Stats and why you can't trust them.

[The Kop Kid]

O.K, I know that I have previously mentioned the seeming difference between some players stats and how they actually perform before but I ask the jury to consider new evidence which has come to my attention. Basically -and this is as far as I am concerned a MASSIVE flaw- the stats shown for a given player do not accurately reflect how a player will actually perform on the pitch. The reason for this is that there is a descrepency between the underlying (real) stats and the ones displayed.



This bit is OPTIONAL READING (not my writing):

"During the game, all the Playing Attributes are shown with a number between 1 and 20. Actually it is divided into two genres in the program,

One is the Low Range Data. It is equal to the number you can see in the game.

The rest one is the High Range Data. It is restricted in the range of valid byte value (-128 to 127), this attribute isn't shown directly in the game, neither can be calculated in a fixed scale. It takes the hidden High Range Data of RA(Real Ability) and the CA(Current Ability) of the player into one complex formulae to get a result FA(Fore-Ability) with the low limit of 1, when FA is found more than 20, it will all be shown as 20.


The short version of all of this is that how a player actually performs is governed for main attributes by a number between -127 and 127, which we do not see in the course of playing a game. So a striker with say a value of 127 for finishing will be a goal scoring god whereas a player with finishing of -127 will not be able to hit a barn.


O.K, SO HERE'S THE PROBLEM...

The stats which we normally see during gameplay do not correspond to the unseen stats, the ones which affect what the guy really does on the pitch.

Example:

Current Celta Vigo Game.

Player 1: Daniel Fredheim Holm.

Visible Stat In Game: Finishing 19.

Unseen (Real) Stat: 8 (out of a max 127)




Player 2: Vagner Love

Visible Stat In Game : Finishing 16

Unseen (Real) Stat: Finishing 36 (out of max 127)


What this means is that Vagner love is a significantly better finisher in the game even though Daniel Fredheim Holm has a much better visible stat. This is not an isolated case. I checked it with a number of player with dribbling, finishingstats etc. and what it amounts to is that the stats are massively unreliable.

Also, before anybody brings it up, this has nothing to do with the potential ability and current ability of players. Fredheim Holm is currently at 180 of a possible 180, whereas Vagner Love is 145. Those are totally irrelevant.

I realise that this is a rather large post and that yes I do have too much time on my hands, but I happen to think that this is a big issue for gamers....O.K, and I need to pretent to be busy at work. The biggest factor in the game is finding and buying the best players based on their visible stats and if those are unreliable then it means that the whole experience is gambling to an extent. What you see is not what you get.

So many times I've read something along the lines of "his stats aren't very good but he's an awesome player". Conversely I've bought players with finishing of 20 who couldn't bury a shot if their lives depended on it. Ever have two players with the same stat for something but one is great at it and another average at best? Well, now you have your answer.

I really feel that SI needs to address this issue and make the visible stats more reliable and reflect more accurately the true abilities of any given player.

 

[X-Ter]

Fantastic post. Now I know what all those numbers mean on the other real time editor. Thanks!

 

[Socrates]

Great post Kop Kid, now it all makes sense. Man this is really retarted, i dont see how Love can have a better finishing when his stat only says 16... I really dont see why SI did this to the game, i mean it would be better if it was just off the visible stats that we see.

No wonder Baptista and Griasles are long shot gods compared to others, i bet they have a higher not visible stat for that.

 

[rpvankasteren]

Actually, the current ability is weighed against those stats, and the in-game number comes from that.

The 8 -> 19 vs 36 -> 16 is done through the current ability. Holm just has a higher current ability. Vagner Love has high finishing, when compared to other stats.


180 of 180 doesn't matter, it's 180 vs 145.

Also: Mental and hidden stats have a great effect on how the player uses his talents.

 

[Tom]

yeah im with Ruud on this one, there has been many discussions about this on SICommunity forums and from what I CAN tell, it seems that ever since CM3 Sports Interactive have used the current/potential ability to effectivley 'calm' down or 'balance' the actual stats, hence the current ability will effect certain attributes within the game. I remember reading an actual post by Mark Duffy who reinforced an idea such as this.

So while your post is informative and makes excellent reading, it is innevitibly flawed from its conception due to the fact you have not taken into account other "confounding" variables such as the ones Ruud has mentioned.

TROD.

 

[rpvankasteren]

Yeah, and Vagner Love is as inconsistent as, well, anyone can be.

 

[The Kop Kid]

Actually, the current ability is weighed against those stats, and the in-game number comes from that.

The 8 -> 19 vs 36 -> 16 is done through the current ability. Holm just has a higher current ability. Vagner Love has high finishing, when compared to other stats.


180 of 180 doesn't matter, it's 180 vs 145.

Also: Mental and hidden stats have a great effect on how the player uses his talents.

Erm..I do believe that I said that the 180 ability thing was irrelevant for precisely the reason that there was no point in going down that road. Your point about mental and hidden stats makes no sense whatsoever in this context as I am talking about individual stats.
Also, Vagner Love is just a case in point so I don't see how his consistency is relevant. I made it pretty clear in my original post that the problem was a widespread one.

yeah im with Ruud on this one, there has been many discussions about this on SICommunity forums and from what I CAN tell, it seems that ever since CM3 Sports Interactive have used the current/potential ability to effectivley 'calm' down or 'balance' the actual stats, hence the current ability will effect certain attributes within the game. I remember reading an actual post by Mark Duffy who reinforced an idea such as this.

Nope. Sorry Tom but that arguement is seriously flawed. The game makes it's calculations on those -128/127 numbers which are the final result of the RA and CA.

