Potential Ability - should it exist?

[Sir Stuart Wilson]

The limit refers to ability. Everyone has potential at everything not just football. There is always a limit to how good anyone can get no matter what they do. I think you are misunderstanding me. If both average 9.50 over a season they most like have a very high limit. Any player in real life currently has a limit to how good they can become no matter what they do.

Yes but the way PA is guessed by a researcher and has then a set limit regardless. Potential is really not that simple. You look at a player with the attributes to be a player and pretty much assume he has the ability(CA) and then with proper coaching and nurturing,confidence, the right breaks and a whole lot of luck he may just begin to improve upon his current ability and talent(attributes) therefore how could you possibly say whether or not he can improve or not by setting an unknown number at the start of the game regardless how the player is handled?

[jod123]

Yes but the way PA is guessed by a researcher and has then a set limit regardless. Potential is really not that simple. You look at a player with the attributes to be a player and pretty much assume he has the ability(CA) and then with proper coaching and nurturing,confidence, the right breaks and a whole lot of luck he may just begin to improve upon his current ability and talent(attributes) therefore how could you possibly say whether or not he can improve or not by setting an unknown number at the start of the game regardless how the player is handled?

PA is the best a player can become with everything going his way. So he has lots of luck the right breaks, training, nurturing, playing time, no serious injuries and everything.

Just because they have a certain PA does not mean they will reach it. It is just the best they can become in the ideal circumstances.

[Sir Stuart Wilson]

PA is the best a player can become with everything going his way. So he has lots of luck the right breaks, training, nurturing, playing time, no serious injuries and everything.

Just because they have a certain PA does not mean they will reach it. It is just the best they can become in the ideal circumstances.

So why does pretty much every player in FM reach that potential surely they don't all have access to the best coaching, facilities, luck etc....

You are just advocating a very good reason for the current system to be changed.

[jod123]

So why does pretty much every player in FM reach that potential surely they don't all have access to the best coaching, facilities, luck etc....

You are just advocating a very good reason for the current system to be changed.

They don't as far as I know. I think I read in the manual before that very few players will reach their PA.

Anyway if they did all reach it then that does not mean that a fixed PA should be removed, it would mean the training and so on should be tweaked to have a bigger impact. A fixed PA reflects reality. Another thing is nobody would know that there was a fixed PA or anything if they did not use editors so I don't see the problem.

[DiamondPirate]

I think PA should stay, and still be input by researchers. However, on game generation there should be a random factor. So if a player has CA 100 PA 150 in the database then at the start of a new game it should assign a PA of anywhere between 100 and 175, but weighted so that it averages 150.

Note: 175 is a number I picked out of thin air - there should be a formula to work out the max - with the min being CA. I don't know what this formula should be, but the further PA is from CA the greater the spread of possible values.

Thats already been implented, for players who's potentials arent obvious, they are given a -#, which gives them a random PA within a certain brackets, IE -10 gives a PA of between 170-200, -9 is 150-180 (Not exact, but you get the picture)

[DiamondPirate]

So why does pretty much every player in FM reach that potential surely they don't all have access to the best coaching, facilities, luck etc....

You are just advocating a very good reason for the current system to be changed.

I've seen plenty of players not reach thier potential. Hardly any of my players fail to reach peak, but thats because I use them right, as do most players. Some kids that got hit bad with injuries though have failed to reach the potential, even with me. And its way more common with the AI.

[Sir Stuart Wilson]

They don't as far as I know. I think I read in the manual before that very few players will reach their PA.

Anyway if they did all reach it then that does not mean that a fixed PA should be removed, it would mean the training and so on should be tweaked to have a bigger impact. A fixed PA reflects reality. Another thing is nobody would know that there was a fixed PA or anything if they did not use editors so I don't see the problem.

Well the training and coaching have been the poorest aspects of the CM/FM series and have shown no improvement at all either in content or implementation and in game are pretty much cosmetic in relation to the game.

Maybe the actual PA system of only 0-200 is now too low and compact and maybe should be increased to 0-1000 but again I believe the game has to come away from its' reliance on the CA PA system and become more focused on attributes linked to performances etc...

I also like the idea from the guy earlier of recalculating PA every 6 months or so as a sort of summary of that players past three months such as the expression "You are only as good as your last game".

[jod123]

Well my opinion is PA should be fixed as that is the way it is in reality. CA should obviously fluctuate and that is where "You are only as good as your last game" is accounted for.

For example, in reality Drogba is nowhere near as good this season as he was in other seasons but that does not mean his "PA" in reality is reduced. He still has the same potential but has just lost some of his CA for whatever reason.

