An Evolution of Potential Ability

So, there's a couple of threads in Data Issues and the like related to how CA/PA should be a thing of the past, but nothing really in general discussion regarding our perceptions.

Something frequently mentioned here is that people feel the current fixed number is restrictive in terms of attributes and overall ability, and also feel that your club's facilities and coaches should have more impact. This, of course needs to be balanced.

There are two types of player in the game. Real and Newgen. The younger real players are given a potential ability by researchers, and newgens are simply given a fixed number. At present, even this random PA can be glaringly inaccurate within a few months of game release. Equally we all want to feel that our achievements and growth both as managers and clubs should lead to being able to develop players better, and many feel that CA/PA is no longer relevant.

I'm ignoring CA since (data issues aside) it obviously is relevant. PA on the other hand, needs looking at.

I think overlapping and potential range are the key changes here.

So, Overlapping is a simple issue. Rather than having the current setup which simply leads to 90% of the players using the same players from a potentials shortlist all year, we have a lot more variation in what can be achieved. Somehow signing a -9 at Annan Athletic and watching him develop into a beast in the Scottish Third and upwards is not realistic.

A player expected to be world class (-10) might have a PA between 140 and 200. Rather than a guaranteed 170-200. A very good player (-9), might be 120-180. Take the -10, if he never moves to a club with adequate facilities, coaches, tutoring, and ultimately first team action he should struggle to make it much beyond 140. If he moves to Manchester Utd on the other hand a combination of ALL the above factors along with his age should help him develop, aided or hindered by his reputation, attitude, or professionalism accordingly.

Example -

Young -9 Swede stays in Sweden. It's not the country or league that limits him, it's the club, it's facilities, and his mental attributes. The leagues and countries need to have league standards and reputations that can grow in FM11 so winning the Champions League 6 or 7 times in a career with AIK in Sweden will in due course end up with a highler league reputation, more money for the clubs, better facilities as a result. You could therefore develop players to a very high standard if you were dominating Europe, but whilst Sweden remains a footballing backwater he'll only develop to a lower area of his range dependent on his mental, hidden and physical attributes.

Equally, the same -9 moving to Manchester Utd might not be getting an first team action, might lack determination, might be a tad unprofessional and injury prone so while he develops as a result of his surroundings he is potentially hindered by the above and will possibly still not make a very good PA.

There's far too much talk in the current PA debate that centres around "I have a big club and thusly all of my players shall verily become Lionel Messi/John Terry/Steven Gerrard/Iker Casillas [delete as applicable]" This is folly. Whilst the club and facilities undoubtedly should have a bearing, so should the players mental attributes, his natural fitness, his injury proneness etc.

League and Nation reps need to be as fluid as club reps proportionately. If this is finally cracked in FM11, allowing for proper career games in smaller countries, you can then tie PA's to a wider spectrum as Potential Range. This means that a -8 with all the right stuff going for him could easily develop to being as good as a -10 with an attitude problem and lesser facilities or tutoring for example. How may times have you spent 10 years dominating Europe as Anderlecht, Standard, Aberdeen, IFK Gothenberg, Tours FC or Sporting Lisbon only to find that you can't keep players because your league rep remains poor? When you do well, you create more Euro spaces for other teams in your league, they in turn get more money, grow facilities, get bigger stadiums, but struggle to keep players or coaches? So, this needs fixing as part of PA, or the PR I'm suggesting.

PA might be altered to PR, actually becoming a range rather than a fixed number, so:

Alphonse Areola -10, means he could end up with a potential range that would currently be defined as between 140 and 200

Fed Macheda -9, means he could end up with a potential range that would currently be defined as between 120 and 180

Shane Duffy -8 means he could end up with a potential range that would currently be defined as between 100 and 160

Areola's attitude might suck, and he rarely gets games and is not tutored well with PSG having perhaps poor coaches so may end up a lesser player than Shane Duffy who is exposed to first team football, with perhaps better coaches, better tuition, and an equal standard of training facilities. Macheda has a massive advantage at Manchester Utd, but could just as easily be tempered by his own mental and physical limitations.

It's down to you as a manager to develop the facilities through your board, sign the best coaches, think cleverly about tutoring, exposing the player to first team games either at home or on loan. Training is a big factor too. After that the players mental stats and hidden stats play a part in the development.

What you get is a player moulded by you, and your actions, and your club's capabilities.

A player scouted with a tool should not show a fixed number, especially since PA is meaningless when used without umpteen other factors, it should show the potential range of the player. His ability? That's up to you to work out given the parameters. (In fairness, I dislike such tools, and am only referencing them in terms of what we would see as a give number)

Further, regens don't have random PA's they have a fixed one, which is a failing in the game mechanic. A young 14yo italian at Carrarese cannot realistically be defined as a world beater at 14 whilst playing in the footballing depths of Italy. So, again, instead of PA, there should be PR. Young Roberto Handbaggio should have a range, a range that can me maxed out with all the right conditions.

If he never get's picked up he'll stay at the lower end of his PR in the lower leagues of Italy.

If he get's taken to Chelsea but is a lazy stroppy feck he'll do better as a result of surroundings and facilities but not become great.

If he get's taken to Chelsea and is a determined professional he'll maybe become the next Lampard.

