Current ability cost of of attributes - position breakdown

[lance101]

Years ago there was a very active thread on this forum discussing the current ability (CA) cost of different attributes. There was a lot good analysis done at the time (and a lot of speculation), but game fans did not have a tool that could allow them to verify their hypotheses. That has changed. The in-game editor, made available by SI to fans of the game through Steam, has an interesting feature, not previously made available to fans in the pre-game editor: an item labelled "recommended current ability" (RCA). This item lets gamers know, upon editing a player's attributes, what should be the CA chosen for the player, if the editing gamer does not want the game to revert some of the attribute changes.

What this feature also allows is for players to trace, position by position, the CA cost of different attributes. I have attempted to do that. The methodology I used was the following:

- I set all attributes to 8.

- Then, I set all positions to 1, except the position in analysis. (A player with 8 in all attributes, has 39 RCA in all positions).

- Then, for each attribute, I first set it to 1, recorded the RCA, and then set it to 20 and recorded the RCA.

- To obtain the CA cost of an attribute at a given position I used the following formula: (RCA when attribute is 20 - RCA when attribute is 1)/19

To validate the results, I repeated the process with all attributes set to 7, except for the attribute in analysis. Unfortunately, the results were not entirely consistent. This indicates that there is something else going on, and that the formulas are not as simple as I imagined. Upon finding this inconsistency, my initial hypothesis was that the formulas were somewhat non-linear (e.g. going from 15 to 16 in an attribute may cost more CA points than going from 8 to 9 in the attribute). My analysis of some of the attributes did not show this, but I did not have the patience to do the analysis to rule this out for all attributes. My new hypothesis is that not all units of CA have the same "thickness" (e.g. the increase in attributes allowed between CA 92 and CA 93 may be different from the increased in attributes allowed between CA 93 and CA 94). I found some evidence for this hypothesis, but I do not have the time or the patience to fully verify it.

Despite the caveats explained in the previous paragraph, I still think it is useful to let you all know of my findings. I believe that the true CA costs of different attributes are close to the numbers I found. The CA costs of attributes are presented in the next post.

[lance101]

GK:

Cost in CA units of 1 unit of the attribute - Attributes

2.5 - Handling, Reflexes

1.65 - Aerial Ability, Command Of Area, Communication, Kicking, One On Ones, Bravery, Concentration, Decisions, Positioning, Agility

0.92 - Throwing, Acceleration, Strength

0.6 - Weak Foot

0.35 - Anticipation, Composure, Leadership, Team Work, Balance, Pace

0.125 - First Touch, Creativity, Workrate, Jumping Reach, Stamina,Technique

0 - Eccentricity, Free Kicks, Penalty Taking, Rushing Out, Tendency To Punch, Aggression, Determination, Flair, Off The Ball, Natural Fitness

Side Defenders:

Cost in CA units of 1 unit of the attribute - Attributes

2.15 - Tackling, Concentration, Decisions, Positioning, Acceleration, Pace

1.2 - Marking, Anticipation, Agility, Stamina, Strength, Weak Foot

0.53 - Crossing, First Touch, Heading, Passing, Technique, Bravery, Composure, Creativity, Teamwork, Workrate, Balance, Jumping Reach

0.125 - Corners, Dribbling, Finishing, Free Kicks, Long Shots, Long Throws, Penalty Taking, Leadership, Off The Ball

0 - Aggression, Determination, Flair, Natural Fitness

Center Backs:

1.9 - Heading, Marking, Tackling, Concentration, Decisions, Positioning, Acceleration, Jumping Reach, Pace, Strength

1.5 - Weak Foot

1.05 - Anticipation, Agility, Stamina

0.42 - First Touch, Passing, Bravery, Composure, Leadership, Workrate, Balance

0.125 - Corners, Crossing, Dribbling, Finishing, Free Kicks, Long Shots, Long Throws, Penalty Taking, Technique, Creativity, Off The Ball, Teamwork

0 - Aggression, Determination, Flair, Natural Fitness

Midfielder Center:

2.25 - Weak Foot

2 - Passing, Creativity, Acceleration, Pace

1.15 - First Touch, Long Shots, Tackling, Technique, Anticipation, Composure, Decisions, Workrate, Agility, Stamina, Strength

0.47 - Dribbling, Finishing, Marking, Concentration, Off The Ball, Positioning, Teamwork, Balance

0.125 - Corners, Crossing, Free Kicks, Heading, Long Throws, Penalty Taking, Bravery, Leadership, Jumping Reach

0 - Aggression, Determination, Flair, Natural Fitness

Attacking Midfielder Center:

2.7 - Acceleration, Pace

2.25 - Weak Foot

1.8 - Passing, Creativity

1 - Dribbling, Finishing, First Touch, Long Shots, Technique, Anticipation, Composure, Decisions, Off The Ball, Work Rate, Agility, Stamina, Strength