[Sir Stuart Wilson]

Well my opinion is PA should be fixed as that is the way it is in reality. CA should obviously fluctuate and that is where "You are only as good as your last game" is accounted for.

For example, in reality Drogba is nowhere near as good this season as he was in other seasons but that does not mean his "PA" in reality is reduced. He still has the same potential but has just lost some of his CA for whatever reason.

So why did he not have the same potential on previous versions of the game?

If we went by his attributes, and ability you could obviously have seen he would be a decent player if played properly in a decent team hence his huge rise in PA since that happened to him at Marseille and then Chelsea.

[jod123]

So why did he not have the same potential on previous versions of the game?

If we went by his attributes, and ability you could obviously have seen he would be a decent player if played properly in a decent team hence his huge rise in PA since that happened to him at Marseille and then Chelsea.

Because the researchers get it wrong obviously. If you go back in previous versions of the game was Rory Delaps long throw attribute at 20? I don't know if it was or wasn't but researchers can get it wrong. I would be pretty sure it was not a few years ago anyway. Players are re-evaluated every year by the researchers and so all CA and PA and all the other stats are looked at once again and changed if the researcher feels they should be.

[Sir Stuart Wilson]

Because the researchers get it wrong obviously. If you go back in previous versions of the game was Rory Delaps long throw attribute at 20? I don't know if it was or wasn't but researchers can get it wrong. I would be pretty sure it was not a few years ago anyway. Players are re-evaluated every year by the researchers and so all CA and PA and all the other stats are looked at once again and changed if the researcher feels they should be.

That is why we are having the discussion of whether to have PA. And it proves SI should be looking at developing the current ratings system as it is too basic and very limited.

[jod123]

Why because researchers can be wrong? They can be wrong on CA just the same so how are you supposed to simulate the players in the game realistically? They can be wrong on anything not just PA.

[x42bn6]

The limit refers to ability. Everyone has potential at everything not just football. There is always a limit to how good anyone can get no matter what they do. I think you are misunderstanding me. If both average 9.50 over a season they most like have a very high limit. Any player in real life currently has a limit to how good they can become no matter what they do.

I disagree. The only limits a human being has are those which are dictated by extreme limits - for example, human beings will not be able to lift up buildings by themselves any time soon, as our body structures cannot allow any further than a certain amount.

And you're wrong about the "high limit" part. I had a youngster playing very well in my youth and first-teams - he averaged around 7.30 in his first-team league appearances. However, I found out later that he wasn't developing - because he was at 130/131 or something. It is hard to believe that he could not be any better than this as he was still young. Players don't suddenly stop developing by playing well. They stop developing when playing badly.

Let's take Arteta. I don't know what his CA/PA are but let's just say he's 170/175. Let's also say he moves to Barcelona, displaces Henry (say) and then forces his way into the first-team squad at Barcelona.

His previous limit has just gone up.

While it is true that a player will never reach a certain level, we do not know what that level is, so we should not try and guess it.

My guess is that this limit you speak about is the likes of "Blue Square North players will never make it in the Premiership". While this may be generally true, it can never be discounted. This "rule" can be "proved" without PA - because players at that level never develop enough to play in the Premiership. They do not need this artificial ceiling stopping the outside case where someone simply breaks through.

[Kewell08]

I actually agree that PA is a good idea as a cap on a player's development.

But I also agree with a fair portion of the OP -- that a scout can't look at someone and think, "Well he's not good for his age now, but he'll be a star later."

I read somewhere else something along these lines: "I saw Ryan Giggs when he was this old, and this kid is as good/nearly as good/better than that, so he could develop into a legend." But RT-- is right, there is no way of assessing a player's potential outside of current ability and age. That's why I hate seeing a scout report on a 19-year old which says he has reached his full potential, because that just doesn't make sense. Even if he isn't going to get any better, everyone in the world will assume that as a 19-year old he is going to get better.

[looknohands]

I've gone through many varying opinions following the numerous posts and I'm struggling to find an answer that can fix the possible problem. Having a fixed PA in the game, by its very nature, raises the issue of people exceeding expectations. In real life it's easy to say, "Oh, we were wrong about so-and-so, he had the potential to be a great player all along...we just didn't see it." That's the part that doesn't seem to be integrated into the game- say, for instance, I signed someone with a CA of 95 and PA of 100 for an MLS team and, as luck would have it, that very same player turns in the best season of any player in the league. In real life, an adjustment would be made to his PA for the following patch/release, as the player has proven himself to be the most talented in the league. In the game, however, nothing will happen, aside from his CA possibly maxing out. So what happens if the same player repeats as MVP the next season? You better believe that in real life a foreign club would come looking to sign him. In the game, what foreign club is going to shell out the cash for a player with a PA of 100 and no hope of "improving"?