In simple terms. CA/PA not dead, but certainly a limitation, both in terms of youth development and long term careers.

Anyone follow this?

Thoughts? Discussion?

I disagree with your interpretations. A player's potential does not change based on his location, training, or so forth, by the time he's 16. However, his ability to reach his PA does change. The situations you have described are CA changes, not PA changes. A player is not guaranteed to reach his PA.

Regens don't have ranges- neither do real players! Once a game starts, their PA is fixed.

The young Italian will still have the potential or not to be world class at his age, but the issue is whether he will reach it. His potential does not change.

I completely agree with SCAIG. In real life a persons potential is pretty much fixed by the time they are 15/16 (when they would enter the FM game world) Their potential will not change based on coaching or facilities or anything else. All that will change is their ability to reach their set potential.

I think the way a player's CA develops has improved in the last couple of FM's, but there can still be more variation.

To be blunt, this is just plain wrong.

You're applying ideas to PA that should be, and already are, applied to CA.

If a player is at a club with better facilities and has good mental attributes etc., this increases the chance of that player's CA reaching the level of his PA, but the PA still needs to remain fixed. Having a good attitude will help you to be all that you can be. "all that you can be" is your PA. Having a bad attitude will mean you never fulfil your potential, your potential is your PA.

Example: Let's say I have a player with high potential and a good attitude. If I had him tutored by a succession of lazy, undetermined, unprofessional idiots, then his development might change completely - his attributes would cease to rise, as he did less and less work in training, and so his CA would stay roughly at whatever level it had reached.

But, if I saw this happen, I would look at him and say "what a waste of potential." because his PA would still be high - it's just that he'd have no hope of reaching it.

His effective PA - the level he could actually now be expected to reach - would be far lower than that he had when he had a good attitude, but making the actual PA value change is only of benefit to those who cheat and look at what is meant to be a hidden stat. Players with a high PA don't automatically reach it: if you go to whatever players are on your mental short-list at the beginning of a game, and edit their professionalism and determination right down, but leave their PA as high as before, they will never reach it. You see it a lot, later in the game, with a bunch of South American regens who start with very high stats, but have crap personalities and never develop their CA.

Furthermore, it is training facilities and mental attributes, along with first-team action and the avoidance of injuries, that determines a player's development, far more than the league that he plays in. You've tried to shoe-horn a request for dynamic league reputation (I've stated my reservations about this in other threads, can't be bothered to repeat here) into your argument also, but if you have high-class facilities you can generate great players in any league - you just won't be able to stop them wanting to leave.

Brazil doesn't have the highest league rep, but it generates the most quality players, because a ton of its clubs have top training facilities and talent pool. Then bigger clubs poach their players.

I disagree with your interpretations. A player's potential does not change based on his location, training, or so forth, by the time he's 16. However, his ability to reach his PA does change. The situations you have described are CA changes, not PA changes. A player is not guaranteed to reach his PA.

Regens don't have ranges- neither do real players! Once a game starts, their PA is fixed.

The young Italian will still have the potential or not to be world class at his age, but the issue is whether he will reach it. His potential does not change.

Which is what I've said, and is also why a players potential should remain a range not a fixed number. No player is guaranteed to reach fixed PA, we know that, it's the fixed PA that's a horribly rigid mechanic.

Regens: I said

Further, regens don't have random PA's they have a fixed one

I'm sorry, but I believe a players potential changes depending on his surroundings. Wayne Rooney becomes a vastly better player at Utd than he ever would have at Everton. Theo at Arsenal.

PA too rigid for me I'm afraid. It works Short Term, but is a lousy implementation in long terms games.

Either way, disagreement is good, debate needs having based on the bi-monthly threads in other areas.

I didn't misread, you got my phrasing wrong. I meant "Yes, regens don't have a ranged PA- neither do real players".

The Wayne Rooney at Everton had a PA of 190, for example, but wouldn't reach it because he was playing in the wrong environment. His CA may only go up to 175 or something. That doesn't make his PA 175, it just means he fails to reach it. If he moves to United, he has the ability to reach 190 due to the better coaching, CL football, chance to get in the England squad, etc. His PA doesn't increase, but rate of CA increase does.

To be blunt, this is just plain wrong.

You're applying ideas to PA that should be, and already are, applied to CA.

If a player is at a club with better facilities and has good mental attributes etc., this increases the chance of that player's CA reaching the level of his PA, but the PA still needs to remain fixed. Having a good attitude will help you to be all that you can be. "all that you can be" is your PA. Having a bad attitude will mean you never fulfil your potential, your potential is your PA.

Example: Let's say I have a player with high potential and a good attitude. If I had him tutored by a succession of lazy, undetermined, unprofessional idiots, then his development might change completely - his attributes would cease to rise, as he did less and less work in training, and so his CA would stay roughly at whatever level it had reached.

But, if I saw this happen, I would look at him and say "what a waste of potential." because his PA would still be high - it's just that he'd have no hope of reaching it.