0.37 - Tackling, Concentration, Positioning, Teamwork, Balance

0.125 - Corners, Crossing, Free Kicks, Heading, Long Throws, Marking, Penalty Taking, Bravery, Leadership, Jumping Reach

0 - Aggression, Determination, Flair, Natural Fitness

Wingers:

3.55 - Acceleration, Pace

1.8 - Crossing, Dribbling

1.63 - Weak Foot

1.05 - First Touch, Technique, Anticipation, Composure, Creativity, Workrate, Agility, Stamina

0.42 - Finishing, Long Shots, Passing, Tackling, Concentration, Decisions, Off The Ball, Teamwork, Balance, Strength

0.125 - Corners, Free Kicks, Heading, Long Throws, Marking, Penalty Taking, Bravery, Leadership, Positioning, Jumping Reach

0 - Aggression, Determination, Flair, Natural Fitness

Forward Center:

3.2 - Acceleration, Pace

2.15 - Weak Foot

1.55 - Finishing, First Touch, Heading, Composure, Off the Ball, Jumping Reach, Strength

0.85 - Dribbling, Technique, Anticipation, Agility

0.32 - Crossing, Long Shots, Passing, Concentration, Creativity, Decisions, Positioning, Workrate, Balance, Stamina

0.125 - Corners, Free Kicks, Long Throws, Marking, Penalty Taking, Tackling, Bravery, Leadership, Teamwork

0 - Aggression, Determination, Flair, Natural Fitness

I did not do this analysis for sweepers, wing backs, defensive midfielders and side midfielders because I do not use these positions in my current tactic. I may do it at some point if there is interest.

[lance101]

The comments I make below are based in the idea that the CA costs of different attributes reasonably reflect the impact of these attributes in the Match Engine for the different positions (if that is not case, then the issue is why should the attributes be weighted in this way, if they have very different impacts):

- Acceleration and Pace - They are crucial, for pretty much all positions, but in particular for Wingers and Strikers. This seems reasonable to me, but it may be a bit of a surprise for most that, especially for forwards, these attributes are by far the most important (more than double the weight of finishing).

- Weak Foot - The ability to use the weak foot well is crucial in almost all positions. This has been the case for a long time in FM. I feel that the importance of the weak foot is very overrated on FM. In particular, why should it matter for goalkeepers???

- Decisions seem to matter more for defenders than for other positions. I'm a bit surprised by this. In particular, given the description given to decisions in the manual, I thought the attribute would be the most important to midfielders.

- First Touch is a fairly important attribute in attacking positions. It is more important than technique and as important as finishing for strikers.

- Leadership is more important for GK and Center Backs than for other positions. Could this mean that leadership is not just about being a good captain, but may also have to do with coordinating the defense??

- Balance is a fairly unimportant attribute for most players, including for wingers and strikers. SI should clarify what the attribute really means.

- Teamwork is also a fairly unimportant attribute. I found that a bit surprising, especially for midfielders.

- Heading and Jumping are fairly unimportant for midfielders center. I do not think this is an accurate representation of reality.

- Long Throws - Even thought it always one of the least weighted attributes, I feel that weighting of long throws is too high, except for perhaps side backs. Most players will never even attempt them!

- Preferred foot - The preferred foot of players does not factor into CA for side players. While I can see the case for wingers, given that a left footed or a right footed left winger can be effective in different ways, it seems to me, that given how most teams play, a left footed left back is preferable to a right footed left back. This is not the case in FM. A left footed left back will have the same CA in FM as a right footed left back with the same attributes.

[TOUGHGUY]

You are doing god's work, son.

[duduric]

So, set pieces DO use CA....

[hannahrr]

I had a 15 year old regen that was 6'3," a winger, didn't have great passing, dribbling, or crossing, but had great tackling, speed, penalty taking, technique, teamwork and workrate. As a result the IGE stated he was rated 49 CA as a winger with 157 PA. With no editing (apart from his name, as he was called Simon Cowell) I turned him into a 95 rated DC in a year through training his heading and marking on heavy. I'm not a huge fan of the IGE, but the recommended rating thing is marvelous.

[CMdan44]

- Weak Foot - why should it matter for goalkeepers???

- Heading and Jumping are fairly unimportant for midfielders center. I do not think this is not an accurate representation of reality.

- Long Throws - Even though it always one of the least weighted attributes, I feel that weighting of long throws is too high, except for perhaps side backs. Most players will never even attempt them!

the first is obvious: back passes. If the ball is coming into a 'keeper's weak foot and he is being pressured by an opponent, his ability to effectively strike the ball with his weaker foot is crucial. Look at Boruc last month(?) shifting the ball from foot to foot before being dispossessed and conceding.

the second is fair, central midfielders rarely have to head the ball. If they can do so it can be an advantage but in most games where the ball is in the air it bypasses the midfield and when the midfield is being used a lot, it is usually with balls on the floor. Heading is not a vital skill in this position, but the ability to do it is an advantage, so it is right that it has just a little weight.

your third argument is flawed. Any player proficient in long throws will attempt long throws, I play for a team whose striker is a long-throw specialist. In a way, the argument is similar to the one about M©'s and heading, it's not vital for players to have a long throw but it is a bonus. Therefore it makes sense for it to have some weight when it comes to CA calculations.

