Dynamic Potential in FM?

[forameuss]

Would be similarly interesting if for FM24, SI purely managed to obfuscate and hide the raw PA value so that it was nigh on impossible to view.  I imagine a lot of these complaints disappear.

[wazzaflow10]

On 16/07/2023 at 04:41, santy001 said:

But this in turn is the problem, you've got a top goalkeeper, already playing well and in a fantastic position. So what if he cannot develop any further? At a 154 CA in any position you can be playing as one of the best in the world if your attribute spread permits it. 

As someone who rates players for the game, we do rate players differently to how the game generates players. There is something of an obvious reason behind this however. We are humans guessing to the best of our ability what a players potential is likely to be. The game, when it generates a new player, knows that players potential. 

This gave me an idea while that might placate this dynamic potential crowd. It's somewhat of an in-between for the people who play with attributes completely off and those like me who want the ratings since I can't actually observe the players.

The attributes of the player should reflect what the coaches or scouts currently think rather than hard-coded values. Its almost meaningless when you get a report that says "player x is now a leading premier league player" when their attributes haven't changed at all. They're still the same player and by the time you get that message you know they're in form. The attributes should move with the coach/scout perception of that player. A player might only be a 140 CA/PA player but if they're consistently getting 7.2, 7.5, 8.0 ratings in PL games perhaps the attributes reflect a player at 150 or 160 or even 170! Its up to us then as managers to decide "is this their true ability or just a hot streak?" Likewise if a player starts on a poor run of form and their ratings drop. We'd have to question is this a decline in ability, reversion to actual ability, or just temporary blip in form due to fatigue or mismanagement etc.  

I think this would add some dynamic elements to the transfer market for the player as well since its easily an area a human can outperform the AI in.

Buying a hotshot youngster from a lower division? Maybe your scout told you he was a PL caliber player but he turned out to be in the right system at the right time. Now moving to a bigger club where he's no longer the focal point means his skillset isn't being utilized as efficiently so his ratings start decreasing. Do you off load him at a discount and cut your losses or try and help him rediscover his form because you believe he can be better?  

I think that would reflect reality a bit better since if someone was lighting the league up for long enough such that the general consensus was "player x was previously only thought to be a decent/good league player and is now a good/leading league player" they would presumably get a CA boost. I could be wrong in how you do your job so please correct me if that's not how it works. 

Should add the ME would use their actual CA ability (however it works) so that you can't cheese the game this is purely a skinning/UI exercise. 

[santy001]

@wazzaflow10 the game already has this in place, for the AI at least. I regularly lure the AI into overpaying for youngsters who are in a hot streak or have played adequately in a handful of games that don't have the potential or are less likely to fulfil their potential. I've also made it something of an expected pattern I can bring in mid 20's players who have under-performed at their last club and quickly turn them around and get a profit from them. 

This is from a network game with a friend and so there is a bit more pressure on me to buy players at their peak as well to ensure I remain competitive. But across the duration of our save I'm up £1.6 billion in the transfer market. There's only been 1 player I've sold against my wishes in that time and it was a player I agreed to sell if a £200m bid came in.

Giving game time to players at a point they can make an impact and in a system that suits them will yield results in terms of how they're perceived by the AI. 

Players make the same judgements but for very different reasons. We all have our own criteria by which we judge players. For many this will be attribute based as its the easiest place to start, but you can identify players under-performing/over-performing their attributes and such through a whole lot of methods in game. I don't think the game needs are artificial construct which ultimately would only serve to add confusion for players who couldn't get to grips with it or would feel cheated when players dropped to their actual levels. 

[Jonthedon26]

7 hours ago, wazzaflow10 said:

This gave me an idea while that might placate this dynamic potential crowd. It's somewhat of an in-between for the people who play with attributes completely off and those like me who want the ratings since I can't actually observe the players.

The attributes of the player should reflect what the coaches or scouts currently think rather than hard-coded values. Its almost meaningless when you get a report that says "player x is now a leading premier league player" when their attributes haven't changed at all. They're still the same player and by the time you get that message you know they're in form. The attributes should move with the coach/scout perception of that player. A player might only be a 140 CA/PA player but if they're consistently getting 7.2, 7.5, 8.0 ratings in PL games perhaps the attributes reflect a player at 150 or 160 or even 170! Its up to us then as managers to decide "is this their true ability or just a hot streak?" Likewise if a player starts on a poor run of form and their ratings drop. We'd have to question is this a decline in ability, reversion to actual ability, or just temporary blip in form due to fatigue or mismanagement etc.  