His effective PA - the level he could actually now be expected to reach - would be far lower than that he had when he had a good attitude, but making the actual PA value change is only of benefit to those who cheat and look at what is meant to be a hidden stat. Players with a high PA don't automatically reach it: if you go to whatever players are on your mental short-list at the beginning of a game, and edit their professionalism and determination right down, but leave their PA as high as before, they will never reach it. You see it a lot, later in the game, with a bunch of South American regens who start with very high stats, but have crap personalities and never develop their CA.

Furthermore, it is training facilities and mental attributes, along with first-team action and the avoidance of injuries, that determines a player's development, far more than the league that he plays in. You've tried to shoe-horn a request for dynamic league reputation (I've stated my reservations about this in other threads, can't be bothered to repeat here) into your argument also, but if you have high-class facilities you can generate great players in any league - you just won't be able to stop them wanting to leave.

Brazil doesn't have the highest league rep, but it generates the most quality players, because a ton of its clubs have top training facilities and talent pool. Then bigger clubs poach their players.

Ok, thanks. I understand what you are saying, and agree that much of this does apply to CA.

But should potential be so easily fixed at a young age? Even the researcher themselves use random PA on youngsters because we cannot aspire to know the actual potential of anyone until they have developed?

Ok, thanks. I understand what you are saying, and agree that much of this does apply to CA.

But should potential be so easily fixed at a young age? Even the researcher themselves use random PA on youngsters because we cannot aspire to know the actual potential of anyone until they have developed?

In RL a persons PA is set in stone the moment they are born whether they ever reach it or not is down to many other factors.

Researchers use a random PA because they don't have access to a "Genie Scout" in real life

I think, with regens, CA should be generated first, and then PA generated dependent on the CA, determination, and so forth. This should decrease the number of players who unexplainably peak when everything says they should continue, though I do think some should continue to rise.

I think, with regens, CA should be generated first, and then PA generated dependent on the CA, determination, and so forth. This should decrease the number of players who unexplainably peak when everything says they should continue, though I do think some should continue to rise.

If you go by the coaching/scouting stars when regens are created then this does actually happen. By viewing PA by the coaches/scouting stars it does appear to be variable and not fixed.

The problem with PA is when it is viewed as a number and not by the mechanisms that the game gives us to view it.

The problem with PA is when it is viewed as a number and not by the mechanisms that the game gives us to view it.

Absolutely right. If you do not view PA and use only the tools given to you by the game then you get a pretty realistic development model. This model encompasses all kinds of players including wonderkids, players who peak early, players who peak later etc etc.

If you go by the coaching/scouting stars when regens are created then this does actually happen. By viewing PA by the coaches/scouting stars it does appear to be variable and not fixed.

The problem with PA is when it is viewed as a number and not by the mechanisms that the game gives us to view it.

Absolutely, I've just experienced it myself.

Had a DR in my youth team rated at 1.5* potential, never did anything for the youth team. I gave him a two year deal as an emergency backup when he was 19 as his wages were next to nothing and I was short of DRs.

First season in the reserves he progressed rapidly, averaged 6.9 and PA went to 2.5*

Second season got a chance in the first team, played well and ended up with 20 odd games at over 7.00. PA went to 3.5* and he is now a fixture in the main squad.

Not sure I have the strength to start arguing over PA again, but... PA doesn't have to exist if we can ensure that on average a player's peak CA will be what we would assign a PA to if we knew prior to peaking.

It also fails to somewhat model someone who may have "multiple" ceilings and discovers these by being a "late bloomer", meaning that a player's PA may not be best modeled as a solid number, but perhaps something that moves (slowly?), possibly up and down.

Take a youngster with CA 80, PA 150. Thanks to a hard ceiling the probability of him having CA 151 is zero while the probability of him having CA 0 (multiple injuries, lazy bugger, rubbish player, etc. etc. etc.) is not zero (quite close obviously) but still greater than that of him reaching CA 151.

If I placed 100 identical youngsters like this in-game and plotted their peak CAs, I suspect (conjecture, my computer isn't fast enough to confirm this ) it will look similar to a normal distribution (skewed perhaps) with mean maybe at 140 but truncated at CA 150, something like this: http://img121.imageshack.us/img121/5711/normh.png

However, we are missing the "tail" of the distribution. For low PAs (or simply small CA-PA differences) this is akin to a player peaking early and never developing further.

I think this illustrates a possible problem with a solid ceiling - PA can be thought of as a generic sort of average taking into account all plausible circumstances and their probabilities. However, once a player goes through an unusual, i.e. far from average, career path (i.e. low PA but the player suddenly starts punching way above his weight, say a late bloomer), this average isn't appropriate any more. Think trying to drive a car down a hill but veering off slightly onto a slippery slope - you've deviated from the average and that previous average no longer appropriately describes your journey.

Those with solid ceilings and good mental attributes have their futures determined from the start and I don't believe this can be appropriate in all cases. For two players close in PA but not too close (i.e. 10 PA higher for one player) the lesser player will never be better than the other. Yet in some circumstances a very different set of circumstances, i.e. a manager takes a liking to the lesser youngster and starts playing him irregardless, and he receives much better training facilities - he has gone far off his average track to the extent that his PA is no longer seems appropriate.

I don't believe that "it's just a number"/"you're not meant to know it anyway" is a valid argument, the reason being that it assumes PA is correct and that you can construct any number of plausible arguments around it to suit any scenario. However, I don't believe you can assume PA is correct to... Justify PA being correct. I think it would be better to assume nothing and then argue about PA in this sense.