[hannahrr]

I always make sure my CMs have good heading and positioning. Long balls from goal kicks often end up being contested by the CMs, so to have players that can be in the right spot to win those is crucial. The good thing also is that it doesn't bloat their CA and therefore their wage demands.

[mojojojo101]

I'd be interested to know how others will use this information to alter how they approach player development.

As I can see, you could either try to make use of the low CA cost of less favoured attributes, creating a more well rounded team or focus on high cost-high gain attributes to get players who may well be speciallists in their area of the pitch but are weaker when asked to do more complex things.

Initially, as a kind of gut reaction type thing I am tempted by the former option, in an attempt to for instance create a full back with very high finishing/off the ball/technique who can offer a significant goal threat from an area of the pitch the oppositon would not be a expecting a direct goal threat t come from.

[lance101]

the first is obvious: back passes. If the ball is coming into a 'keeper's weak foot and he is being pressured by an opponent, his ability to effectively strike the ball with his weaker foot is crucial. Look at Boruc last month(?) shifting the ball from foot to foot before being dispossessed and conceding.

the second is fair, central midfielders rarely have to head the ball. If they can do so it can be an advantage but in most games where the ball is in the air it bypasses the midfield and when the midfield is being used a lot, it is usually with balls on the floor. Heading is not a vital skill in this position, but the ability to do it is an advantage, so it is right that it has just a little weight.

your third argument is flawed. Any player proficient in long throws will attempt long throws, I play for a team whose striker is a long-throw specialist. In a way, the argument is similar to the one about M©'s and heading, it's not vital for players to have a long throw but it is a bonus. Therefore it makes sense for it to have some weight when it comes to CA calculations.

CMdan44,

I agree that I'm exaggerating when I say that the weak foot plays no role in a keeper overall ability. But should it be more important than a keeper's anticipation? Than a keeper's balance?

I disagree that central midfielders don't have to head the ball. I see some, like Yaya Toure, heading the ball all the time.

I think that long throws should have some weight. But I don't think they are on par, in terms of usefulness with corners, free kicks or penalty taking. Only a few teams ever employ them. Are the players in those teams the only players in the world with enough arm strength to send the ball to the area? I don't think so. It just isn't a common soccer tactic. Besides the set piece attributes, I don't think that long throws should be on par with dribbling for a side back (I think dribbling is under rated CA-wise for side backs - lots of side backs play like wingbacks and have large offensive roles); with teamwork for a center back; with jumping and heading for a center midfielder; or with tackling for a striker.

[samdiatmh]

what about the hidden attributes (that get achieved through natural progression) - how do they affect CA?

all this kinda shows is that those god-like newgens that you see are almost impossible (where they have like 20s everywhere)

brb, training my GKs like Chilavert, given that FKs take up no CA

also, can you try this for coaches too, i'm curious to see how this works

[lance101]

Hidden attributes have no CA cost.

I wanted to do the same for coaches, scouts etc, but the in-game editor tool does not show the recommend current ability (RCA) for staff, so it would be a lot harder and less accurate to try to analyze them and find the exact costs of different attributes.

[phnompenhandy]

This has enormous implications for training.

Correct me if I'm wrong, but doesn't it suggest that if you give a player indivdual training on quickness, it should take a long tine to increase one point in pace and acceleration, whereas indiv training on lower weihtings should result in more rapid increases.

[vlobben]

This is good work, but it raises more questions than it answers

1. How does retraining positions factor into this?

2. I assume these values are for the natural position. What about accomplished positions, awkward positions etc?

3. As the poster above suggested, how does attribute weight factor into speed of increase? Does it matter at all?

[jayahr]

Amazing work!

Interestingly this also leads to the conclusion that players with high attributes in the free categories are necessarily better than those with lower ones.

[braceletwinner]

I disagree that central midfielders don't have to head the ball. I see some, like Yaya Toure, heading the ball all the time.

I think that has more to do with his height and jumping ability than with being a midfielder. Loot at a team like Arsenal. Have any of their midfielders headed a ball in midfield, ever? I know that they have, but in general you want your midfielders passing the ball, so heading isn't as important. Useful if one has it, as in Toure's case, but not critical.

[braceletwinner]

2. I assume these values are for the natural position. What about accomplished positions, awkward positions etc?

What if a player is Natural in multiple positions? I have a player that is Natural at DL and DM. If he increases an attribute, is the cost averaged?

[GSevensM75]

Using the pregame editor, these multipliers don't seem right? I have used Brad Friedel as an example. His CA according to your ratios is 359. In the pre game editor it is 137 out of 200.

I may be missing something obvious here mind you, but is the IGE ranked out of 200?