My suggestion, which ties into much of what of others before me have recommended, is the concept of a fluctuating PA during the game (I'd italicize that last part if I could figure out how.) The current PA system would be used- researches and co. would still give as precise PA as possible. What would differ in my version, however, is that this starting PA number would be more of a reflection of the odds that a given player can maximize his potential, and also have the ability to shift both positively and negatively based on performance in training and on the pitch. Here's how it would be implemented:

A player is assigned a PA of 130. Obviously, the general consensus is that he probably (again, wish I knew how to italicize!!!) doesn't have what it takes to make it in the Premier League. So one of two things could happen: he could sign for a team in a lower division and see significant playing time, allowing him to prove his ability on the pitch and, should his performances be of sufficient quality, allow him the chance to play for a team at a higher level. In this case, his PA should steadily rise as the player proves himself at a higher and higher level and, should he prove himself to the point that a Premier team takes notice, he could fit into a Premier squad and have the chance to sink or swim. Should he disappoint at this level and perform poorly his PA should begin to decrease- it's obvious that he can't "cut it" on the highest stage and his true place may be in a lower league/division. The second option, however, would be for a Premier League team to sign him at the start, while his stats (both CA and PA) are low. Obviously the player would be expected to fail...but what if he doesn't? Surely it'd be tough to throw someone with a CA of 115 onto a Premier team and have them perform well, but if they somehow thrive shouldn't they be rewarded? Wouldn't it be natural to reconsider the player's potential? Say it was your team that signed the player, and over the course of the season (or several) with a combination of good training results and satisfactory performance on the pitch (even as a sub in limited action) it's my belief that the player should see his PA receive a boost. Granted, this would be very difficult and time consuming to pull off, but it should be a possiblity nonetheless.

In a nutshell:

Initial PA serves as an indicator of what league a player is expected to reach and limits the initial CA, preventing higher leagues from spending all of their time and resources on developing talent lower-rated talent and instead focusing on well-known players and "can't-miss" prospects. Performance on the pitch and in training (as well as injuries and limited playing time), however, are the ultimate deciders of a players true potential, and affect the PA attribute accordingly, meaning that although (in theory) every player in the game would be capable of playing in the EPL, few would be able to sustain their success and avoid injury long enough to climb all the way from a Blue Square side all the way to the top.

[Kewell08]

looknohands, I can sort of see where you are coming from but don't really agree.

The Drogba example was used above - he had a good season at Marseille, came to Chelsea and his in-game PA was put up by the researchers.

The problem here is that he always had the potential to play for Chelsea in the EPL, and it's just that nobody knew. Unfortunately, the way the game is designed requires the researchers to enter an absolute value as to the best they will ever be. In Drogba's case, that value should always have been as high as it is now. Of course it wasn't, because the researchers have no way of properly assessing that.

That's my main problem - not with the idea behind the system but the fact that it requires the researchers to provide an arbitrary stat out of 200 without very much to work from. Your system would work better in game but doesn't really mirror real life very well, because two good seasons doesn't actually make any difference to the best you can ever be.

So here's the challenge: finding a system that strikes a balance between realism and mirroring the real world and the practicalities of making the game work. I think the current system is too skewed towards the former -- it's basically a realistic representation of real life, but it can't work perfectly in game because nobody knows the maximum potential of any player.

[jod123]

And you're wrong about the "high limit" part. I had a youngster playing very well in my youth and first-teams - he averaged around 7.30 in his first-team league appearances. However, I found out later that he wasn't developing - because he was at 130/131 or something. It is hard to believe that he could not be any better than this as he was still young. Players don't suddenly stop developing by playing well. They stop developing when playing badly.

That is very different. My post was in response to a person who gets an average rating over the season of 9.50 so for that to happen they must have very high CA and PA.

[borivoje213]

Sorry if someone has already posted this (clearly an admission that I haven't read the entire thread)...

But surely if theres not need for potential ability then why have a current ability atall? Why can't a player's ability simply be a product of their existing attributes. Why can't their attributes dictate everything about how good a player is, after all, they're supposed to represent every fascet of a football player's game.

After all, in real life isn't the awareness of a player's potential down to spotting that he shows a spark in a particular area of his game? Something that sets him above the rest and shows that he has a flair for some aspect of the game.

I'm inclined to agree with the thread creator, it would be more realistic to see people developing as a result of coaching environment and career stability.

Unfortunately I just don't see how someone could make the randomness of it all work. People have different opinions of players in the real world, they see a different player, not a set of numbers which quite clearly tell you all about them. Which is why sometimes they get it wrong and players fizzle out, or sometimes a player goes overlooked.