I've put forward a suggestion of a soft ceiling to alleviate this missing tail probability, so that a player that hits their PA can still go above this PA but perhaps much slower reflecting the fact that his circumstances are indeed very different but he will always be hindered by his "talent level". Then we may see more players from low leagues work their way up through grit and luck as their PAs were low to begin with but they found unusual circumstances to justify moving away from their average.

A bolder approach would be to use PA as a sort of "benchmark" average so perhaps it's a normal distribution centred on the PA instead. In other words a player is allowed to succeed (i.e. go beyond their PA) in a similar way to how they can fail (i.e. not reach their PA) - yet for all intents and purposes the average peak CA will be on average their assigned PA.

An even bolder approach is to not use PA at all and use game mechanics to ensure that the most probable to succeed are those with high talent, ambition and professionalism, as well as being at a good club with good facilities. Then their average peak CA is essentially a function of all these things. PA can be considered to be their "talent" level. The advantage of such a model is that if things like their ambition changes (tutoring?), then their "projected future" (i.e. peak CA) will change. Yet we don't need to know what their projected future is a priori (i.e. at present) - we let the game take care of that. This is likely beyond the realms of current processing power though.

This way instead of scouts looking at a virtual number, all they will look for are talented youngsters with their personality and circumstances - and project a "PA" value from there.

Personally I feel CA and PA will be gone within the decade anyway. CA is essentially a weighted average and therefore unnecessary in terms of processing (think putting any number for an attribute you want in the database without having CA bring it down later). PA is in some ways a weighted total and therefore can also be removed in terms of processing. However some sort of "talent level" is required - if not for regens then for real players - but this does not mean it is the weighted total - just some component of a weighted total.

I think the PA system works just fine. There's a logic to it and it serves its purpose in lower leagues (gives clubs cash boosts from transfers) and big teams (nurture your talent).

It only matters if you use a third-party program for scouting.

In RL a persons PA is set in stone the moment they are born whether they ever reach it or not is down to many other factors.

Researchers use a random PA because they don't have access to a "Genie Scout" in real life

tbh i don't agree with that at all, PA isn't based entirely on a players natural talent. Most youth coaches will say that between the ages of 8 and 13 arethe crucial years of development for a young player, if they don't receive good coaching in that time natural talent won't take them all the way, it only counts for so much.

tbh i don't agree with that at all, PA isn't based entirely on a players natural talent. Most youth coaches will say that between the ages of 8 and 13 arethe crucial years of development for a young player, if they don't receive good coaching in that time natural talent won't take them all the way, it only counts for so much.

What you are talking about there is CA not PA.

Natural talent = high PA, good coaching = increase of CA.

A player who doesn't receive the coaching still has a high PA but will never reach it while players who do receive the coaching can still never exceed the limitations of their talent.

The main problem I have with PA is that they are saying there is this fixed invisible barrier that a player cannot pass.

I don't think that's how it works, players can always improve their game, whether it be by new and better coaching.

Also that PA spans not only technical but mental stats and physical stats as well. That makes no sense. A player would be able to progress mentally and this wouldn't have an effect on his playing skills.

An even bolder approach is to not use PA at all and use game mechanics to ensure that the most probable to succeed are those with high talent, ambition and professionalism, as well as being at a good club with good facilities. Then their average peak CA is essentially a function of all these things. PA can be considered to be their "talent" level. The advantage of such a model is that if things like their ambition changes (tutoring?), then their "projected future" (i.e. peak CA) will change. Yet we don't need to know what their projected future is a priori (i.e. at present) - we let the game take care of that. This is likely beyond the realms of current processing power though.

This way instead of scouts looking at a virtual number, all they will look for are talented youngsters with their personality and circumstances - and project a "PA" value from there.

I think that by using the in-game systems for revealing PA instead of looking at the numbers you actually get something very close to what you are describing here.

PAs according to scouts/coaches do actually change. They are not static, unlike the PA numbers which we cannot see. They are also a judgement on the player based on a variety of factors, not just a report on the PA number.

What you are talking about there is CA not PA.

Natural talent = high PA, good coaching = increase of CA.

A player who doesn't receive the coaching still has a high PA but will never reach it while players who do receive the coaching can still never exceed the limitations of their talent.

I can see what you're saying that a player with no training in their life will never reach the heights of football even if they could have with proper coaching, and that is certainly the way PA is defined in the game. But irl a player doesn't have a hope in hell of reaching a top level if they aren't coached well between 8 and 13, even if they receive excellent coaching after that. To me coaching from a young age can define a players PA.

I can see what you're saying that a player with no training in their life will never reach the heights of football even if they could have with proper coaching, and that is certainly the way PA is defined in the game. But irl a player doesn't have a hope in hell of reaching a top level if they aren't coached well between 8 and 13, even if they receive excellent coaching after that. To me coaching from a young age can define a players PA.

Whilst that may be true for RL its fairly irrelevant for FM as 8-13yos aren't in the game.

In FM terms a player with 200PA would need to come into the game with a fairly high CA to start with if he is to reach his PA. This could be viewed as having had good training between 8 & 13yos if you want to see it that way.