The other thing I am wondering is whether key stats for a position aren't counted towards CA until they get to a certain point. E.g. first 5 points of Handling for a GK doesn't count towards his CA, but every point after that is 2.5 or something.

[lance101]

This has enormous implications for training.

Correct me if I'm wrong, but doesn't it suggest that if you give a player indivdual training on quickness, it should take a long tine to increase one point in pace and acceleration, whereas indiv training on lower weihtings should result in more rapid increases.

That is a possible theory, but I'm not sure if that is the case. I guess your intuition is that if it takes less CA to for a certain attribute to increase and a player spends an equal amount of time and effort training it, the attribute should increase faster. I guess that is a possibility, but I have no evidence for it. It may be that when you ask a player to focus on a low weighted attribute, the player actually focuses less time on it than when you ask a player to focus on high weighted attributes.

That being said, in a related matter, I know that acceleration, pace and I believe also agility and balance, become very hard to increase when players are no longer teens.

[lance101]

This is good work, but it raises more questions than it answers

1. How does retraining positions factor into this?

2. I assume these values are for the natural position. What about accomplished positions, awkward positions etc?

3. As the poster above suggested, how does attribute weight factor into speed of increase? Does it matter at all?

The weightings in the original post apply to a player with a single natural position and everything else at 1. I have not fully investigated what happens when players have more than one position, but I have learned some things. I'll illustrate what I have found with Cristiano Ronaldo as an example (CA spoilers for Ronaldo below):

- At the beginning of the game, Ronaldo's CA is 192. Ronaldo is natural at AM L; accomplished at S C and AM R; competent at AM C; and unconvincing at M R and M L.

- When I change Ronaldo to be natural at S C and every other position to 1, Ronaldo's recommend current ability (RCA) is 195. This means that Ronaldo's ability as a S C is higher than his CA. If I made these changes to Ronaldo's positions and didn't change his CA, within a few days the game would adjust some of Ronaldo's attributes down, so that his RCA would match his CA of 192.

- I also tried changing Ronaldo to natural as a D R and every other position to 1. When I do this, Ronaldo's RCA becomes 134. If I changed Ronaldo's positional attributes in this way and didn't lower his CA, within a few days the game would bump up his attributes across the board so that his CA as a DR would reach 192.

- Then I made Ronaldo natural at D R and S C. When I do this, his RCA is 178. This is between 195 (RCA as a SC) and 134 (RCA as a D R). However, it is not a simple average. In some earlier games, players with proficiency in many different positions had, often times, very very good attributes. I believe this had to do with the way the game averaged the RCA of different positions to calculate the overall CA. SI seems to have compensated for this potential bug by making the average biased towards the position the player is best at.

- When, besides being natural at D R and SC, I also make Ronaldo natural at D C (Ronaldo RCA when he is only a D C is 137), Ronaldo's RCA becomes 170. It seems like the weight given to the RCA in the position the player is less good at becomes even smaller when the player is proficient in many positions.

- I also tried starting Ronaldo as a S C only and then, one by one, increase his D R proficiency. Up until D R = 10, Ronaldo's RCA does not change. Starting at 11, Ronaldo's RCA steadily decreases as his DR proficiency increases, reaching 178 when his DR proficiency reaches 20. So, positions are "free" up to 10, and then gradually start to affect the CA weights of attributes.

So, my take is that the CA costs of attributes for players with multiple positions are an average of the CA costs of attributes of the different positions. However, this average CA cost is not a simple average (nor even a "simple" weighted average based on the proficiency in the positions). Rather, the CA cost of an attribute is "biased" towards the CA cost of the attribute in the position the player is best RCA-wise.

[lance101]

Using the pregame editor, these multipliers don't seem right? I have used Brad Friedel as an example. His CA according to your ratios is 359. In the pre game editor it is 137 out of 200.

I may be missing something obvious here mind you, but is the IGE ranked out of 200?

The other thing I am wondering is whether key stats for a position aren't counted towards CA until they get to a certain point. E.g. first 5 points of Handling for a GK doesn't count towards his CA, but every point after that is 2.5 or something.

A player with 6 in attributes has a RCA of 0 with any natural position (also when I changed finishing to 5 in a striker, the in-game editor showed a RCA of -1; and when I changed finishing to 7 in a striker, the in-game editor showed a RCA of 1). This implies that you cannot just multiply the CA costs I posted originally by the attributes and get a player's CA. You have to subtract from that calculation, approximately, 6 x sum of the CA costs of all attributes.

That being said, I still wouldn't expect you could get the exact CA of a player by doing that calculation. As I explained in the original post, there is some quirks in the way the whole CA weighting of attributes works that I couldn't fully figure out. But the number should be in the neighborhood.

[lance101]

Amazing work!

Interestingly this also leads to the conclusion that players with high attributes in the free categories are necessarily better than those with lower ones.