......perhaps if players attributes were all constantly invisible and you had to make your mind up based on how you feel they played and the opinions of other people around you. After all, thats about as realistic as you can get. Match ratings? every paper gives them so thats realistic, and i don't always agree with a player's rating. star ratings and profiles from coach reports and scouts? they're opinions of others so thats fairly realistic.

who'd be up for the challenge of that?

[Kewell08]

I think CA needs to exist only because of PA. It really doesn't, because you can just look at the attributes. But there does need to be some sort of PA to cap players' development, and that can only work if there is a corresponding CA to cap. I just ignore CA generally, seeing as you can't see it in game anyway.

[looknohands]

looknohands, I can sort of see where you are coming from but don't really agree.

The Drogba example was used above - he had a good season at Marseille, came to Chelsea and his in-game PA was put up by the researchers.

The problem here is that he always had the potential to play for Chelsea in the EPL, and it's just that nobody knew. Unfortunately, the way the game is designed requires the researchers to enter an absolute value as to the best they will ever be. In Drogba's case, that value should always have been as high as it is now. Of course it wasn't, because the researchers have no way of properly assessing that.

That's my main problem - not with the idea behind the system but the fact that it requires the researchers to provide an arbitrary stat out of 200 without very much to work from. Your system would work better in game but doesn't really mirror real life very well, because two good seasons doesn't actually make any difference to the best you can ever be.

So here's the challenge: finding a system that strikes a balance between realism and mirroring the real world and the practicalities of making the game work. I think the current system is too skewed towards the former -- it's basically a realistic representation of real life, but it can't work perfectly in game because nobody knows the maximum potential of any player.

I completely understand where you're coming from...What I'm trying to do is make the game less reliant upon the fixed PA as determined by a researcher. My hope is that with my system the game could accurately reflect a situation like Drogba's, wherein a player from a lower league that comes to a higher league and excels there would see an adjustment that reflects what a researcher (or anyone using the editor) would have to implement themselves. Personally, I wouldn't like it if my new signing somehow tears up the division one year against the odds, but is crippled the following years by his low PA. With my system, that player would gain PA, (theoretically) giving him the chance to continue his domination for an extended run. The flip side would apply, as well. That promising wonderkid you spent a fortune on in the offseason: watch his value plummet as his poor play indicates his potential wasn't nearly as high as hoped! Yeah, I know this technically can occur with the current set-up (not reaching his full potential and all) but it's the idea of a concrete ceiling that limits a player's potential that irks me, and this is the best solution I can come up with.

[davesearle1984]

......perhaps if players attributes were all constantly invisible and you had to make your mind up based on how you feel they played and the opinions of other people around you. After all, thats about as realistic as you can get. Match ratings? every paper gives them so thats realistic, and i don't always agree with a player's rating. star ratings and profiles from coach reports and scouts? they're opinions of others so thats fairly realistic.

who'd be up for the challenge of that?

That is EXACTLY what I think it should be like! As long as AI managers/scouts/coaches etc dont go by attributes or PA/CA.

Make it so every player has a fixed value as they do now, but make them all invisible...all the time. Give the user the option to add their own values or get their scouts/coaches to asses the players and give the attributes they think, or a range of attributes they think (eg Pace 11-14), and the less they have seen on the players the wider the range. The more well known the players, the more well known/accurate the stats. Or if someone is playing in a poorer league than you they can appear to be amazing, with fantastic stats, and then fall short when they come into the league you're in. Also if you sign someone who is unknown, they may play brilliant at first chasing everything and never stop running (like Zaki) and you can see their "work rate" stat at 18-20, even if the true value is 10, but they were just trying extra hard as a new player trying to prove themselves, and eventually either the user could amend it or the coaches would spot it and change their mind too.

I think this would add an extra dimension to the game, extra realism. I want to sign a flop. I want to spend money on a superstar and get a Rebrov. Because at the moment you can look at any player, get a scout report immediately, see all/most of their stats and know if they are any good. Its just too easy to find quality players.

Anyway thats completely off point so sorry about that

[Ljuba82]

If you don't look at the potential abilities of the players then it works quite nicely. The only problems that arise are when people start looking at values in the game that are not supposed to be seen.

Agree...

Young players with very high PA are very good when they are young (have CA of 100+).

They have to be, otherwise you couldn't get new Messi and Fabregas until they are 20.

So it's ok to me, in average better younger players have bigger PA.

[ddw31089]

I appreciate this thread, but I think the argument that PA should not exist is flawed.

This argument is best summed up like this. Do you feel any randomly selected human being can theoretically develop into a Pele or Maradona? If you don't feel they can become as good as Pele or Maradona then you are saying there is a limit somewhere, but you are just no defining it. Which means you actually feel there is a PA.