TBH I think the biggest problem with the CA/PA system is that in RL very few players reach their PA whilst on FM far too many do.

EDIT

Sorry missed your last point about coaching defining a players PA.

IMO the coaching has an affect on the PA a player can reach but this will always be less or equal to his natural PA which is set at birth.

Honestly I think the notion of there being a fixed "potential" for a player is fundamentally flawed. What holds players back (or allows them to move on) is personality (professionalism, ambition, determination...) and genes. PA sort of works as an acceptable abstraction of the latter, but is nonetheless flawed as it incorporates nearly all mental attributes as well. Players with the right approach do not suddenly stop learning, nor do they stop developing assuming physical limitations or external factors don't hold them back (bad coaching, low-level fooball, non-footballing conditions...).

Long-term I think the player developmental model needs to move away from PA and instead attempt to implement realistic ability development (CA if you will) based on realistic factors (both personal and external), without a roof or with one that's limited to just the relevant attributes. A player's "potential" is nothing more than an estimation based on all such factors. Personality attributes already play a huge part in player development, there is a Natural Fitness attribute that can be expanded on, and we have a number of external factors (club, country, league...) present, so I don't think this would be unfeasible to implement, though time-consuming to balance.

PA is not a real life stat. CA is a real life stat. Unless SI have a time machine they will never know what any players PA is. I think the games works very realistically and if people didnt have the editor of Scout we wouldnt be having this debate.

Also, how do we know that in real life players dont reach their PA?

PA is not a real life stat. CA is a real life stat. Unless SI have a time machine they will never know what any players PA is. I think the games works very realistically and if people didnt have the editor of Scout we wouldnt be having this debate.

Also, how do we know that in real life players dont reach their PA?

Exactly my thoughts.

I still think that if the PA value was truly hidden and all we had to go on were coach/scout reports then this argument would not crop up as often as it does. People get too caught up with the debate over whether a PA value is realistic, that is exactly why it is hidden and not designed to be looked at.

Well, to a degree. Someone who can't help but check every player's CA/PA will obviously notice this a lot more, but PA is both present and noticeable in the game through reports and players no longer improving. What I think makes very little sense is that some players in the game who appear to have every chance of developing (the right approach in terms of personality, good coaches and facilities, first-team football...) will not past a certain point because of this set potential. So what's that "potential" supposed to represent that's holding him back, exactly?

What I think makes very little sense is that some players in the game who appear to have every chance of developing (the right approach in terms of personality, good coaches and facilities, first-team football...) will not past a certain point because of this set potential. So what's that "potential" supposed to represent that's holding him back, exactly?

How do you define 'a certain point' though? Do other people actually perservere with players once the coaches have declared them at their peak or do they keep going and see what happens?

Personally, I think you can still get more out of a player even once the coaches have declared them at their peak. Redistribution of attributes through good training and the teaching of PPMs can still help a player to improve, even if their CA is going no where.

The main problem I have with PA is that they are saying there is this fixed invisible barrier that a player cannot pass.

I don't think that's how it works, players can always improve their game, whether it be by new and better coaching.

Also that PA spans not only technical but mental stats and physical stats as well. That makes no sense. A player would be able to progress mentally and this wouldn't have an effect on his playing skills.

Think of a young kid, say 16. He's technically utterly brilliant. He can land a ball on a sixpence, or put it in any part of the goal from any distance, and he holds the world record for keepy-uppies. He's also very good at dribbling around cones. However, his movement is terrible, and he has no mental prowess whatsoever. He's very unfit and slow, he can't turn properly, he falls over all the time and he's weak. Does he have the potential to be a world class player? Probably not.

Mental, technical and physical sides all affect how "good" a player is. Has Wayne Rooney's recent improvement been down to technical increases, or has the overriding factor been better movement and reading of the game? Would Gabriel Agbonlahor be as good if he was slow?

I think that by using the in-game systems for revealing PA instead of looking at the numbers you actually get something very close to what you are describing here.

PAs according to scouts/coaches do actually change. They are not static, unlike the PA numbers which we cannot see. They are also a judgement on the player based on a variety of factors, not just a report on the PA number.

Not true. In mathematical terms, what we have is the following:

PA=E[f(ambition, professionalism, training facilities, ...)]

What I'm asking for is:

Peak CA=f(ambition, professionalism, training facilities, talent level, ...)

E[Peak CA]="PA"

Where talent level is some function of PA. i.e. for a real player, a researcher will set Messi's "talent level" to maximum and the least-talented amateur player, maybe Aruna Dindane, to a minimum (or is that too harsh?). However, "PA" generally doesn't need to come into the equation - a player will simply develop along the lines of this function f. Of course these attributes (ambition, professionalism, etc.) may change - maybe even a player's talent level will change - and this will simply affect the Peak CA. It will of course be unknown what the actual Peak CA is but on average the Peak CA will be what you may assign as PA in the database (for real players) or generated in-game (for regens).

There indeed exists an explanation to every scenario based on PA, but when used to defend PA is essentially begging the question.

The basic issue is that if PA is not fixed, i.e. there is not a hard cap on maximum player ability set when a player enters the game world, then there is nothing to prevent managers from maxing out a squad of 200 CA players.