I agree with you for Determination and Natural Fitness, and for some of the hidden attributes (e.g. important matches and consistency). However, that is not that clear cut for aggression and flair. High aggression could be bad, for instance, for a bad tackler, as it can cause him to keep fouling. High flair can be bad for a player with low technique, because it can lead him to try flair moves that he can't complete.

[samdiatmh]

A player with 6 in attributes has a RCA of 0 with any natural position (also when I changed finishing to 5 in a striker, the in-game editor showed a RCA of -1; and when I changed finishing to 7 in a striker, the in-game editor showed a RCA of 1). This implies that you cannot just multiply the CA costs I posted originally by the attributes and get a player's CA. You have to subtract from that calculation, approximately, 6 x sum of the CA costs of all attributes.

That being said, I still wouldn't expect you could get the exact CA of a player by doing that calculation. As I explained in the original post, there is some quirks in the way the whole CA weighting of attributes works that I couldn't fully figure out. But the number should be in the neighborhood.

so saying this (using GMSevens75 Friedel example) then he has 37 stats over 6

conveniently the difference (222/6) gives a round number

so maybe attributes are CA free till they get to 6, then start costing CA at the rate posted above

care to test this out?

a ST with 6 pace and then with 20 pace (which should see a RCA increase of around 45 [(20-6)x3.2=44.8] according to your calculations)

edit: and positions seem CA free until they get to 10 in them (which I think is the orange colour)

love the work, it's really helpful for those that create new players, and what CA to give them

[lance101]

so saying this (using GMSevens75 Friedel example) then he has 37 stats over 6

conveniently the difference (222/6) gives a round number

so maybe attributes are CA free till they get to 6, then start costing CA at the rate posted above

care to test this out?

a ST with 6 pace and then with 20 pace (which should see a RCA increase of around 45 according to your calculations)

No, attributes are not "free" up until 6. According to my testing, their cost seems similar below and above 6. What happens is that a player with CA = 1 has an allotment for attributes that is approximately 6 x CA cost of all attributes. You don't start with all attributes at 1 when CA is 1.

A player with 1 in all attributes and 2 in corners has RCA = -99 (when I put all attributes to 1 the in game editor shows RCA=0 -> this is probably because the editor is not prepared for that extreme).

[Freakiie]

The positional stuff is highly interesting. Sounds like in theory you could take a player that hit his PA at reasonably low age, retrain him in a position for which he doesn't have the attributes to lower his RCA and then fill up the missing points with attributes for his main position.

What would happen if you'd remove Ronaldo's positions and just make him a natural GK/AML, turn his GK attributes to 1 (If that's possible with the IGE) and then let the game readjust his attributes so they suit a CA of 192?

[lance101]

The positional stuff is highly interesting. Sounds like in theory you could take a player that hit his PA at reasonably low age, retrain him in a position for which he doesn't have the attributes to lower his RCA and then fill up the missing points with attributes for his main position.

What would happen if you'd remove Ronaldo's positions and just make him a natural GK/AML, turn his GK attributes to 1 (If that's possible with the IGE) and then let the game readjust his attributes so they suit a CA of 192?

From my testing, Freakiie's strategy would only work for players that are proficient on a small number of positions. I just tried adding D R to Ronaldo (without removing any of his initial positions) and his RCA jumped to 195! The game seems to have some built in mechanism to penalize players with lots of positions (and thus to contain this exploit). But according to the results I discussed above (post #20 of the thread), the exploit still exists for players with only one or two positions.

I don't think you can retrain a outfield player to become a GK I added GK to Ronaldo and he become just a GK outside of the editor screen.

[dennie]

lance101 This is very interesting stuff have you tested some more positions/speciality role?

keep up the good work

[Johnyfooty]

I don't think the attribute gravity has changed much, so this should be correct. Can't recall from which version it is,I have it bookmarked for quite some time.

[flatLINE]

I must say that your estimates are pretty close. I tested the Attribute weight on a random regen Striker (see screenshots bellow). For full stats of the regen check the screenshots.

I've based my calculations on the assumptions that I've correctly understood the following, from the posts in this thread:

a) An attribute with a value of 6 has a weight of 0, therefore a player with 6 in all attributes will have CA 0

b) If an attribute is above/bellow 6 (ie. Corners 8, which above 6 by 2), that value is multiplied by the weight of the attribute and then added/substracted in the final breakdown of what the projected CA should be, along with all other attributes.

c) Weaker foot doesn't factor in for some reason, at least not with this regen player. The regen I tested your theory on has 4 in Weaker foot, which is 2 bellow 6 = -4.3 off from the final CA estimate.

So, assuming that I understood correctly how calculations work, these are the estimates for the regen's CA:

Note: Values in BOLD letters are attributes and their weight, values in GREEN increase CA, while values in RED decrease CA, and the final breakdown of CA is at the end of the list.

