FC Bayern Forum

Coman wrote:Wow your only answer is the Whoscored rate ? Hahahaha. Sorry I watch game. You watch Whoscored rate. Try better !

I've been posting a number of whoscored ratings lately and we've had some discussions before as to what how reliable their ratings are so I did a little further reading on the topic. Turns out, whoscored has a proprietary algorithm which takes into account around 200 variables per player/event but what is more important is it DOES take into account both the outcome (successful or not), location on the pitch (deeper events are given higher points) and it does penalize the player for unsuccessful events. It also takes into account how the event influenced the game.

The thing is whoscored takes raw data from Opta which seems to be the leading provider of live sports data. From Germany I couldn't find Bayern among its clients but it turns out Bayer (no wonder Reschencke was so good at scouting) is in as well as Bild and Kicker. Other clients (from a list of more than 100 household and not so known names) include Bloomberg, CNN, MU, Sky, Barza & Atlético and the like.

So before we get into another ignorant post such as, 'try better', I highly doubt we can. I found this interesting description from Opta's own data description page:

"For each event logged we add a series of descriptors: a pitch
coordinate, a player and team reference plus a time-stamp. These
variables enable us to provide hundreds of statistics on where
events occurred, which players were involved and when they took
place, meaning that the detail and range of information on every
ball contact is the largest available from any provider.
This level of detail means Opta data separates different types of
pass, such as through ball, headed flick-on and long balls. We can
tell which tackles were goal saving, from those which regained
possession in the final third, or even what percentage of shots
were left-footed volleys compared to diving headers.
We collect up to 2000 detailed events, live, during every match in
the competitions listed below..."

Hope this helps in evaluate how good or unreliable these statistics ratings can be.

And Whoscored says Xabi Alonso is our 5th best player this season.



That's all I have to say

Coman wrote:And Whoscored says Xabi Alonso is our 5th best player this season.

And I agree. That is exactly an average position. I don't think he has been quite calamitous (the guy can pass through the lines in case you haven't noticed) which certainly makes up for his abysmal lack of pace. He's also had some important goals, for a defensive midfielder.

Looks like a spot on ranking. If anything maybe Lewandowski and Robben can be switched, so can Alonso and Costa. Other than that perfect.

tflags wrote:

Coman wrote:And Whoscored says Xabi Alonso is our 5th best player this season.

And I agree. That is exactly an average position. I don't think he has been quite calamitous (the guy can pass through the lines in case you haven't noticed) which certainly makes up for his abysmal lack of pace. He's also had some important goals, for a defensive midfielder.

That's the problem with stats.

A player with 90% success pass and 80% of them for the goalkeeper will have a better rate than a player with 80% success pass but all forward.

Same for the mistake. A ball lost in his own area who gave a chance for the opponent = a ball lost after a cross in the opponent area.

Same for the bad defense. A good dual rate ok. But how many time Alonso was late and let his teammates alone ?

Etc.

ramsej84 wrote:2016 in numbers

Sent from my HUAWEI VNS-L31 using Tapatalk

Statistics I like!

Simply the greatest club in the world (and no, the number of CLs is not the sole criterion)

Coman wrote:

tflags wrote:

Coman wrote:And Whoscored says Xabi Alonso is our 5th best player this season.

And I agree. That is exactly an average position. I don't think he has been quite calamitous (the guy can pass through the lines in case you haven't noticed) which certainly makes up for his abysmal lack of pace. He's also had some important goals, for a defensive midfielder.

That's the problem with stats. A player with 90% success pass and 80% of them for the goalkeeper will have a better rate than a player with 80% success pass but all forward. Same for the mistake. A ball lost in his own area who gave a chance for the opponent = a ball lost after a cross in the opponent area. Same for the bad defense. A good dual rate ok. But how many time Alonso was late and let his teammates alone ?

And this is where you continue to be wrong. You seem not to listen. Not only are forward, side and backward passes differentiated, passes back to the keeper from forward positions have negligible impact on ratings. And for some reason I just stated that position on the pitch is one of the primordial influences on ratings; a missed passed in his own area which led to an opponent's goal would be severely penalized against a pass or cross (crosses are not passes by the way) on the last third.

But for some reason this discussion is futile if you continue to dismiss the importance of data in which the world of sports runs since the 1970s. So I bid you farewell as I don't see anything to gain from these discussions.

Ok ok. Can't wait to see one of our player being really bad and get one of the best rate in Whoscored.

Maybe you will understand that stats could be useless if his performance cost the qualification for Bayern.

I already see reactions : "Alonso 8,1 ? What the hell he was shit and our main problem. And Robben only 6,8 while he was our best player".

We'll see if you come with your stats logics.

Ben Mayhew updated his scatter graphics looking at the shot dominance and the attacking & defensive effectivenes of football clubs. Surprise, surprise, Bayern are in the “green areas” in the Bundesliga and also in the consolidated view of the top four European leagues.

Can't see tweet? Click here!
Can't see tweet? Click here!