Positional ability overrated in FM - Manchester United today

In my opinion the positional ability is way overrated in FM (and has always been).

Players with accomplished in a position will see their star assessment go down between 1 to 2 stars vs their best position. Players without at least competent for a position are completely useless in those positions even if they have attributes very well soothed for them (or at least the AI managers believe that, because they would never think of playing for instance Lampard as a striker instead of some non league striker). I have done some computation of my own, and believing in the star ratings of players, each positional point away from 20 costs a player 6 CA points!!

The team Alex Ferguson choose today for the Wolfsburg - Man Utd game completely makes a mockery of this. Man U is playing with Carrick and Fletcher in defense (I believe they both have 1 as a DC in FM), and they do not seem to be suffering very much (It is risky to be posting this at half time, but I might not feel like posting later ).

Carrick has a high versatility. Fletcher's made some key errors though, he completly missed to mark two guys in a Clear Cut Chance scenario. It should've been 2-1 to Wolfburg at half-time. I do agree that positioning is slightly overrated in FM, but not wholly.

I am not defending we should do away with positional ratings, but I believe the weight given to it in the game is way exaggerated.

Yeah I agree. And the fact that it takes AGES to learn a new position is also quite stupid. Especially when the 'new' position is already very similar to the natural position

Yeah I agree. And the fact that it takes AGES to learn a new position is also quite stupid. Especially when the 'new' position is already very similar to the natural position

It doesn't take ages if the player is versatile, determined and professional, or even one of these. I signed an AML with no WBL abilities at all and within a season he's got, according to Genie, 18 for WBL. During that time he's played maybe 20 games there.

Someone with higher versatility and with greater determination could become natural in a position in less than a year - not long at all.

The process of getting as new position has indeed speeded up a lot since FM09... In fm09 I got a 31 AMC with no skill at FC whatsoever and he got to accomplished in half a season (this was facilitated by me playing him there for the half a season)... Nevertheless, I think the performance of a player shouldn't be lowered just by playing out of position.

I think they should be really. If a player has spent his whole career playing MC and is suddenly asked to play DC or DR even if he has the ability to do so, it may take a while for him to adapt his game mentally.

I made a mistake on the last post: I meant: I think the performance shouldn't be SO lowered just by playing out of position.

Depends on his versatility though. I'd imagine Essien would be much better suited to playing at DC than Carrick for example - in part due to his stats but mostly due to his mental abilities and general versatility. Neither would play as well as someone like Wes Brown though - even if they're better players.

Look at this thread: http://community.sigames.com/showthread.php?t=174893

I did some research, and if my results are correct, you will understand why I think that the effect of positional attributes is overrated in FM.

Yeah. I've read that - good post.

BUT.. the moment you start playing someone out of position they get better at playing there. If they play there just once or twice, you'd expect them to play worse than in their natural position - but as they play, assuming they've got the versatility and the stats to do so, they'll get much better at playing there.

The problem is that irl positional ability doesn't really exist. All players have their personal set of abilities, along with their strengths and weaknesses. Based on that they are chosen for a role which suits them best.

I.e. any player could in theory play in any position, but of course they will not perform well in any position because their abilities do not match with what is required for every single position.

There is such a thing as being used to play in a certain position which adds to the pure ability factor, but that is far away from being set in stone as much as the positional ratings suggest.

Still, they should be kept as users are used to that and irl also players do regularly play in certain positions only.

In FM10 I too have the impression though that the weight given to positional ability is way too big.

I don't take much notice of positions in FM, it might say 1 and it might say that his performance will drop, but if a DM has 15 i heading, marking and tackling, he'll do a job regardless of positional ability. If an MC has 15 in finishing and composure, i'd happily play him up front because of an injury crisis and i'd ignore the position advice. My golden rule is; attributes > everything else.

I don't take much notice of positions in FM, it might say 1 and it might say that his performance will drop, but if a DM has 15 i heading, marking and tackling, he'll do a job regardless of positional ability. If an MC has 15 in finishing and composure, i'd happily play him up front because of an injury crisis and i'd ignore the position advice. My golden rule is; attributes > everything else.