Physical

Acceleration 3.2 +19.2

Agility 0.85 +1.7

Balance 0.32 +0.96

Jumping 1.55 +1.55

Natural Fitness 0 0

Pace 3.2 +16

Stamina 0.32 -0.32

Strength 1.55 0

Mental

Agression 0 0

Anticipation 0.85 -4.25

Bravery 0.125 -0.625

Composure 1.55 -1.55

Concentration 0.32 -0.96

Creativity 0.32 -1.6

Decisions 0.32 -1.28

Determination 0 0

Flair 0 0

Influence 0.125 -0.5

Off the ball 1.55 -1.55

Positioning 0.32 -1.6

Teamwork 0.125 +0.125

Work rate 0.32 +1.28

Technical

Corners 0.125 -0.5

Crossing 0.32 -1.6

Dribbling 0.85 -3.4

Finishing 1.55 -1.55

First touch 1.55 0

Free kicks 0.125 -0.625

Heading 1.55 -3.1

Long shots 0.32 -1.6

Long throws 0.125 -0.625

Marking 0.125 -0.625

Passing 0.32 -1.6

Penalty taking 0.125 -0.625

Tackling 0.125 -0.625

Technique 0.85 -3.4

Weaker foot 2.15 -4.3 ??? (Note that I have not included the Weaker foot attribute in the final breakdown)

Projected CA 6.705 (Player's ingame CA is 5)

Projected CA (Weaker foot factored in) 2.405 (Player's ingame CA is 5)

In this case it seems that the final CA estimate was closer to the player's actual CA when Weaker foot was not factored in. However, it might prove much more reliable to test this by calculating a better player, ie. a high end player. Still, in both cases, the final estimate is rather close.

So, as you look at the screenshots, my calculations were, ie. Marking is 5 bellow 6, so 5 x 0.125 = 0.625. So it's -0.625 from the CA value. Once this is done for every attribute, they are all added/substracted depending on how much above/bellow 6 they are.

[flatLINE]

What just crossed my mind is that there is a possible hole in the whole SI attribute system, given that attributes have a different weight for every Playing Position. This in turn can maybe be exploited.

So for example, to make a superstar Defender:

1) I can buy a 15 year old regen Striker with as low CA as possible, high PA (170+), high Ambition/Professionalism/Determination, high Versatility and with around 15 in Acceleration/Agility/Pace, but otherwise very poor in all other attributes (6 or bellow, except when it comes to free attributes).

2) Put him on a training regime for Central Defenders to improve defending abilities, because training defensive attributes is cheaper for Strikers than it is for Defenders. Ie. Each Tackling point weighs 1.9 CA point for Defenders, but only 0.125 for Strikers.

3) The regen Striker can then get to 20 Tackling for only 2.5 CA, 20 Marking for only 2.5 CA, 20 Bravery for only 2.5 CA, 20 Positioning for only 2.5 CA, etc. All in all, getting ALL defensive attributes up as high as possible would be dirt cheap, since the regen is a Striker.

4) Once the regen Striker enters his 20s, train him in the Central Defender position, which he shouldn't have any problem adjusting to due to his Versatility. And voila, a "Natural" Striker that became an "Accomplished" Central Defender, and has tons of CA points to spare.

Of course, I have not tested this but I'm wondering if anyone has more info on whether or not this theory is correct/incorrect. I'm aware that the whole Training concept became much more automated and streamlined in the last two editions of FM, and isn't as in-depth and customizable as it used to be. But even with the new training system that offers less control over player development, if this could be implemented even to a limited extent, it would be a serious exploit. Needless to say, 'doping' players like this would give a huge unfair advantage to the Human Player, and if it's possible I'll start spying on other Human players in my network games, in case they attempt to resort to this.

[forameuss]

What just crossed my mind is that there is a possible hole in the whole SI attribute system, given that attributes have a different weight for every Playing Position. This in turn can maybe be exploited.

So for example, to make a superstar Defender:

1) I can buy a 15 year old regen Striker with as low CA as possible, high PA (170+), high Ambition/Professionalism/Determination, high Versatility and with around 15 in Acceleration/Agility/Pace, but otherwise very poor in all other attributes (6 or bellow, except when it comes to free attributes).

2) Put him on a training regime for Central Defenders to improve defending abilities, because training defensive attributes is cheaper for Strikers than it is for Defenders. Ie. Each Tackling point weighs 1.9 CA point for Defenders, but only 0.125 for Strikers.

3) The regen Striker can then get to 20 Tackling for only 2.5 CA, 20 Marking for only 2.5 CA, 20 Bravery for only 2.5 CA, 20 Positioning for only 2.5 CA, etc. All in all, getting ALL defensive attributes up as high as possible would be dirt cheap, since the regen is a Striker.

4) Once the regen Striker enters his 20s, train him in the Central Defender position, which he shouldn't have any problem adjusting to due to his Versatility. And voila, a "Natural" Striker that became an "Accomplished" Central Defender, and has tons of CA points to spare.