I think this would add some dynamic elements to the transfer market for the player as well since its easily an area a human can outperform the AI in.

Buying a hotshot youngster from a lower division? Maybe your scout told you he was a PL caliber player but he turned out to be in the right system at the right time. Now moving to a bigger club where he's no longer the focal point means his skillset isn't being utilized as efficiently so his ratings start decreasing. Do you off load him at a discount and cut your losses or try and help him rediscover his form because you believe he can be better?  

I think that would reflect reality a bit better since if someone was lighting the league up for long enough such that the general consensus was "player x was previously only thought to be a decent/good league player and is now a good/leading league player" they would presumably get a CA boost. I could be wrong in how you do your job so please correct me if that's not how it works. 

Should add the ME would use their actual CA ability (however it works) so that you can't cheese the game this is purely a skinning/UI exercise. 

This is very much like Out of the Park Baseball. You get your own scouts ratings of your players, plus a neutral league rating of the players, but they are just their personal ratings from them, and at the end of the day it is up to you who to pick based on the multitude of stats available in the game. I like this approach as it makes it feel a little more realistic to me. 

[wazzaflow10]

 

5 hours ago, santy001 said:

This is from a network game with a friend and so there is a bit more pressure on me to buy players at their peak as well to ensure I remain competitive. But across the duration of our save I'm up £1.6 billion in the transfer market. There's only been 1 player I've sold against my wishes in that time and it was a player I agreed to sell if a £200m bid came in.

Giving game time to players at a point they can make an impact and in a system that suits them will yield results in terms of how they're perceived by the AI. 

I'm very much in the middle of this right now in FM21. Real Madrid want Anthony Martial because I have him scoring goals for fun. I got messages stating transfers would be at 60M then 75M and now we're up to 90M. His attributes are exactly the same as they were when the season started. It becomes very easy to say I'll happily sell you for 90M because I can replace you with a better player who the AI has devalued b/c of form but I know has better attributes than Martial. I don't think it's an exploit its just easier to do as a human because we can be more principled on what we buy/sell. I think it's easier to make it harder for the human than it is to code the AI to behave like one.

5 hours ago, santy001 said:

@wazzaflow10 the game already has this in place, for the AI at least. I regularly lure the AI into overpaying for youngsters who are in a hot streak or have played adequately in a handful of games that don't have the potential or are less likely to fulfil their potential. I've also made it something of an expected pattern I can bring in mid 20's players who have under-performed at their last club and quickly turn them around and get a profit from them. 

I think it would be much hard to identify those players if you saw an underperforming player in their mid 20s and the attributes you saw were reflective of their underperformance. Now you'd really have to gamble on your scouts being right that a player has more potential than he's showing. I don't want to take too many liberties with your researching job but if a player had a bad season wouldn't they be docked a few points across their attributes in the next version? Maybe you've kept their PA the same but their CA has declined. I feel like once the game is released players have a pretty stable career arc of growth from 16-24, peak from 25-32, decline from 32+. And this is true for the population and how it should be if you looked at CA growth on average. However, If you look at the same player across different versions there's much more fluctuation of attributes to reflect their performance each year. 

 

5 hours ago, santy001 said:

Players make the same judgements but for very different reasons. We all have our own criteria by which we judge players. For many this will be attribute based as its the easiest place to start, but you can identify players under-performing/over-performing their attributes and such through a whole lot of methods in game. I don't think the game needs are artificial construct which ultimately would only serve to add confusion for players who couldn't get to grips with it or would feel cheated when players dropped to their actual levels. 

I don't know if it would be an artificial construct. It's more aligning the attributes with the star ratings your players have from coaches/scouts. It would fluctuate much like real players do year to year based on their performance. I don't think the swings need to be crazy like gaining/losing 5 points. But having upticks and downticks over the course of a players career based on their performance would add some dynamism to world imo. Nothing changes under the hood we're just masking the ability to 100% know their abilities. I think this would also add some importance to the JPA skill for coaches/scouts. I feel JPA right now is meaningless to hire because I can see their CA and once you scout a player you can just search your scouted players for their attributes and filter. JPP is much more important since it's something I can't know without an editor. 

I think it can be a toggle much like the fog of war for scouting so people who want to reduce the confusion or feel cheated don't need to play with it. I just think the game would feel more alive if both CA and PA were ultimately masked and you relied on the game telling you attributes and potential.