That's exactly how it should be

Have you made experiences with doing that? Does actual performance not suffer too much?

If so, it might be more of a cosmetic issue concerning the stars ratings only.

I feel that players put in good technical and physical shifts, but sometimes can be a bit mentally lacking. IMO that's the way it should be. Bryan Robson said as much in midweek- it takes time to adjust and learn what you should be doing. He also said it was easier to move backwards than forwards.

I'm a versatile (terrible) footballer, but I always took time to adjust to a new position.

In FM, you can get a good performance out of a player in a strange position. If he has low versatility, the law of averages means he'll probably get worse, if he's versatile, he'll get better. I have two wingers on "swap position" at Liverpool, one natural AMR, one natural AML. The AML seems to perform better cutting in from the right. He's "unconvincing" fwiw.

That's exactly how it should be

Have you made experiences with doing that? Does actual performance not suffer too much?

If so, it might be more of a cosmetic issue concerning the stars ratings only.

I've played midfielders in forward positions and defensive midfielders in central defence or wing back, the results aren't as good as the real thing, but they aren't as bad as the positional rating would have you think.

I think the weighting is probably out a bit. No, a DMC shouldn't fall apart if played at RB, and an MC shouldn't completely fall apart at DC, but we should definitely see some poor play.

However, when a position is somewhat similar and the attributes are beneficial to that position, when I have had to play someone in a "red" position (EG, right winger out on the left, DC at RB) they've never got a terrible rating - some 6.5s but never much lower - but when I do this I give them slightly more conservative instructions than I would someone who plays that position naturally, which is what I am sure Fergie would have done with Messrs Fletcher and Carrick. I have noticed, though, that the lower down the leagues, the worse the impact is. For example, at Leeds I'd made all my subs at half time due to being 2-0 down against Carlisle, and due to an injury had to play Prutton at RB. We'd already come back to 2-2 but Prutton's performance went from around 7.0 to 5.6 even though as we held on to draw. But as Man City, using Tevez as an AMC for the second half (yellow) or SWP as an AML for a whole game (red) or de Jong at CB for 30 mins (red) the ratings did not dip nearly so much, in fact not much lower than their average rating.

So I guess it goes on ability, maybe CA, not sure. But definitely could do with re-weighting.

agree it is over-rated. i think if a player has stats suitable for that position he should be able to play allright regardless if he is red in that position.

I think the weighting is probably out a bit. No, a DMC shouldn't fall apart if played at RB, and an MC shouldn't completely fall apart at DC, but we should definitely see some poor play.

I think we do, though it is limited. However, poor play should be limited because of attributes, not because of positional sense i.e. an MC playing in a DC role with decent DC atts, but a 4 in marking, should be caught out because of that attibute, not just because he's an MC.

Perhaps the positional rating should be based on attibutes rather than knowledge of the real life player. Very few players irl have a 1 out of 20 in a majority of positions, they are far more versatile than that e.g.

Wayne Rooney: ST - 20, AMC - 18, AMRL - 16, MC - 14, DMC - 10, DRL - 8, DC - 4.

It's more realistic than 20 in one and 1 in the rest.

I think we do, though it is limited. However, poor play should be limited because of attributes, not because of positional sense i.e. an MC playing in a DC role with decent DC atts, but a 4 in marking, should be caught out because of that attibute, not just because he's an MC.

Perhaps the positional rating should be based on attibutes rather than knowledge of the real life player. Very few players irl have a 1 out of 20 in a majority of positions, they are far more versatile than that e.g.

Wayne Rooney: ST - 20, AMC - 18, AMRL - 16, MC - 14, DMC - 10, DRL - 8, DC - 4.

It's more realistic than 20 in one and 1 in the rest.

Agreed for the most part. That is probably true, though in the end it doesn't really matter as long as even with a positional rating of 1 we see some decent performances if the attributes are okay for that position.

I don't think though that a positional rating should just be the result of the attributes (actually then we wouldn't need it at all anymore, we'd just have to look at the attributes). Although I think that this is quite realistic as long as some factor of practise in a certain position still comes into play, but also this approach would probably lead to some flaws of players with unusual attributes for their position not being recognised as being able to play in their rl position.