Of course, I have not tested this but I'm wondering if anyone has more info on whether or not this theory is correct/incorrect. I'm aware that the whole Training concept became much more automated and streamlined in the last two editions of FM, and isn't as customizable as it used to be. But, if this could be implemented it would be a serious exploit. Needless to say, 'doping' players like this would give a huge unfair advantage to the Human Player, and if it's possible I'll start spying on other players in my network games, in case they attempt to resort to this.

I have no idea what you just said...

But seriously, it's an interesting test. Obviously this is something that no-one would do off their own backs as a genuine strategy, but looking at your reasoning it's hard to argue against it. Whether it would work or not...not sure, you'd need a much smarter guy than me to work it out.

In terms of your final step, changing the player to be a defender. If you've gained all these attributes based on a player being a striker, so pretty much "cheap training", and then change him to a defender, what happens to his CA? Does it still mirror the weightings for a striker, or does this now change? Would this result in a crash? A CA-Mutant? The attributes rapidly falling? Would be an interesting test certainly.

[Johnyfooty]

The attributes would take a sharp decline to balance out, but he should be able to keep most of the stats high.

The problem is, you can't put a player to a role training regine without training in that role as well. Which means that in order to train him as a midfielded, you have to train him for the midfielder position as well at the same time.

Then he will become accomplished/natural in the MC position, which will already shift the gravity of his CA costs. And then the gravity will shift even more when training him as a defender.

Not to mention that the position training will take 40-50% of his regime, so it goes to waste.

And also, it's not that it's easier for a ST to increase his tackling. It is cheaper, not easier. The ease at which a stat increases is based mostly on how high that stat is. I.E, it's easier to train for 1->5 than it is for 19->20.

[flatLINE]

Then he will become accomplished/natural in the MC position, which will already shift the gravity of his CA costs

This pretty much fully answers what I've been wondering, and eliminates the possibility for any type of exploiting. Good. I've just checked myself, and confirmed that you cannot train a Striker for Central Defender's attributes. Oddly, I wasn't aware that you cannot put a Striker on a Central Defender regime, as I never had any reason to check if this is possible lol. Thanks for the input.

As for the rest of your reply, I'm aware of all of it. However, if not for the limit on what regime each position can undergo (ie. no CD regime for STs), the cheap factor would shine as an exploit eventually, regardless of how long it would take to get stats up (some game time would definitely speed things up), and once desired stats are satisfactory, position training would commence. Good to see SI thought of this, it would be a terrible exploit.

[Ljuba1982]

Nice thread. I remember you Lance from my old thread

http://community.sigames.com/showthread.php/58666-Just-found-something-amazing!!!-Must-see!!!?highlight=found

[Coolcup]

Does retraining a player to a new position cost CA points?

[toluteriba11]

Ive been forced to make an account because you sir, deserve to be commended for this work. What you have put up here is literally work we should pay you for! WELL BLOODY DONE!

[CurnRaisin]

Hi, Nice thread, I am always interested in the inner workings of the game mechanics.

On the point of retraining a player, although you cant do roles with out retraining your player in that position, as in a SC training in the roll of Central Defender, you could theoretically train your new 15yr SC with great potential in the new position of CD train him in the roll or central/ball playing defender up until the point he gets awkward(orange on the position screen) stop training in the new position, since you said up till a position rating of 10 it doesn't affect the weighting.

Then train him using the individual attributes focus rather than the role focus, or vice versa .

But this probably all wouldn't really be worth it as I think(just a theory) That the weighted attributes for each position also be trained on more, i.e regardless of your focus a CD will still improve on Tackling/marking a SC will still improve on Finishing first touch etc. so you will just end up using up less CA to get the defensive attributes up wile your player uses more up to get the SC attributes up so when you finally make the swap if will sorta balance out, it could work on well rounded players, regens with a lot of average stats or them regens that start with a lot of high physical/mental and lower technical

I do think that the position system needs a rework, especially on regens, as unless its a player controlled team you will rarely see a regen change positions similar to Garth Bale or even players like Stephen Gerard and Ryan Gigs. Bale went from DL->ML/AML/AMC/AMR/SC. While Gerard was a DMC/DR and went on to be mainly AMC/SC and now is dropping down to imo nearly play as a regista irl in a DMC role again, never mind the fact that in more than a few FMs Gerard could reasonable play nearly all the positions on the pitch. Regens tend to be a lot more ridged in their positions, sure you might find one that's natural DMC/MC/AMC/MR/AMR but that player will never leave them positions, IRL if you had a player like that I guarantee that his positions in FM would be something slimier to Gerard, Henderson or Bale with them being able to play a lot more positions up to awkward/competent as well

I had an idea I posted last year but, all the post history got wiped so I cant get it now, but basically it was that a players attributes determine their positions to an extent maybe up to competent or the threshold of it affecting a players CA (10), which is awkward if I'm not mistaken. So basically if you take the generic roles, and not the specialist roles(so like central defender, full back, central midfielder, wide midfielder etc.) the game looks at a players attributes and averages them for that role and put points into that position up till 10. Say 10 average will get 1 points in that position,11 average will get 2, 12 gets 4, 13 gets 6, 14 gets 8 and 15 gets 10. So for example you have a CD who also has good passing and first touch then he will get a few points in DMC and MC position. That way The AI would start using them in them positions and you would see players progress into different positions. Problem is that good players will end up with a load of awkward positions, and every MC will also have DMC and AMC, but I would much rather that if it made positions more fluid, rather than what we have now and never seeing the AI retrain players. It would need to be tweaked and better balanced but the current system of retraining positions is far to ridged and the AI never seem to use it unless a player already has at least awkward/competent in that position.