Fredheim Holm has current ability of 180, Visible finishing stat of 19

Vagner Lover has current ability of 145, Visible finishing stat of 16

Therefore, as far as the gamer is concerned Holm is going to be the better finisher on ALL appearances.

BUT

The database calculates that Vagner Love dispite all VISIBLE indications actually has a better finishing stat. You guys can go on about current ability and weighing values as much as you want but in the end it all comes down to the game displaying that one player has a better finishing stat whilst in fact the other does.

TRY THIS...

Find a striker with finishing 20 and a reasonably low hidden count, say a +16. Play him for 10-15 games in a team where he has scoring opportunities. Then use the in-game editor and change the hidden count to 127. He still has a visible stat of finishing 20, he still has the same potential ability. His "confounding variables" are still exactly the same. You will find a huge difference in the amount of chances he finishes. Why? Because the -128/127 count is 'God' and not the corresponding visible stat which is often at variance to it.

 

[Tom]

so if your theory is right then Darren, why would SI do this? and how long would it take the scouts to go through all this hassle?

TROD.

 

[rpvankasteren]

I meant to say it doesn't matter Holm has fulfilled his potential. His current ability is higher than Love's, and that's why the lower RA becomes a higher FA.

 

[Socrates]

Yeah, and Vagner Love is as inconsistent as, well, anyone can be.

Well actually in my CSKA game he did pretty even though he didnt have that great of a stats.

I also agree with Kop Kid, it seems like the hidden varible is more powerful than anything else. This is the reason why unkown players with crap stats can still be good.

 

[The Kop Kid]

I don't think that this is some wacky conspiracy on the part of SI, I simply think that their maths (for want of a better word) is flawed. You seem by your tone to think that I am having a go at SI for the sake of it, which is not the case. I wouldn't bother if it was EA Sports for example as they never give a sh*te what gamers think anyway. I think that this is something that needs to be looked at though for future versions of the game and a more accurate method found for displaying the stats.

I don't quite understand your point about the scouts (?)

 

[rpvankasteren]

Ok, so you have RA x CA = FA, right?

Now looking at the RA for
Holm: 135 (out of 255)
Love: 164


Multiply that with CA (I know SI probably use a linked value, but still)

Holm: 135 x 180 = 24300
Love: 164 x 145 = 23780

So Holm would indeed have better finishing than Love.


I have no clue as to how they balance it out, but there must be a factor introduced somewhere else

 

[rpvankasteren]

FMM 1.10 actually calculates (on estimate) the FA from RA and CA. It shows you the FA in a tooltip window, when hoovering over the RA value.

 

[Daz]

what is RA?

 

[The Kop Kid]

RA is Real Ability. A number ranging between -128 and 127. Every player has one. 127 for an ability (i.e dribbling, finishing, longshots) awesome, -128 means Bognor Regis substitute.

 

[rpvankasteren]

I think the Bognor Regis substitute can still have an Real Ability of 0, but his Current Ability would be in the region of 2 and 3.

The real ability is more an indicator of where the player's talents lie (does he have good dribbling, for his level of play?), and the current ability shows how big the talent actually is (his level of play sucks big time).

 

[The Kop Kid]

Now you are contradicting yourself. If as you say on the previous page an ability is determined by the CA x RA,which gives the FA which then is broken down into a visible stat of between 1 and 20 then the RA is one of two equally weighted factors in determining the players final level of performance on the field. It's a simple axb=c.

Again, the problem is that once a stat hits 20 you can no longer see the variable in performance caused by the RA. Here's an example. We'll use Berbatov from my current game.

We open up FMM.

His in-game stat for finishing is 20

He has current ability of 180, and real ability of 162 for finishing.

So, 180x162= 29160. This is what determines how well he is going to finish.

NOW....

We change the players RA from 162 to 255.

We now have 180x255, which gives 45900.

The players in-game finishing stat is still 20.

THE PROBLEM BEING THAT....

although the in-game stat for finishing is 20 both before and after we made our changes, the actual number which is used to calculate what the player will actually do on the pitch is massively different, having changed from 29160 to 45900. A 47 percent increase in finishing abilty.


FROM MY CELTA GAME. FOR THE DOUBTERS.

Long Shots.

Freddy Guarin. In-game stat:20 CA 174 x RA 181= 31494 for FA

Andy Jones. In-game stat:20 CA 74 x RA 231= 17094 for FA

Juan Roman Riquelme. In-game stat:17 CA 185 x RA 139= 25715 for FA

Three players. Freddy Guarin and Andy Jones both have an in-game stat of 20 for long shots and yet Freddy is actually almost twice as good at them as Andy Jones. Riquelme has a long shot stat of 17, 3 less than Andy Jones, but is also significantly better than him at them. This basically amounts to the in-game stats being bollocks as a representation of a players true ability at a given attribute.

SI need to build a stat system which more accurately reflects the differences between different players abilities on the pitch.

As I have said before this is certainly not an attack on SI as I think that overall they are a good company with an excellent game, but this is a major flaw and it does really need fixing if the game is to continue to improve.

 

[Socrates]

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.

 

[rpvankasteren]

Well, I feel SI aren't using the full range of -127 through 127 for RA. Have any of you ever seen anyone with an RA lower than -50?


Anyway, it's not supposed to be a simple axb=c, as it's said to be a "complex formulae". That's why I said I didn't know exactly how it's balanced. Although the Riquelme example does throw me off a bit.

 

[Parra Power]

this is why you've got to look at how a player performs and the ratings/goals/assists/cleansheets/tackles/dribbles/shot%/whatever before buying the guy