But I do actually have an experience of playing someone out of position completely and still getting a good result lol

I accidently put my striker in the right back position but he still ended up scoring a hat trick.

I don't think though that a positional rating should just be the result of the attributes (actually then we wouldn't need it at all anymore, we'd just have to look at the attributes). Although I think that this is quite realistic as long as some factor of practise in a certain position still comes into play, but also this approach would probably lead to some flaws of players with unusual attributes for their position not being recognised as being able to play in their rl position.

I don't think either of us know for sure what way it works Is it attributes > positional ability or vice versa. I might do a test tonight, just FMRTE - Legend the player, if attributes are most important that player will be able to play any position and get excellent ratings.

I dont get how a player can play as a natural in say AMR, accomplished as AML and ST but then are unconvincing as a AMC. If a player can play on both wings and as a striker then surely they can play behind the front 2. Positions arent really that much different. You see it more and more with players now. They are a lot more versatile than say 10 years ago. Almost all wingers now can play on both sides of the pitch due to the players swapping sides in games. OK they might not be as good as on their natural side but its not like they are that out of position they start running the wrong way or tripping over their own feet.

Players need to be a lot more versatile IMO. Especially the attacking players but also the defensive players. I cant think off the top of my head but what is Owen Hargreaves rating for RB. He's played there quite a few times and never looked out of place.

Can someone please explain why the game should be changed on the basis of evidence from one match? Even at that, Carrick an d Fletcher werent great.

Can someone please explain why the game should be changed on the basis of evidence from one match? Even at that, Carrick an d Fletcher werent great.

If you read the thread and not just the OP, which was pretty light hearted, you'll realise that people are trying to determine how it works, not asking for changes because we think that it works ok atm.

Can I just add, this is the second time i've experienced Elmo thinking he's smart and trying to belittle someone, getting caught out and not even having the good grace to apologise or admit his mistake. Think that says it all about the lad.

I think we do, though it is limited. However, poor play should be limited because of attributes, not because of positional sense i.e. an MC playing in a DC role with decent DC atts, but a 4 in marking, should be caught out because of that attibute, not just because he's an MC.

Perhaps the positional rating should be based on attibutes rather than knowledge of the real life player. Very few players irl have a 1 out of 20 in a majority of positions, they are far more versatile than that e.g.

Wayne Rooney: ST - 20, AMC - 18, AMRL - 16, MC - 14, DMC - 10, DRL - 8, DC - 4.

It's more realistic than 20 in one and 1 in the rest.

Yes, I agree with that. A DMC with high marking, tackling, heading, anticipation should be able to slot into a DC role better than an MC with low marking but high passing.

But I think the trouble with giving players more varied positional ability is that, doesn't it eat up the points from CA?

I had a few occasions where I tried to make someone natural or accomplished with fmrte, but they kept falling back to the original state if I didn't increase the CA a little.

there's a problem in that being versatile in FM costs CA pts (stupidly). mind you i've never paid attention to this and ronaldo, with no extra training had 3 40 goal seasons in a row for me in FM08 (before he ended up scoring 40 goals that season irl i may add so it was pure ingenuity on my behalf! ) up front with 3 AMs behind him.

Ivanovic is currently Chelsea's right back (along with Bosingwa), and playing very well. I put him in as right back in FM, and he instantly gets a report card of "Someway off the standard of Boswinga," as if Bosingwa is twice as good as him at that position.

i'd agree that it is overrated. and i don't like the way they balance out a players versatility. i didn't know about this issue, and i just retrained aguero to winger since i already have dzeko and a superregen up front, and he goes down in most of his most crucial attributes, like finishing and composure. i don't think that is realistic, especially since he was already competent there.

Ivanovic is currently Chelsea's right back (along with Bosingwa), and playing very well. I put him in as right back in FM, and he instantly gets a report card of "Someway off the standard of Boswinga," as if Bosingwa is twice as good as him at that position.

I'd say Bosingwa probably is twice the player Ivanovic is tbh.