Every regen/newgen coming through the youth ranks at any of the top clubs in the world will be able to be the next Messi in terms of overall ability. The manager simply then has to pick and choose the players with the best starting attributes and bobs your uncle, never have to sign a player again.

The current system produces a player with huge potential and great starting attributes about once every 2-3 seasons if you are lucky. Remove PA from the game and anyone managing at the top clubs in the game will be able to turn any player into the best in the world.

PA and CA mechanics are deeply entrenched in the game. You can't change the way these work without radically altering a huge number of knock-on effects, and simply removing PA will potentially destroy realistic gameplay at the top clubs.

The real question is: Is a fixed PA as soon as a player enter the gameworld realistic irrespective of Age? Is it realistic for a 130 PA player joining Manchester United at age 14 with excellent starting attributes to hit 130 CA around the Age of 19 and never be able improve his overall ability?

I'm not sure that is realistic, and I would definately support looking at ways of enabling PA to be improvable under certain ages under certain strict conditions, but it needs to be done with 1.9 eyes firmly focused on gameplay balance and knock-on effects across the entire game.

Whilst that may be true for RL its fairly irrelevant for FM as 8-13yos aren't in the game.

In FM terms a player with 200PA would need to come into the game with a fairly high CA to start with if he is to reach his PA. This could be viewed as having had good training between 8 & 13yos if you want to see it that way.

TBH I think the biggest problem with the CA/PA system is that in RL very few players reach their PA whilst on FM far too many do.

EDIT

Sorry missed your last point about coaching defining a players PA.

IMO the coaching has an affect on the PA a player can reach but this will always be less or equal to his natural PA which is set at birth.

Yeah i totally agree in fm that a player thats 14 or 15 should have a fixed PA, what i was saying was completely irrelevant to fm its just my point ofview irl.

The real question is: Is a fixed PA as soon as a player enter the gameworld realistic irrespective of Age? Is it realistic for a 130 PA player joining Manchester United at age 14 with excellent starting attributes to hit 130 CA around the Age of 19 and never be able improve his overall ability?

I'm not sure that is realistic, and I would definately support looking at ways of enabling PA to be improvable under certain ages under certain strict conditions, but it needs to be done with 1.9 eyes firmly focused on gameplay balance and knock-on effects across the entire game.

Plenty of players peak early and never get any better.

I dare say that Michael Owen's best years will be remembered as 18-20. Admittedly injuries played a part here. But what about Francis Jeffers? he looked excellent as a kid, was better than anyone else in his age range, seemed to be mentally there, but just never improved.

That 130PA/CA 19 year old at United had the ability to become a decent footballer. United helped him achieve that, but there isn't much further he can now go. Plenty of these will exist at United in reality, kids who look good, peak early and don't have anywhere else to take their game so quietly move on.

As others have said, the in game recreation of PA works well. The system only breaks when you look behind the curtain. Without these behind the curtain controls though things would become silly.

No matter how long you let an average player, say Narry Geville, play for United he'll never be in the top 5 in the world. He could peak early then just mature a little rather than outright improve for the rest of his career. Would he be a 130PA player? No, because United would see he had stopped improving and moved him on. 150/160 player? Sure if there's no-one better for the position they may well keep him. With a PAless system, we'd see Narry has good mental attributes and is pretty good at a young age. Therefore being at United, he'll carry on improving until he's better than Roberto Carlos, or whoever the current benchmark would be. This is further from reality than the idea of him hitting a wall.

What I would, however, support is the idea that the closer to the actual PA a player gets, the harder it should be to increase. This alone would mean most players would never quite touch their PA (players with 195CA would tend to have 200 PA rather than the 195 they have currently), players would always have a chance of improving, etc. It would mean that even that 19 year old at United, with what would be 130CA and 135PA, would have room to improve. He wouldn't stagnate. In fact, being at United would give him the best chance of hitting his actual potential, while it would still be out of reach at 19.

I really do disagree with people saying that PA should be able to change. It is not a real stat, it is a feature in FM to cap how good players can get. I'm sure the comments made about real players potential is true...that it can change dependant on other factors. We shouldn't be looking at the PA in FM as real.

It is the CA that is the most important stat here. The example of Michael Owen. In FM world he would be given a high PA. At the start of his career his CA would rise towards the top of his PA but then fall again. This doesn't mean that his Potential changed does it?

I think the main thing here is that perhaps the coding that assigns PA to newgens / regens needs to be looked at? Or how CA can change throughout a players career - where he plays, how many games, training, injurys?

CA does change throughout a players career dependent on where he plays, how many games, training, injurys and a dozen or so mental factors inbetween.

As SFraser said above I would also support looking at ways of enabling PA to be improvable under certain ages under certain strict conditions as per my OP, but it needs to be done with that focus on gameplay balance and a close eye on certain side-effects across the entire game.

A major failing of the game at present is the level of predictability seen in GP&TG wherein a select 90 or so players are the be all and end all of the first few seasons until regens arrive en-masse. Plenty of FM's former -9's have gone on to be mediocre in real life, as such random PA's should be a much wider banding in FM to allow for more variety.

