn't due to career defining injuries shows clearly that Football Manager as it's currently produced is a very pale shallow reflection of reality in this area.

Sorry for rambling. Hope someone from SI actually takes the time to look into this properly.

 

[Maaka]

1 hour ago, qaz said:

My point is that potential ability is an outdated idea and isn't real. Football Manager now has enough depth to drop the statistic entirely and instead double down on things like experience, personality, environment, coaching, facilities, injuries etc to calculate how a player will develop. Just taking into account the number of footballers who could have become top-class players but didn't due to career defining injuries shows clearly that Football Manager as it's currently produced is a very pale shallow reflection of reality in this area.

 

Yes, but no. I see your point, but even so, it's not that the "potential" as a concept is outdated. As this is a game based on numbers, it needs something that can be quantified to tell how good a player can be. And (unfortunately) it's not like anyone can become world stars, purely based on facilities, injuries and coaching (and others). What the game does (or at least, the researchers), is setting a cap at how good they believe a certain player can become, and within this range, said player can improve in-game on behalf of the factors you mention.
As has been mentioned before, it is possible for a player to excel at a level higher than what's percieved by his CA, due to tactical settings, and the right attributes at the right time etc, but that doesn't make the PA values obsolete.

AFAIK, the fact that FM is some sort of a very advanced spread sheet (figure of speech), makes it necessary to have potential that can be labeled with a number, quantified.

[qaz]

56 minutes ago, Maaka said:

Yes, but no. I see your point, but even so, it's not that the "potential" as a concept is outdated. As this is a game based on numbers, it needs something that can be quantified to tell how good a player can be. And (unfortunately) it's not like anyone can become world stars, purely based on facilities, injuries and coaching (and others). What the game does (or at least, the researchers), is setting a cap at how good they believe a certain player can become, and within this range, said player can improve in-game on behalf of the factors you mention.
As has been mentioned before, it is possible for a player to excel at a level higher than what's percieved by his CA, due to tactical settings, and the right attributes at the right time etc, but that doesn't make the PA values obsolete.

AFAIK, the fact that FM is some sort of a very advanced spread sheet (figure of speech), makes it necessary to have potential that can be labeled with a number, quantified.

Messi didn't pop out of the womb juggling a soccer ball

Of course he has some physical advantages others don't . But there is no reason why any baby born tomorrow (of fit health and body), with the right training, coaching, personality, mindset, luck with injuries, nutrition etc can't eventually be a world class footballer.

There are no hard limits here. . . . 

 

 

Additionally, it wouldn't be difficult for SI to program this effectively given all their expertise and computing power . It's a philosophical choice by Mr Miles Jacobson as to why things are done this way. 

[HUNT3R]

Just now, qaz said:

Messi didn't pop out of the womb juggling a soccer ball

We're not talking about babies just born here. We're talking about 15-16 year olds. They've already developed to such an extent that they all have a ceiling of some sort. 

[qaz]

Just now, HUNT3R said:

We're not talking about babies just born here. We're talking about 15-16 year olds. They've already developed to such an extent that they all have a ceiling of some sort. 

SI are absolutely smart enough to produce a system of player development that keeps game balance without an unrealistic hard cap. 

The funny thing is , while it would be a substantial adjustment , it would actually make the game both more realistic and less burdensome in terms of complexity. 

Again , it's not because it can't be done . It's because the current SI decision makers haven't learnt this yet. 

[HUNT3R]

Just now, qaz said:

The funny thing is , while it would be a substantial adjustment , it would actually make the game both more realistic and less burdensome in terms of complexity. 

No, it'll be a lot more complex and difficult to balance.
 

The cap isn't the issue. It's the development of players. If that was modeled in a more realistic way, PAs (of existing players) could be 'looser' to allow different ways of development. At the moment, players reach their ceiling too easily.

[santy001]

Biggest problem with any kind of in-game malleable system that can be influenced is this:

It won't hold up against a playerbase determined to maximise the gains through it. 

No matter the merits of it, it just isn't cost effective or remotely viable to develop this kind of system and having to keep deploying fixes to patch up the new stress points players find in this system. Largely it doesn't matter how many years of development or how much money is spent on developing this kind of system, it just wouldn't survive even the open beta time-frame. The game would feel incredibly more complex, until the cookie-cutter "XYZ's FMXX guide to guaranteed 190+ PA players" appeared. 