[poom15005]

Great job Lance ! Really appreciate your work.

I had tried to work on the calculation in FM13 and got the almost exactly the same as your on striker.

Wondering if you can find DM and WB positions' weight?

[poom15005]

- Weak Foot - The ability to use the weak foot well is crucial in almost all positions. This has been the case for a long time in FM. I feel that the importance of the weak foot is very overrated on FM. In particular, why should it matter for goalkeepers???

- Leadership is more important for GK and Center Backs than for other positions. Could this mean that leadership is not just about being a good captain, but may also have to do with coordinating the defense??

I think weak foot should not relate to CA since it is innate ability, players like Carzorla or Adriano can play equally well with both feet. However, it should be able to

improve through training.

I guess player in those 2 positions can play well with only needed attributes, so they might have room for free attributes like leadership.

[toluteriba11]

Defensive midfielder cost of attributes would be appreciated. Thanks in advance

[Zarmael]

I need a bit of help. If i raise players current ability does it raise his stats also? Coz he is better player then? Or no? Also if it does not then how does that current ability and potential ability works? What if I have player with low stats but current ability of 199. Will he be world class or just average?

[Cougar2010]

I need a bit of help. If i raise players current ability does it raise his stats also? Coz he is better player then? Or no? Also if it does not then how does that current ability and potential ability works? What if I have player with low stats but current ability of 199. Will he be world class or just average?

As a player plays and develops he earns CA points, nothing else.

These earned CA points are then converted to attributes based on your training, training is basically a way of weighting the points towards certain attributes.

One thing which can make a player look worse attribute wise is his weak foot. Weak foot also costs CA points so if a player has a good weak foot his attributes will be lower as the CA points have been spent there rather than on attributes.

[Methos]

I tried to replicate the OP method of creating those weights for FM15 and they seem different than FM14. If they are indeed different what conclusion can we make out of it? Also would anybody be interested to know those different FM15 values?

EDIT:

FM14

Forward Center:

3.2 - Acceleration, Pace

2.15 - Weak Foot

1.55 - Finishing, First Touch, Heading, Composure, Off the Ball, Jumping Reach, Strength

0.85 - Dribbling, Technique, Anticipation, Agility

0.32 - Crossing, Long Shots, Passing, Concentration, Creativity, Decisions, Positioning, Workrate, Balance, Stamina

0.125 - Corners, Free Kicks, Long Throws, Marking, Penalty Taking, Tackling, Bravery, Leadership, Teamwork

0 - Aggression, Determination, Flair, Natural Fitness

FM15

Forward Center:

2.47 - Acceleration

1.95 - Finishing

1.68 - Pace

1.42 - First Touch, Heading, Composure, Off the Ball, Agility, Stamina, Strength

1.11 - Dribbling, Anticipation, Decisions, Jumping Reach

0.89 - Technique

0.37 - Crossing, Long shots, Passing, Concentration, Positioning, Creativity, Work Rate, Balance

0.16 - Corners, Free Kicks, Long Throws, Marking, Penalty Taking, Tackling, Bravery, Leadership, Teamwork

Notes:

Acceleration, Pace have been nerfed but still better than other attributes.

Finishing buffed.

Agility Buffed.

Jumping Reach nerfed.

Decisions buffed.

Stamina buffed.

[Guest]

The only conclusion is that the weighting system was adjusted for FM15.

[Linie]

This is gold, can someone please summarize when some sort of consensus is reached for FM16?

[santy001]

Like most systems in the game, the attribute weighting system has evolved and changed in ways that will make it far harder for anyone to draw up a spreadsheet again with any degree of real usability with the actual players generated in the game.

Much for the better I say, it was kind of entertaining when you could take your star striker and retrain him to be competent in MC and DC and it would cause a surge in attributes as that hit one of the sweet spots (if my memory serves) but it was never an intended aspect of the game to be able to effectively artificially increase the CA cap. People will always stumble on things though, I seem to remember there was something people could do which enabled players to just smash the ball in from anywhere on the pitch and a mate of mine used to delete the old AI 4-2-4 formation as he couldn't defend against it which worked fine for him in single player, until he began playing network games with me and some others and we were getting random result changes because the AI was still changing to 4-2-4 and generating a result based on that, just the match engine couldn't show it so a 3-2 win would become a 3-4 loss etc.