Stats and why you can't trust them.

[The Kop Kid]

I wonder if SI will change it from 1-20 to 1-100 or 50 becuase it seems like 20 is too small of a number so show the players true skills.

Yep, that is what I would like to see implemented. A greater division of visible in-game stats. It would also need to take into account a higher max range for the RA x CA as currently at a certain level the in-game stats end whilst the actual ability still varies above it. The Berbatov example highlights this.

RUUD: In answer to your question, yes there are many players with stats below -50.

 

[Tom]

yeah sorry i was just about to say this but arent you just saying that the attributes simply need to be more elabourate, e.g. like Socrates says.

So you're up for for 1-100 instead of 1-20?

CM5 does that

TROD.

 

[rpvankasteren]

If you have a wider range available, why only use 1-20? I agree that's not good. Tell them, Tom. Either use 1-100, or make the engine restricted to using 1-20.

 

[Tom]

personally i think it creates more mysticism having higher and lower 20's as you might put it. Shows that looking at av rating and performance is just as important as the stats themselves. So no, i shall not tell them

TROD.

 

[rpvankasteren]

That's only good if you don't show stats at all.

 

[Áļéאָ]

if they change stats from 20 to 100 ill stop playing the game

 

[The Kop Kid]

Yes, I am saying that a greater amout would be more accurate but much more importantly what I'm saying is that the in-game stats also need to reflect the very maximum that a player can possibly achieve. Whether the max in-game stat is 50 ,100 or whatever is a somewhat lesser issue.

At the moment when a player gets to a certain level from the RA/CA for an attribute the in-game stat reaches 20 and stops there. However, a players ability can go way beyond the minimum number required for that 20 to be displayed.


Here is What I am saying. I'll use some of the previous players to illustrate the point.

Andy Jones. CA/RA of 17094 for long shots. In-game stat:20

Freddy Guarin CA/RA of 31494. Current In-game stat:20


Freddy Guarin: CA/RA of 31494. 84 percent better than Andy Jones 17094. So, based upon Andy Jones in-game stat of 20, a stat which showed Freddy's true ability (relative to Andy Jones) would be an in-game stat of long shots 37 (184 percent of Andy's 20). This cannot occur though as the in-game stats stop at 20.

 

[Tom]

ah, now feck me thats a good point, unless as Ruud says, that isnt totally correct due to complex forumlae

so for the thickheads like me, basically that means that in your opinion FA could possibly be 28 for someone and 20 for someone else even though 20 is shown for both. i.e. one is far better than the other. right?

TROD.

 

[Socrates]

ah, now feck me thats a good point, unless as Ruud says, that isnt totally correct due to complex forumlae

so for the thickheads like me, basically that means that in your opinion FA could possibly be 28 for someone and 20 for someone else even though 20 is shown for both. i.e. one is far better than the other. right?

TROD.

Yes, and imo this all makes sense becasue if you have ever had Grisales/Gurain they both usually have 20 for longs shots after a while but they are the best at it. In my Ajax game Welsey and Hatem Ben Arfa both have 20 for longs shots as well but they dont get nearly as many goals as the other 2. I play Girsales as mc and 90% of his goals are from longs shots.

 

[rpvankasteren]

Probably their mentalities tell them to try other things more.

 

[Socrates]

Probably their mentalities tell them to try other things more.

The thing is that if there long shots are all 20 than all 4 should be equal in that aspect but Wesley and Hatem dont get nearly as many goals as the other two when infact they should according to thier in game stats. I put all of them on longs shots/often but the results are different for each player.

 

[rpvankasteren]

Unless they take less shots at goal. Guarín just keeps shooting at goal.

 

[NIDOKING]

First of all: Football is not logical, so why games that emulates football would be?

I think the FA and RA are not fixed, there are stats called Consistency and Important Matches, also one call Composure, there are also mental stats like Temperament, Pressure and Adaptability, not to forget Liked and Disliked teams/players, and all of this makes the all so almighty FA and RA changes on a match, IMHO.