As Hunter says, the development is the main issue. Players are either on rails to success or not. 

[qaz]

8 minutes ago, HUNT3R said:

No, it'll be a lot more complex and difficult to balance.

Give me a week with the FM source code, a talented SI programmer and a good computer to run simulations 

Not saying what i'd come up with would be perfect, but it would be a huge improvement over the current

8 minutes ago, HUNT3R said:

The cap isn't the issue. It's the development of players. If that was modeled in a more realistic way, PAs (of existing players) could be 'looser' to allow different ways of development. At the moment, players reach their ceiling too easily.

I think the problem is both. But I agree entirely that players reach their PA ceiling far far too easily at the moment.

 

[qaz]

22 minutes ago, santy001 said:

Biggest problem with any kind of in-game malleable system that can be influenced is this:

It won't hold up against a playerbase determined to maximise the gains through it. 

No matter the merits of it, it just isn't cost effective or remotely viable to develop this kind of system and having to keep deploying fixes to patch up the new stress points players find in this system. Largely it doesn't matter how many years of development or how much money is spent on developing this kind of system, it just wouldn't survive even the open beta time-frame. The game would feel incredibly more complex, until the cookie-cutter "XYZ's FMXX guide to guaranteed 190+ PA players" appeared. 

As Hunter says, the development is the main issue. Players are either on rails to success or not. 

 

So what i'm hearing from your post is "i accept the system as it is now isn't right, but I don't believe it can be fixed"?

I doubt i'm any smarter than you, but I think i'd manage to fix it given the right resources.

[Slabbekoorn]

The CA/PA discussion is a difficult one. Who would've known that Vardy would become a great player? It's pretty much impossible to predict for the entire database.

 

However I really like the idea of the 'late bloomer' stat. That would be a great hidden stat that would result in a nerfed CA growth until the age of 25 and an improved CA growth when they've turned 25. 

[qaz]

3 minutes ago, Slabbekoorn said:

The CA/PA discussion is a difficult one.

With every respect, I don't believe the issue is as difficult as most think. It's just we've been conditioned by playing Football Manager for 10+ years to think in a certain way and haven't really stopped and questioned our assumptions.

[santy001]

I actually have huge faith in the current PA system after what must be like 5 years now of up close working with it and seeing the subtle evolution within it across that time as a researcher. My point is that with it being a completely fixed system is pretty much the best solution. I have begun some more developmental specific feedback myself, but the channel my feedback goes through mentions a lot of what I've seen during research and things from research so they aren't on public forums. 

It's difficult to explain, but FM is a game and all games are driven by algorithms and the numbers within them. It may give an illusion of randomness, but as I said players would discover the route to enforce what they want. You might have a talented youngster anyway, but then people figure out a way to enforce "late-bloomer" development on him as well, so even though he's a great player already players artificially unlock a higher level so you could even potentially take a player like Hazard who is probably ballparking 175/180CA and push him to 200CA or something a little further down the line by being able to trigger this "late bloomer" rise in PA and acquisition of it through CA.

PA really ought to be completely invisible to players though I think, CA as well. Especially given that the calculations and "attribute weightings" of FM are getting ever more complex. It means these figures are becoming something increasingly distant from the playerbase. That's not a bad thing, its again a touch on the evolution of the system. They have such a niche application now though. For example, giving Wayne Rooney & Zlatan Ibrahimovic say a PA of 175 for the sake of argument means two extremely different things. That doesn't translate at all in the current system, but it means even though they both have the same PA if they both attained it they would still be massively different players and one of them would be notably better than the other. It seems counter-intuitive to say it that way but its how it works in the game now. 

[qaz]

44 minutes ago, santy001 said:

I actually have huge faith in the current PA system after what must be like 5 years now of up close working with it and seeing the subtle evolution within it across that time as a researcher. 

Fair enough. Happy to agree to disagree.

It's worth remembering tho, this is a single player game. While exploiting the mechanics of an uncapped PA system might be possible (though harder than you probably think), it's a single player game. We should be able to play the game the way we want. 