Take a look at some of the -10's in the editor, and then look at the mental stats? Is it any wonder that so many of them are not the world's best in the game?

As in real life, the game should allow for them to fail miserably rather than simply be guaranteed a high fixed PA.

CA does change throughout a players career dependent on where he plays, how many games, training, injurys and a dozen or so mental factors inbetween.

As SFraser said above I would also support looking at ways of enabling PA to be improvable under certain ages under certain strict conditions as per my OP, but it needs to be done with that focus on gameplay balance and a close eye on certain side-effects across the entire game.

A major failing of the game at present is the level of predictability seen in GP&TG wherein a select 90 or so players are the be all and end all of the first few seasons until regens arrive en-masse. Plenty of FM's former -9's have gone on to be mediocre in real life, as such random PA's should be a much wider banding in FM to allow for more variety.

Take a look at some of the -10's in the editor, and then look at the mental stats? Is it any wonder that so many of them are not the world's best in the game?

As in real life, the game should allow for them to fail miserably rather than simply be guaranteed a high fixed PA.

I agree, the CA does change, not sure if that was clear in my last post.

I still believe that we wouldn't be having this debate if the PA stat was kept hidden. It sounds like how the PA is derived is the problem here rather then the PA being able to change. I strongly believe that the PA shouldn't be able to alter, the CA does this.

I agree that the game should allow players to fail miserably rather than simply be guaranteed a high fixed PA. This doesnt mean that the PA should change, rather the effects on the CA made more accurate.

.. and perhaps that PA should have considerably wider bandings.

I agree that CA/PA should be hidden, and suspect that many of the 'fixed PA' supporters are heavily 'tool' dependent and would not wish this to be changed.

Yes wider bandings is definately the way forward I think. I must say that as a 'fixed PA' supporter I have never used a CA/PA tool in any FM game. Don't see the point myself.

I must say that as a 'fixed PA' supporter I have never used a CA/PA tool in any FM game. Don't see the point myself.

Same here, and I think by not looking at them I am possibly oblivious to any problems it creates, which makes it look like a good model from my perspective.

A major failing of the game at present is the level of predictability seen in GP&TG wherein a select 90 or so players are the be all and end all of the first few seasons until regens arrive en-masse. Plenty of FM's former -9's have gone on to be mediocre in real life, as such random PA's should be a much wider banding in FM to allow for more variety.

Take a look at some of the -10's in the editor, and then look at the mental stats? Is it any wonder that so many of them are not the world's best in the game?

As in real life, the game should allow for them to fail miserably rather than simply be guaranteed a high fixed PA.

The game does allow for that.

There's a young centre back from South America who I believe has a -10 but may have a fixed potential. He's frequently bandied about as one of the two best young centre backs in the game, and on PA, he is. However, because he has quite a low reputation, he agrees to sign for quite weak sides. Also, I think some of his mental attributes are random, so he can be generated with a poor attitude.

I've seen screenshots, in OTF, of this player going to a mid-table Turkish club... and not getting in the first team. He stagnated, and ended up going backwards, because he's not getting football, and if he was, it's hardly the best league for a future Maldini to be playing in.

CA is by no means guaranteed to reach PA, though in ideal conditions, it usually will.

A major failing of the game at present is the level of predictability seen in GP&TG wherein a select 90 or so players are the be all and end all of the first few seasons until regens arrive en-masse. Plenty of FM's former -9's have gone on to be mediocre in real life, as such random PA's should be a much wider banding in FM to allow for more variety.

As in real life, the game should allow for them to fail miserably rather than simply be guaranteed a high fixed PA.

I have no issue with increasing the banding in the situation you describe above.

I fully agree that too many users focus on hidden gems that they have read about on forums rather than playing the game for themselves.

As per the Data Issues thread, I think we need extra bands.

Currently, we have -1, -2, etc., up to -10. -10 is 170-200 iirc, -9 is 150-180, -8 is 130-160, and so forth. Those may be slightly wrong, but that's generally right.

On top of those, I suggest we have a set of wide bands. -11 could be 140-200, -12, 110-170, -13, 80-140, -14, 50-110, and -15, 20-80. These would be assigned to 16-19 year olds. Not all 18 or 19 year olds would have one, of course, but the less proven ones (those who have never or barely played in a professional league) would. Once a player gets to 20, and sometimes earlier, the current ranges can be used, and once the researcher has a very good idea of his potential, a set PA.

Furthermore, I'd like to see a few more ranges introduced in the middle of the current ranges. For example, -8.5 being 140-170, and -9.5 for 160-190, just for a slight increase in precision. Due to the baffling nature of the CA scale (130 is good enough for a first team player at one of the best teams in the world... yes, there could well be 70 CA points between the best and worst players in a top club's side), I think that sort of precision is only needed at the top end, as the differences at the bottom end are too small to register.

Think of a young kid, say 16. He's technically utterly brilliant. He can land a ball on a sixpence, or put it in any part of the goal from any distance, and he holds the world record for keepy-uppies. He's also very good at dribbling around cones. However, his movement is terrible, and he has no mental prowess whatsoever. He's very unfit and slow, he can't turn properly, he falls over all the time and he's weak. Does he have the potential to be a world class player? Probably not.