I don't think either of us know for sure what way it works Is it attributes > positional ability or vice versa. I might do a test tonight, just FMRTE - Legend the player, if attributes are most important that player will be able to play any position and get excellent ratings.

They way I've always assumed it works (mostly because it's the way I'd code it) is that the positions act as some kind of multiplier on certain attributes, dependant on how related each attribute is to your position. So if you're played out of position, your positioning and off the ball would be lowered a lot (because these are very dependant on position), maybe your anticipation would be slightly lowered (because anticipation is partly related to your position, i.e. reading the game), and finishing wouldn't be lowered at all, because your ability to finish has no relation to your position on the field.

They way I've always assumed it works (mostly because it's the way I'd code it) is that the positions act as some kind of multiplier on certain attributes, dependant on how related each attribute is to your position. So if you're played out of position, your positioning and off the ball would be lowered a lot (because these are very dependant on position), maybe your anticipation would be slightly lowered (because anticipation is partly related to your position, i.e. reading the game), and finishing wouldn't be lowered at all, because your ability to finish has no relation to your position on the field.

I wouldn't have thought so, positioning is a defensive attribute so it shouldn't come into play until you're in a defensive position or situation, same goes for off the ball, it's an attacking attribute and there's no reason for it to change depending on position.

A player won't suddenly anticipate play at a lesser ability because of the position their in, the attribute should take any failings in their game into consideration.

I wouldn't have thought so, positioning is a defensive attribute so it shouldn't come into play until you're in a defensive position or situation, same goes for off the ball, it's an attacking attribute and there's no reason for it to change depending on position.

A player won't suddenly anticipate play at a lesser ability because of the position their in, the attribute should take any failings in their game into consideration.

Basically what I'm saying is, players have knowledge of playing in certain positions, over and above their attributes. Rio Ferdinand knows how to play centre half. He also has all the attributes to play well at fullback (pace, tackling, positioning, passing etc), but he has no experience of playing there so he clearly shouldn't have 18 positioning when playing at a fullback (no idea if his positioning is 18, but you get the idea). So the obvious solution is to reduce players position-related attributes by a factor dependant on their ability in that position.

I'm including anticipation because if very much comes from reading the game - knowing how an opponent is likely to react in a given situation. You are less likely to be able to do this if you are playing in an unfamiliar position. Obviously position is not the only factor, which is why I said it should only be slightly reduced.

Have to disagree with anyone arguing for the removal of the positional ratings as assessed by the researchers.

It's better to think of this not as "positional ability" but as "positional experience".

So a player rated 20/20 has always played that role. Someone rated 10/20, it's not his normal role, but he's done it enough to know what he's doing. Someone rated 1/20 has never played that role.

Any role on the football pitch requires a different set of disciplines, a different type of game. So even if a player has the relevant skills, he will still have to get used to the intricacies of that role.

I do think FM has many flaws in this system. The one being referred to in the OP is one of my favourites. Given the choice, in FM, the AI will always use a player of 15+ over a player of more ability, even if that means using a 16 y.o. with CA 40 instead of that CA 150 30 year old that would be out of position. The game needs to value experience.

I don't think re-training is too slow, but I do think it can be polished. I'd like to see all players assigned a "training position", with an optional secondary position, so they are always working on that side of their game.

I also think positional ratings should be flexible. There should be no fixed position. In reality, if I use a guy who came through at left-back as a central midfielder every game for 5 years, he should be 20/20 at MC and his DL rating would be down to about 10. But in FM, that'd never happen.

Basically what I'm saying is, players have knowledge of playing in certain positions, over and above their attributes. Rio Ferdinand knows how to play centre half. He also has all the attributes to play well at fullback (pace, tackling, positioning, passing etc), but he has no experience of playing there so he clearly shouldn't have 18 positioning when playing at a fullback (no idea if his positioning is 18, but you get the idea). So the obvious solution is to reduce players position-related attributes by a factor dependant on their ability in that position.

I'm including anticipation because if very much comes from reading the game - knowing how an opponent is likely to react in a given situation. You are less likely to be able to do this if you are playing in an unfamiliar position. Obviously position is not the only factor, which is why I said it should only be slightly reduced.