[santy001]

The only thing I can say is the potential for late bloomers is already there, the potential for a PA which fluctuates over time as far as everyone can tell is already there. 

It really doesn't lie at all in any changes to the PA itself though. It's all in the hidden mental attributes, the right changes there would be the perfect place for this.

[enigmatic]

1 hour ago, qaz said:

Fair enough. Happy to agree to disagree.

It's worth remembering tho, this is a single player game. While exploiting the mechanics of an uncapped PA system might be possible (though harder than you probably think), it's a single player game. We should be able to play the game the way we want. 

People can play the game the way they want. For people that want to turn players into a late bloomer despite the researchers who watch them and the clubs that don't sign them not thinking they have the potential to be a late bloomer, data editors exist.

 

The other point I'm surprised santy didn't make is that it's not just about exploiting mechanics, it's also about getting the basic mechanics to function as people expect. Right now if a youth player he watches appears to be likely to be good enough for the first team in Stoke in future, but frankly probably not Messi, he sets one value (or value range) which ensures that if things go well for him he will be good enough for Stoke's first team, won't be the next Messi *and* ensures the AI non-players supposed to be better at scouting are likely to identify this. Mental attributes and injury are the icing on top of that which can cause players to speed up or slow down their development, but they invariably won't reach a playing level which (99 times out of 100) they won't in real life.

Replacing that cap (and factor in AI scouting) with an implicit growth pattern based on a function of several separate values purportedly linked to that individual's attitude, learning speed, experience and physique just makes the whole thing a lot murkier. And ironically, probably less realistic. People didn't rave about teenage Ronaldo's physique or learning capability or attitude towards the game, but they certainly talked about him having the potential to be one of the best players in the world.

[russianorphan7]

On 10/6/2016 at 01:19, borivoje213 said:

I'm guessing this floating PA would need some kind of Cap on it to prevent players from getting too good/bad too quickly.

Because if it's just rolling and open then there might as well not be a PA and just have a CA that reaches upper and lower limits based on performance.

Is that close to the kind of thing you had in mind?

I think you can have a natural cap. Basically, if a player is outperforming his CA on the pitch, then that differential between his performance and his CA would create a pulling effect on his CA to gradually match his performances. This pulling effect would be slow, though, requiring a player to outperform his CA for an extended period of time (depending on the differential gap, it could be a season or even 2-3 or more seasons). There would be other variables at play, notably strength of competition, training facilities, staff, teammate quality, and system of play, that would effect this growth, as well. Also, if a player is older than 22-24 range, then these stat increases would have to obviously be mostly mental and technical (strength, stamina, balance might improve), and based on how the player is performing. If he is bagging goals with his head, shots off through balls, long shots, etc., then those relevant stats would be the ones improved. This concept would similarly apply for midfield, defense, and GK play. The natural cap would always be that player's performance, though. Even with slowly improving stats, the performances would ideally level off or revert to the mean eventually, hencing capping a players CA so he doesn't just spiral out of control and become a Messi.

This is obviously a much more sophisticated mechanism for growth which I think better reflects how players actually improve in real life, and it would require an overhaul involving some mathematical modeling and significant testing/simulation to get it just right (you don't want league 2 players becoming messi, or vice versa). Ultimately, I think something like this would be the goal, though. In real life, there isn't really a hard cap on someone's potential. Natural physical gifts aside, how a player develops is largely a function of their personality (which is malleable) and work ethic, the environment they are in, and ultimately how they actually play in games.

[russianorphan7]

4 hours ago, enigmatic said:

People can play the game the way they want. For people that want to turn players into a late bloomer despite the researchers who watch them and the clubs that don't sign them not thinking they have the potential to be a late bloomer, data editors exist.

 

The other point I'm surprised santy didn't make is that it's not just about exploiting mechanics, it's also about getting the basic mechanics to function as people expect. Right now if a youth player he watches appears to be likely to be good enough for the first team in Stoke in future, but frankly probably not Messi, he sets one value (or value range) which ensures that if things go well for him he will be good enough for Stoke's first team, won't be the next Messi *and* ensures the AI non-players supposed to be better at scouting are likely to identify this. Mental attributes and injury are the icing on top of that which can cause players to speed up or slow down their development, but they invariably won't reach a playing level which (99 times out of 100) they won't in real life.