If that's not this way, try to explain why, for example, Tevez could play 100 times better for Boca than for Atlético Madrid, or why Kuijt plays so well at Holland but do on EPL?

That's my opinion, I don't know much of it, but I think my points make sense.

Ah, and let the stats at maximum 20 or the game will desbalance, specially LL Teames that got promoted to a new level.

 

[The Kop Kid]

ah, now feck me thats a good point, unless as Ruud says, that isnt totally correct due to complex forumlae

so for the thickheads like me, basically that means that in your opinion FA could possibly be 28 for someone and 20 for someone else even though 20 is shown for both. i.e. one is far better than the other. right?

TROD.

Yep Tom, that's basically what I'm saying, although it's actually even worse than that IMO with a player at times actually having a worse in-game stat for a particular attribute than another whilst actually being better at the given skill. For example, Riquelme has a vastly greater CA x RA for long shots than Andy Jones but an in-game long shot stat 3 points less at 17 than Jones 20. He also has better supporting stats (important matches, consistancy, positioning etc etc.) but I'll get to that in a minute.

Regarding the "complex formula" idea I think that there is still a descrepency regardless. Freddy Guaring is as I mentioned 84 percent better at long shots than Andy Jones based on the CA x RA process alone, so he should have an in-game long shot stat of 37 (relative to Andy's 20) based upon that. 184% of 20=37.
Now, even if other factors such as consistency, important matches etc. are then taken into consideration by the 'complex formulae' that 84 percent (or 17 point in-game stat) gap is not going to be even close to being made up.
In fact, in the example of these two players Freddy Guarin also has better stats than Jones for just about everything else (mental and physical) so hypothetically unless the complex formula ignores ALL player data in it's calculations the gap between the two given players in-game long shot stat should be even greater and not less than the original 84 percent, with Freddy perhaps having an in-game stat of 40 or so for long shots relative to Andy's 20.

I know I'm like a dog with a bone on this issue but I think that the game could really be improved if this problem was ironed out. FM is a good game and the new features for 2006 look great, but there is deffo a problem with the relationship between the calculation and presentation of the game's player data IMO.

 

[Tom]

darren, obviously i know a hell of a lot about FM but im not even gonna pretend i totally understand your argument. What i suggest is that you go and post all this in the actual SICommunity forums.

TROD.

 

[The Kop Kid]

Wouldn't it be a hell of a lot simpler if you just asked your pals over at SI what formula is used to come up with the final player stats. It wouldn't be like they would be giving away the game code. That way we could prove beyond any doubt if there is indeed a flaw in the process as I believe.

 

[Tom]

not really Darren, i have no idea what im on about. If you wanna put together something then ill ask.

TROD.

 

[The Kop Kid]

O.K . Try this. It's pretty complicated but I'll try and keep it simple.

You: "Can you tell me what process (formula) is used to calculate the player stats for finishing, long shots etc.that we can see in the game?"

Expected Response from SI: "F*ck off, we're not relling you." Possibly followed by "nah, nah, na, na nah."

You never know though, they might release the info. I can't see how it could damage them.

 

[Tom]

darren, include your thoughts on what you think it might be, otherwise i sound like a twat

like e.g. i think it is .......

TROD.

 

[The Kop Kid]

How does asking for the formula that is used to calculate the in-game player stats make you sound like a twat? If you ask your lecturer a question at Uni do you feel that you have to falways give your theory on the answer first.


If I knew the exact process I wouldn't need to be asking. I simply can't find any method of calculation that makes the in-game stats accurate given the underlying figures. The CA/RA combination seems to be the main factor for determining a players ability at a given attribute, though I expect that the other stats such as consistancy are also used as a filter before the final number is arrived at.

Ask the question on my behalf if you want, that way I'll look like the twat. Unless SI give an answer though, which I doubt, I think I'll leave this topic be though as we're starting to go in circles and I think that everything has been covered. It's still flawed though.