Ah right, yes, I agree with that. In which case attributes would be amended (theoretically) due to positional ability, dropped down or boosted up perhaps. In which case it would have to be positional ability and not positional experience.

In which case it would have to be positional ability and not positional experience.

To be honest, they're pretty much the same thing in my mind. Certainly in FM terms anyway. A players attributes are his abilities, his positions just indicate where he best uses these abilities. And the only way he is going to gain the ability to use these attributes in certain positions is by gaining experience of playing in those positions. The two go hand in hand. If that makes sense.

Kinda, but you have to assume that players in certain positions will be better placed to deal with certain positions because of their attributes and ignoring their experience. It would be unfair to force a DM through the same experience filter, per se, as a ST, there should be a predetermined ability, which can be improved and I don't think that's done well enough atm. A majority of players have 20 in one position and 1 in the rest, that's unrealistic.

Kinda, but you have to assume that players in certain positions will be better placed to deal with certain positions because of their attributes and ignoring their experience. It would be unfair to force a DM through the same experience filter, per se, as a ST, there should be a predetermined ability, which can be improved and I don't think that's done well enough atm. A majority of players have 20 in one position and 1 in the rest, that's unrealistic.

But that's the beauty of the system I've suggested. If we moved Rio Ferdinand to fullback, he'd still do a good job because he's a good tackler, he can pass and he's quick. However he's likely to get caught out of position because his positioning has been reduced, which seems perefectly fine and reaslistic to me.

There's already an inbuilt penalty for players making 'bigger' changes of position becuase the key attributes for the new position are likely to be different. If you moved Ferdinand to left winger, he'd struggle more because his dribbling, crossing and off the ball are already low so he's not suited to this position. There's a case for making the amount of reduction in attributes dependant on the amount of position change, which is probably a decent suggestion.

If memory serves, FM already has some kind of sanity filter built in for players with 1 in a a position. I don't know how it works exactly, but was mentioned in terms of players with 20 at AMR and 1 at MR not playing like players with 1 positional ability if you played them at MR, the game boosts them up to something like 15. I'm not sure if it works for other positions though.

They way I've always assumed it works (mostly because it's the way I'd code it) is that the positions act as some kind of multiplier on certain attributes, dependant on how related each attribute is to your position. So if you're played out of position, your positioning and off the ball would be lowered a lot (because these are very dependant on position), maybe your anticipation would be slightly lowered (because anticipation is partly related to your position, i.e. reading the game), and finishing wouldn't be lowered at all, because your ability to finish has no relation to your position on the field.

bigdunk: I have always wanted to believe that it would work this way... but I fear that it actually works by scaling every attribute related to CA down by the factor that I refer to in the beginning of the thread. When CA moves some attributes change around almost uniformly, with the exception of pace, acceleration, balance and agility, which are less changed by a CA move. These would be the most obvious attributes for us to spot a change in the match engine if affected, so the way CA module is implemented, cleverly, makes this type of uniform lowering of skill very hard to spot, if true.

I believe this post by Dave C is spot on! Lots of things I have been thinking myself!

Have to disagree with anyone arguing for the removal of the positional ratings as assessed by the researchers.

=> I agree entirely with you, just think they have too big an influence.

It's better to think of this not as "positional ability" but as "positional experience".

So a player rated 20/20 has always played that role. Someone rated 10/20, it's not his normal role, but he's done it enough to know what he's doing. Someone rated 1/20 has never played that role.

=> Yeap, I think this should be the way to go, also by substituting the current term "natural" (which seems to imply proficiency from birth ) for something that describes the highest possible level of experience.

Any role on the football pitch requires a different set of disciplines, a different type of game. So even if a player has the relevant skills, he will still have to get used to the intricacies of that role.

=> Yeap, agree with this too.... although some skills/attributes make that transition smoother.

I do think FM has many flaws in this system. The one being referred to in the OP is one of my favourites. Given the choice, in FM, the AI will always use a player of 15+ over a player of more ability, even if that means using a 16 y.o. with CA 40 instead of that CA 150 30 year old that would be out of position. The game needs to value experience.