Replacing that cap (and factor in AI scouting) with an implicit growth pattern based on a function of several separate values purportedly linked to that individual's attitude, learning speed, experience and physique just makes the whole thing a lot murkier. And ironically, probably less realistic. People didn't rave about teenage Ronaldo's physique or learning capability or attitude towards the game, but they certainly talked about him having the potential to be one of the best players in the world.

Players should not have to mess around with the editor to correct the rigid, oversimplified mechanics of the game. I'm not saying the issue is easy or SI isn't doing a decent job and trying hard to improve, but it is fair for us to give feedback requesting a more sophisticated growth system in future games. 

I think the problem with your last paragraph citing Ronaldo is that people actually do attribute his amazing growth to work ethic, and not just some nebulous and inevitable potential. See statements by Sir Alex Ferguson and others about how hard he trained to grow from an initially raw talent to a world class player. He became the player he is today through tireless dedication to his craft, in addition to some physical gifts (that required personal development, as well, I might add) that he might have been born with. Now, I agree that PA is an okay assumption for a player growth model, but at the end of the day it is an oversimplification that leads to some less than ideal outcomes (which have inspired this thread). 

In real life, potential is just a vague word people throw around as a way of projecting a player's future ability based on what they see now. In truth, there is no hard cap that dictates what a player will become. It is a player's work ethic, personality, and environment that largely dictate how he develops. Now, there obviously should be some natural limiting factor so every player can't become a Messi; we do also have to acknowledge the natural proclivity and physical gifts that some players have, after all. But having said that, the mental and technical attributes (and strength, stamina, balance, and maybe natural fitness to a lesser degree) of a player should be far more variable than they currently are, and so that limiting factor should therefore be more flexible than the hard PA cap that exists now.

This hard PA cap ultimately just forces players to focus on PA and not on the actual development and performance of the player. We should not have scenarios where a player's CA reaches his PA and just stops growing, even if he is still young and (1) has a good personality that enables further improvement (2) continuing to train well (3) out-perform on the pitch and/or (4) exposed to better competition, coaches, and/or training facilities. Nor should we have to take matters into our own hands and just correct PAs when we feel it would be more realistic. Also, a more flexible system would allow for late bloomers that do occur in real life, which is a phenomenon that is not really capture in the current system very well (or at all).

Again, not blaming SI here. A hard cap is a reasonable initial mechanism to manage the complexity of the game, but in the long run it should definitely evolve towards a more flexible and realistic system.

[russianorphan7]

8 hours ago, Maaka said:

Yes, but no. I see your point, but even so, it's not that the "potential" as a concept is outdated. As this is a game based on numbers, it needs something that can be quantified to tell how good a player can be. And (unfortunately) it's not like anyone can become world stars, purely based on facilities, injuries and coaching (and others). What the game does (or at least, the researchers), is setting a cap at how good they believe a certain player can become, and within this range, said player can improve in-game on behalf of the factors you mention.
As has been mentioned before, it is possible for a player to excel at a level higher than what's percieved by his CA, due to tactical settings, and the right attributes at the right time etc, but that doesn't make the PA values obsolete.

AFAIK, the fact that FM is some sort of a very advanced spread sheet (figure of speech), makes it necessary to have potential that can be labeled with a number, quantified.

I think what many people are asking, at least in the relative short term (next few cycles of FM), is to loosen the hard PA cap system by providing a mechanism for players to exceed it when they reach it too early or significantly outperform their CA.

If a young player reaches his PA cap years before he naturally would stop growing, then he shouldn't just abruptly stop growing. That doesn't seem realistic. Instead, there should be some mechanism for the PA to self-adjust to reflect that extra time that he could grow.

Also, say a player is between 24-27 years old and has reached his PA, but is significantly outperforming his CA on the pitch for an extended period of time (say, a season or longer).  There should be a mechanism for his CA to slowly grow to reflect the fact that he is still at an age where he can still tangibly develop his abilities, but just not as quickly or significantly as when he was younger. Obviously, they would have to do simulations and testing to make sure this isn't overpowered, but I think these added mechanisms would make the growth system more flexible and realistic.