While it is true people can continue to improve with the right environment and right coaches, there will be a point where any future improvement is basically non-existent due the physical and mental limitations everyone has. A player can keep improving, but at some stage the improvements will be so limited because of these limitations that when the improvements are analysed statistically in the framework of a video game any improvements are meaningless.

What I mean by that is how you can look at a flattening curve on a graph, and say that while the value is still growing for example from 1.99 to 1.991 to 1.992 to 1.9922 and etc that it will never reach 2. I think this is the same way we can look at PA. If it makes those against PA feel better you can instead feel that no players reach their PA, but that those players who come closest just come infinitely close.

Though I do think less players should come closer to reaching their PA.

[Kewell08]

The problem with PA is that researchers can't get it right so the system doesn't work well. That is all.

[JimbobWWFC]

I appreciate this thread, but I think the argument that PA should not exist is flawed.

This argument is best summed up like this. Do you feel any randomly selected human being can theoretically develop into a Pele or Maradona? If you don't feel they can become as good as Pele or Maradona then you are saying there is a limit somewhere, but you are just no defining it. Which means you actually feel there is a PA.

While it is true people can continue to improve with the right environment and right coaches, there will be a point where any future improvement is basically non-existent due the physical and mental limitations everyone has. A player can keep improving, but at some stage the improvements will be so limited because of these limitations that when the improvements are analysed statistically in the framework of a video game any improvements are meaningless.

What I mean by that is how you can look at a flattening curve on a graph, and say that while the value is still growing for example from 1.99 to 1.991 to 1.992 to 1.9922 and etc that it will never reach 2. I think this is the same way we can look at PA. If it makes those against PA feel better you can instead feel that no players reach their PA, but that those players who come closest just come infinitely close.

Though I do think less players should come closer to reaching their PA.

Nicely said. I agree with this post, right up until the last sentence

[eaglealan64]

Does it matter if two players have the same current ability at 16, play the same number of competetive games, have the same training mentalities etc.? If one is more naturally gifted, he will develop to be a better player.

Exactly - that's what PA reflects.

[sten_super]

I have read most (but not all) of this thread, so apologies if I'm repeating what others have said.

The PA/CA system is not perfect, but I think it's the best we can hope for in the near future. A lot of the issues with it go away if you don't observe it using an editor. However, there are a couple of issues with it, which I've bought up before (for FM08 I think) that remain.

1) The fact that scounts use CA/PA. How this is still the case I'm not sure. It's a massive oversight. A scout should be telling me things along a similar line to those I can work out myself, such as 'is strong in the tackle', or 'looks better than most other centre backs his age', or similar. He should not simply be looking at CA or PA and making that the basis for the report. The most chokingly obvious example of this is, as others have mentioned, when you get a scout report for a 19 year old saying 'is unlikely to improve further'.

2) I'm not sure if this is still an issue in 09, as I've not used any editors to check. But in previous versions far too high a percentage of players were reaching their PA. What percentage of players play to the full level of their ability? Not a lot, I'd wager. There also isn't enough variation in this value for me. I think form should have more of an impact. I don't want to see a 21 year old (or younger, as mentioned above) reach his CA, therefore never improving again, and staying with the same level of ability for 8 years. He will have good times, he will have bad times, when he couldn't hit the proverbial cows bottom with a stringed instrument, and these should be reflected in his CA.

[Sir Stuart Wilson]

I completely understand where you're coming from...What I'm trying to do is make the game less reliant upon the fixed PA as determined by a researcher. My hope is that with my system the game could accurately reflect a situation like Drogba's, wherein a player from a lower league that comes to a higher league and excels there would see an adjustment that reflects what a researcher (or anyone using the editor) would have to implement themselves. Personally, I wouldn't like it if my new signing somehow tears up the division one year against the odds, but is crippled the following years by his low PA. With my system, that player would gain PA, (theoretically) giving him the chance to continue his domination for an extended run. The flip side would apply, as well. That promising wonderkid you spent a fortune on in the offseason: watch his value plummet as his poor play indicates his potential wasn't nearly as high as hoped! Yeah, I know this technically can occur with the current set-up (not reaching his full potential and all) but it's the idea of a concrete ceiling that limits a player's potential that irks me, and this is the best solution I can come up with.

That is a pretty much 'bang on' scenario and statement and proves itself in that every year promoted teams to other leagues have their players CA/PA/Attributes increased by the researchers.

Proving the very limited ability of the current system which I have no doubts leads to some of the long game problems regarding players/transfers etc...

[pires29]

So why in Football Manager games before did his PA not represent this?