Mental, technical and physical sides all affect how "good" a player is. Has Wayne Rooney's recent improvement been down to technical increases, or has the overriding factor been better movement and reading of the game? Would Gabriel Agbonlahor be as good if he was slow?

They are all very good points, but I don't think they should be determined from one pot of PA. Why not 3? One for technical, one for mental, one for physical? That's much more realistic.

Flexible PA is unworkable. Static PA is the only way.

My suggestion would be to break down PA into two categories:

1. Natural talent- the initial PA that is set upon starting the game. All players will develop towards this amount up to the age of 25 or so.

2. Performance- a player that averages 7.0 (or even higher, depending on the level of competition) and over for a season should see a small bump to their PA, say 5 points. This would occur after the age of 25, when a player is no longer being judged by his "potential" but by his actual performance on the pitch.

For example:

In FM, a player with a max PA of 130 will never get better than that, even if he somehow leads the EPL and Champions League in scoring and takes P.o.Y. honors. In real life? You better believe FM researchers would be inclined to boost that player's PA significantly for next year. With my system, however, you'd be able to see an extra 30 points or so to a player's PA under OPTIMAL circumstances before they hit their "peak" around the age of 31-32.

My suggestion would be to break down PA into two categories:

1. Natural talent- the initial PA that is set upon starting the game. All players will develop towards this amount up to the age of 25 or so.

2. Performance- a player that averages 7.0 (or even higher, depending on the level of competition) and over for a season should see a small bump to their PA, say 5 points. This would occur after the age of 25, when a player is no longer being judged by his "potential" but by his actual performance on the pitch.

For example:

In FM, a player with a max PA of 130 will never get better than that, even if he somehow leads the EPL and Champions League in scoring and takes P.o.Y. honors. In real life? You better believe FM researchers would be inclined to boost that player's PA significantly for next year. With my system, however, you'd be able to see an extra 30 points or so to a player's PA under OPTIMAL circumstances before they hit their "peak" around the age of 31-32.

Now this is not a bad idea!!

My suggestion would be to break down PA into two categories:

1. Natural talent- the initial PA that is set upon starting the game. All players will develop towards this amount up to the age of 25 or so.

2. Performance- a player that averages 7.0 (or even higher, depending on the level of competition) and over for a season should see a small bump to their PA, say 5 points. This would occur after the age of 25, when a player is no longer being judged by his "potential" but by his actual performance on the pitch.

For example:

In FM, a player with a max PA of 130 will never get better than that, even if he somehow leads the EPL and Champions League in scoring and takes P.o.Y. honors. In real life? You better believe FM researchers would be inclined to boost that player's PA significantly for next year. With my system, however, you'd be able to see an extra 30 points or so to a player's PA under OPTIMAL circumstances before they hit their "peak" around the age of 31-32.

Some good ideas there, however, I still think that we are forgetting that we shouldn't be comparing PA with real life. PA is a system put in place by SI (a hidden stat!) that is used to level players maximum ability. If a player is good enough to lead the EPL and Champions League in scoring then this is because he has a high PA (and CA of course).

I like the idea of splitting the technial / mental / physical stats into multiple PA stats, I think this may add some real benifits. This would also link in nicely to different training scenarios etc.

I couldn't agree more on the attribute separation. As it currently stands, the physical stats are weighed far too heavily against the technical ones, meaning pretty much any player with a PA under 150 is stuck being either a. a speed demon/genetic freak that'd be lucky to connect two dribbles without losing the ball, b. a wizard on the ball that can be blown over by a strong breeze, c. a creative dynamo that couldn't outrun a sloth, or d. a jack-off-all-trades that doesn't impress at anything.

There must be some way to add a modfier to reflect social conditions tied with whatever stats are applicable. For example the determination factor. Not all players can be botherd with reaching the heights that were assumed by whoever. The regens are afterall code clones and some players in the RW potentially don't reach what is expected. Michael Branch for one. Surely it would make sense to broaden the PA to reflect the RW than have a player in the game that has no comparative to the real player?

They are all very good points, but I don't think they should be determined from one pot of PA. Why not 3? One for technical, one for mental, one for physical? That's much more realistic.

Why not 3? Because it doesn't add anything, it just makes it more complicated.

Let's say Fabregas has a CA of 165 and a PA of 175. As he has low physical attributes, he won't improve them much, and he's already about as good as he'll get technically and mentally due to having nowhere to improve to.

Now, let's say he has a physical CA of 70 and a physical PA of 75, a technical CA of 180 and a technical PA of 185, and a mental CA/PA of 150/160. Same situation, no?

It would also make it harder to compare players in research, and would make training harder to understand.

All three factors determine how good a player is, so combining them all makes sense. After all, splitting them would naturally lead to a "crossing PA", "corners PA", "dribbling PA", and so forth.

Would that be a problem, SCIAG?

I could safely say that unless a miracle happens, Shaun Wright-Phillips will never be as good a crosser as David Beckham so his Crossing PA should be less than 20, maybe 14.

Generalise it the other way round - we let attributes develop naturally without any sort of cap, so in theory we could generalise it further and let CA develop without any sort of cap, letting nature take its course and ensuring Shaun Wright-Phillips will likely never become as good as Becks at crossing barring a miracle.