=> The game needs to value skill more than positional rating! Positional rating at the moment is too overpowered, or at least the AI believes so.

I don't think re-training is too slow, but I do think it can be polished. I'd like to see all players assigned a "training position", with an optional secondary position, so they are always working on that side of their game.

=> Yeah!

I also think positional ratings should be flexible. There should be no fixed position. In reality, if I use a guy who came through at left-back as a central midfielder every game for 5 years, he should be 20/20 at MC and his DL rating would be down to about 10. But in FM, that'd never happen.

=> Yeap!! That's what I would like too! I have been thinking of proposing that myself, but didn't get around writing it!

I wouldn't have thought so, positioning is a defensive attribute so it shouldn't come into play until you're in a defensive position or situation, same goes for off the ball, it's an attacking attribute and there's no reason for it to change depending on position.

A player won't suddenly anticipate play at a lesser ability because of the position their in, the attribute should take any failings in their game into consideration.

When you say positioning is a defensive attribute - do you mean in relation to the game? Of so fair enough but not IRL.

However, I do agree with what was previously said - if a defensive mid has 20 for positioning and he then was played as a centre half - I think his positioning attribute should decrease - postional awareness is different for a centre half and a midfielder - for instance - playing a good back line for an offside trap. Even a great def mid with 20 positioning wouldn't automatically they're positioning in a different positon would be 20.

Hope that made sense!

I wouldn't have thought so, positioning is a defensive attribute so it shouldn't come into play until you're in a defensive position or situation, same goes for off the ball, it's an attacking attribute and there's no reason for it to change depending on position.

A player won't suddenly anticipate play at a lesser ability because of the position their in, the attribute should take any failings in their game into consideration.

But that's the beauty of the system I've suggested. If we moved Rio Ferdinand to fullback, he'd still do a good job because he's a good tackler, he can pass and he's quick. However he's likely to get caught out of position because his positioning has been reduced, which seems perefectly fine and reaslistic to me.

There's already an inbuilt penalty for players making 'bigger' changes of position becuase the key attributes for the new position are likely to be different. If you moved Ferdinand to left winger, he'd struggle more because his dribbling, crossing and off the ball are already low so he's not suited to this position. There's a case for making the amount of reduction in attributes dependant on the amount of position change, which is probably a decent suggestion.

If memory serves, FM already has some kind of sanity filter built in for players with 1 in a a position. I don't know how it works exactly, but was mentioned in terms of players with 20 at AMR and 1 at MR not playing like players with 1 positional ability if you played them at MR, the game boosts them up to something like 15. I'm not sure if it works for other positions though.

Well said and good example with Rio as a full back or (hilariously) a winger. Much the same as what I was getting at but easier to understand!

How about (and this is extreme I know) do away with "positions" and simply have attributes. Leave it to us to decide! It would be easy to figure who should play where - although this woudl be very difficult with LLM.

Thought it worth mentioning. I have a bit of an injury crisis atm and thanks to an injury, mid match, to a central midfielder, I had to bring on a DL in a DM role. His attributes are ok, good tackling, work rate etc. Anyways, 1-0 down in the 92nd minute, cross from deep and he heads in the equaliser, comes away with a 7.4 and proves the point (even if it is only on this occasion) that attributes can help a player play out of position.

A player out of position doesn't know how to best perform in his role, and I think FM tends to get this fairly close to right. It could be scaled down a touch, but not by a great amount.

For instance, stick a hard tackling tough guy DM into CB position, and his stats may be perfectly suited. However, he doesn't quite understand when to close down, when to stay in line, etc. Equally a DM asked to play as an AM might not know when best to do certain things.

I get the impression that low positional rating tends to reduce the decisions, off the ball, positioning and anticipation of the player behind the scenes. Now you could say he'll be just as good at anticipating the play, but that isn't the case. He's not now looking at when the ball will come into an area that as a DM he can attack it, he's looking at it from the viewpoint of a CB.

Players out of position do well technically, but tend to make mistakes / act naively.