Answer... Because his performances didn't warrant it. All indicative to the fact you are only as good as your last performance, as I said there should be no roof for players to hit, some click with certain teams, managers, tactics, areas etc... there is no magic rule.

Performances should be linked to CA, not PA. Take Ronaldinho for example, 3 years ago he was arguably the best player on the planet based on his form so his CA would have been very high but looking at last seasons performances he was a much poorer player, therefor his CA should have been decreased. His potential however should stay the same as he COULD get back to the level of perfomance he had 3 years ago.

In reference to Ronaldo, his PA did not reflect his level in previous games to this one is because it is nigh on impossible to predict exactly how good a player is going to be in the future due to the many variants which could cause PA to change.

[JimbobWWFC]

Of course PA should exist. The problem is when people use things like FMRTE to find out a player's PA. For example, I have never thrown a javelin in my life, so my CA could be around 20 out of 200. I am never likely to get world class distance on my throws by now, even if I were to train for many years. I would get a lot better, but I will never be world class. Therefore my potential is still the same, it is just a question of how close I come to reaching it. Just because a player has a hot season, this shouldn't mean his PA is increased, it means that his CA is closer to (or the same as) his PA. Your PA should never change, because improving as a player is a question of "reaching" your potential, and there will be a point where you cannot improve anymore (due to physical or psychological reasons... as ddw31089 already stated above)

[glamdring]

If you don't look at the potential abilities of the players then it works quite nicely. The only problems that arise are when people start looking at values in the game that are not supposed to be seen.

I might be repeating what others have said here since I am in work and don't have time to read all the responses, but here's my two penny-worth.

I almost never look at PAs (I don't have any kind of scout/real time editor programs so the only time I see them is in the pre-game editor which is very rare and I'm more interested in regens through my youth system then real players anyway), but the stars that your scouts give are based on these. Sure there are modifiers for the scouts' ability and maybe other things, but if I am managing in The Championship, with asperations of reaching the Premier League in a few seasons it's a fair bet that if my Ass Man and all my coaches tell me that a new youth player has the potential to be a decent League 2 striker in the future that he will never become a Championship or Premiership striker. Scots/coaches make mistakes, but if 7 or 8 of them all agree that his limit will be just to be "decent" at 2 or 3 leagues below the level I am at that is pretty much 100% certain.

Even putting that aside, for me it isn't that I look at all the PAs of my new youth players and know they won't be good that makes it disappointing, it's just the knowledge that they are set in stone, even if I don't know them. Even if I ignore my coaches and never use a scout program, if none of my youth players has PA above 50 then whatever I do in training and matches they will never be any good whatsoever which disappoints me. It's different if they have a high PA - they may never reach it so there is some interest and uncertainty there, but if a guy has a PA of 50 then he is 100% certain never to go above that.

[x42bn6]

That is very different. My post was in response to a person who gets an average rating over the season of 9.50 so for that to happen they must have very high CA and PA.

Disagree. The top scorer in La Liga last season was Dani Güiza - not the best in terms of CA or PA. It is possible for a player who is poor in terms of CA to do very well in a top league.

[Scoham]

Very wrong. Everyone has a limit to how good they can be, whether that's football or something completely different.

And what dafuge said, if you don't look at hidden values like PA it's not a problem.

Player 1 and player 2 would both have good (but random) potential at 16 if they were both already 100 CA players at that age.

[x42bn6]

I appreciate this thread, but I think the argument that PA should not exist is flawed.

This argument is best summed up like this. Do you feel any randomly selected human being can theoretically develop into a Pele or Maradona? If you don't feel they can become as good as Pele or Maradona then you are saying there is a limit somewhere, but you are just no defining it. Which means you actually feel there is a PA.

This is true. I put this in the "extreme limit" scenario. However, we have no way of determining how far we get to Maradona's - and the various players roughly on par with him - level.

With the reasoning up there, you could say that "you're only going to be worse than Maradona". Unfortunately, there are 200 CA points from which to choose from in this case. Pick one?

While it is true people can continue to improve with the right environment and right coaches, there will be a point where any future improvement is basically non-existent due the physical and mental limitations everyone has. A player can keep improving, but at some stage the improvements will be so limited because of these limitations that when the improvements are analysed statistically in the framework of a video game any improvements are meaningless.

This is true, but the only true limits we have are the ones where a human body cannot take it because of science rather than mentality. We cannot move planets around. However, if we cannot move a huge rock, there is every chance we could physically develop in order to do so.

Bodybuilders who started skinny are an example of where you can surpass your initial mental boundaries and go further. Yes there comes to a point where development is impossible further, but that sort of scenario is the World's Strongest Man - something like Strength 20?

What I mean by that is how you can look at a flattening curve on a graph, and say that while the value is still growing for example from 1.99 to 1.991 to 1.992 to 1.9922 and etc that it will never reach 2. I think this is the same way we can look at PA. If it makes those against PA feel better you can instead feel that no players reach their PA, but that those players who come closest just come infinitely close.

Though I do think less players should come closer to reaching their PA.

I prefer a logarithmic sort of approach with a truncated pure maximum limit. In some models, it makes no sense to allow infinite models - take, for example, human ages - it is not impossible, if unlikely, for a person to live to 200 years old - so they put in a maximum limit and rescale their probability distributions.

To me, the only limits are the pure limits of human physicality, and that with the correct mindset, anyone can push their boundaries. The problem is, age is a factor. It is too late for players at, say, 28, to decide that they want to develop 100% better. However, they do not simply stop developing. They continue to develop. They continue to get better if slowly.

[x42bn6]

Very wrong. Everyone has a limit to how good they can be, whether that's football or something completely different.

And what dafuge said, if you don't look at hidden values like PA it's not a problem.

Player 1 and player 2 would both have good (but random) potential at 16 if they were both already 100 CA players at that age.

True, Pirlo will never be better than he was at his peak, for example. However, do you have a crystal ball to tell you what that limit is?

If no, does it make sense to try and guess what it is, considering circumstances might change? See Drogba.

[Ah Fat]

Bodybuilders who started skinny are an example of where you can surpass your initial mental boundaries and go further. Yes there comes to a point where development is impossible further, but that sort of scenario is the World's Strongest Man - something like Strength 20?

Agree. Then how many of us can achieve this?

So those who are unable has limited potential.

[Ah Fat]

True, Pirlo will never be better than he was at his peak, for example. However, do you have a crystal ball to tell you what that limit is?

If no, does it make sense to try and guess what it is, considering circumstances might change? See Drogba.

If the researcher do not try to guess what those values are, then anybody can become Maradona in game.

[Sir Stuart Wilson]

If the researcher do not try to guess what those values are, then anybody can become Maradona in game.

Well maybe we should put in accurate attributes for each player and let the computer decide based on all the factors of how this player could potentially improve.... Game time/ performances/ training/ mentality/ happiness/ confidence/ age etc....

Rather than an external researcher putting in an absolute value for each player?

Like was said previous the game could assess each condition and circumstance of a player every 6 months and gauge from there how far they could improve or how far their potential may drop if at all?

[pires29]

Bodybuilders who started skinny are an example of where you can surpass your initial mental boundaries and go further. Yes there comes to a point where development is impossible further, but that sort of scenario is the World's Strongest Man - something like Strength 20?QUOTE]

I may have picked you up wrong here but everyone has the potential to become a bodybuilder, it is not a skill that you are born with so it's impossible to compare that with being a talented football player, yes with training you can improve someones ability as a football player but this will only increase to the level that that human being is capable of achieving.

It is not possible to become better at something than you genetic make up allows you to be.

[Scoham]

True, Pirlo will never be better than he was at his peak, for example. However, do you have a crystal ball to tell you what that limit is?

If no, does it make sense to try and guess what it is, considering circumstances might change? See Drogba.

No but researchers can have a good guess.

Yes of course it makes sense to try and guess. Researchers make mistakes, it's impossible for them not to, but there's a lot of examples of researchers getting it right, and that's not just for the top players.

As already explained, Drogba always had that potential, if it was lower in previous games it was because the researcher underrated his potential.

[Scoham]

Well maybe we should put in accurate attributes for each player and let the computer decide based on all the factors of how this player could potentially improve.... Game time/ performances/ training/ mentality/ happiness/ confidence/ age etc....

Rather than an external researcher putting in an absolute value for each player?

Like was said previous the game could assess each condition and circumstance of a player every 6 months and gauge from there how far they could improve or how far their potential may drop if at all?

No reason for that at all.

Most young players have a random potential, and once players are first team regulars in their early 20s researchers have a good idea of what their potential is.

They might not get it right, but that doesn't make it wrong. Some players have the potential, but never reach it.

[llama3]

personally i think its absolutely fine as it is. there is a limit to how good some players are and should become

[Ant85]

I think some people are missing the point of a potential attribute here.

Sure two players may be similar Player A may have a PA of 140 Player B 180

Say their current ability is 100, if Player A has a good attitude, good facilities etc he will reach his potential. Player B may never progress beyond his current ability despite having a high PA if his attitude is wrong or has bad facilities.

The game does take these things into account!

Of course everyone has a limit to their ability, what you're suggesting is that anyone with motivation and the chance to use top facilities will become a world beater, that's just not true - look at how relatively few players make it through the youth ranks at teams like Manchester United and how many of these become 'World Class'

[Ah Fat]

Well maybe we should put in accurate attributes for each player and let the computer decide based on all the factors of how this player could potentially improve.... Game time/ performances/ training/ mentality/ happiness/ confidence/ age etc....

Rather than an external researcher putting in an absolute value for each player?

Like was said previous the game could assess each condition and circumstance of a player every 6 months and gauge from there how far they could improve or how far their potential may drop if at all?

Certain player just don't fulfill their PA in game.

In my game Inter's Destro has a PA of 180. Yet after 5 years later his CA didn't increase much and it is nowhere near 180. So PA is not a problem as a cap when players never reach it.

While in reality, certain players perform well, train hard, perform well in first team etc. But they don't improve. So the only explaintion is that they have reached their "PA".

[RT--]

I'm not arguing that all players should have unlimited potential. What will limit most player's potential is how good they are at the age of 15, 16, whatever it is they appear in the game. There's already a system in place, in the game, which limits how quickly CA will develop according to how many games players are playing in, at what level, their training facilities, their personality, etc. A slightly random element to the development curve allows for late bloomers.

One other possibility is having certain hidden stats which limit how certain attributes can develop. As an example, pace and acceleration should certainly be limited, strength should certainly be, jumping could be constrained by height, and perhaps some of the tactical attributes limited by an 'intelligence' attribute, or something similar.

My other point is about how the in-game AI estimates potential ability. I'm saying that the way the AI currently 'sees' potential abililty is unrealistic. It should see it in the same way that any of us would, by using current ability, age and maybe personality.

As an experiment, I just added 3 16-year old players to Manchester United's squad: Player1, Player 2 and Player 3. They all have CA100, and exactly the same attributes across the board, but their PAs are: 200 (Player 1), 130 (Player 2), 100 (Player 3). A quick glance at the Assistant Reports shows that their Potential Ability star ratings are given as 5, 4 and 3 stars respectively. So how could the Assistant possibly have come to that conclusion?

[chopper99]

I haven't read all of the posts in this thread, but I've had this discussion many times with many different people on these boards and no-one has yet come up with a better system than we currently have.

Firstly, Dafuge has already stated the most important fact - PA and CA are meant t be hidden. If they remain hidden as they should then you never know which players will turn out excellent and which will turn out a bit rubbish.

Secondly PA exists in real life. No arguments on this one please, every person has a limit to how good they can be at something. Look at it this way, if 10 16 year old players get taken on at Man Utd and get the same training and first team chances will all 10 turn out to have exactly the same ability? No, some will be better than others because some of those players had more potential.

All the arguments about 'well why was this player worse/better in previous games are simply because SI have to use unpaid researchers to gauge each and every players current and potential ability. On many occasions they'll get it right (there have been plenty of brilliant players that I first became aware of through FM), but of course they'll sometimes be less accurate. But this is still the only way to keep the database as up to date as possible.

So yes, PA should exist simply because every player (once they get to the age at which they enter the FM world) has a limit to how good they can be no matter where their career takes them. Quite simply, not every player can be as good as Ronaldo no matter how hard they try. The only thing that should change according to their current form and situation is their current ability, which in turn will govern their ability to reach their potential.

[chopper99]

As an experiment, I just added 3 16-year old players to Manchester United's squad: Player1, Player 2 and Player 3. They all have CA100, and exactly the same attributes across the board, but their PAs are: 200 (Player 1), 130 (Player 2), 100 (Player 3). A quick glance at the Assistant Reports shows that their Potential Ability star ratings are given as 5, 4 and 3 stars respectively. So how could the Assistant possibly have come to that conclusion?

Are they all playing in exactly the same position? I belive that assistants base their ratings on other players in the squad. So a right winger will be rated in comparison to Ronaldo. Or your scouts might not be 100% accurate all of the time, which is again perfectly realistic.

[Ah Fat]

In reality players go through trials before MU signed them.

And before trials they have a lot of game experience which we can't see in game.

So AssMan can judge from them.

[chopper99]

In reality players go through trials before MU signed them.

And before trials they have a lot of game experience which we can't see in game.

So AssMan can judge from them.

Yes, like Scholes, Giggs, Neville and co did when they were taken on. There will also have been many other players signed to the acadamy at the same time who didn't make it big.

Why? Because they never had the potential.

[Sir Stuart Wilson]

Yes, like Scholes, Giggs, Neville and co did when they were taken on. There will also have been many other players signed to the acadamy at the same time who didn't make it big.

Why? Because they never had the potential.

But to have potential they have to be good at their age/group/level in the first place!