Derby County v Bristol Rovers: Player Ratings

[Ellafella]

Summary

Thanks to you for your Player Ratings for the Rovers game last Saturday; 34 of you took the time and the trouble to provide your figures & I salute you for it. 34 is the highest number of raters thus far. It was a very memorable game for the first-half hatrick and a blistering display of full-bloodied attacking play by David McGoldrick.  The highest compliment I can think of is to say it was Hector-esque; those of us who had the privilege of seeing Kevin Hector play on the mud will also see the similarities - great mobility, superb speed of thought and instinctive finishing. It really was that good. It is no surprise then to see that McGoldrick topped the average rating by a country mile with a score of 9.37. 8 raters gave him a perfect score of 10, and his lowest rating was 9.00. 

The average aggregate rating was 87.4 - the highest of the season thus far. Mendez-Laing was 2nd highest with 8.20 - our most consistently highest rated player of the season so far. Collins chipped in with an impressive score of 8.14 as for once the front line outscored midfield and defense. It was all about that intensity of play from the start which blew an on-form Rovers side away.   Average player rating was 7.08 (mode 7.0). 6 players scored above this average and 8 players below. The red dotted line on the graph shows the cut-off. 

Thanks again for your time and effort and your comments in the ratings which I really enjoy reading. It always makes me realize just how very knowledgeable Derby County fans are. We are gathering a big database of scores through this process and I will start to provide some additional kinds of analysis at the season's mid-point.

Edited November 2, 2022 by Ellafella

[Ellafella]

Just want to make sure people see this so they continue to give player ratings - gets quickly buried otherwise. 

[CBRammette]

Thanks for doing this. The most interesting thing is how scores vary in the ratings when we all watch the same match. No wonder we have diverse opinions on here generally!
 I am not trying to add extra work but just wonder if those in the ground generally score differently as you can see a player's all round play, positioning, tracking back etc (but may be influenced by the crowd also) compared to those online who see from the cameras perspective with no influence from the crowd. 
Anyway thank you! ?

[sage]

14 minutes ago, CBRammette said:

Thanks for doing this. The most interesting thing is how scores vary in the ratings when we all watch the same match. No wonder we have diverse opinions on here generally!
 I am not trying to add extra work but just wonder if those in the ground generally score differently as you can see a player's all round play, positioning, tracking back etc (but may be influenced by the crowd also) compared to those online who see from the cameras perspective with no influence from the crowd. 
Anyway thank you! ?

The range is interesting. 

Sometimes I used to do a range in marks for games. Biggest gap in scores was 4 if you ignore silly sods giving very low scores 

[Ellafella]

12 hours ago, sage said:

The range is interesting. 

Sometimes I used to do a range in marks for games. Biggest gap in scores was 4 if you ignore silly sods giving very low scores 

Indeed. I’ve posted previously about all the different factors affecting ratings; different posters clearly have different thresholds for giving scores - which adds “noise”. 
 

The 4-2 against Gas was interesting because the ranges were lower and there was an homogeneity of scores for McG. It seems that when things go well - ie we win and there’s clearly a team good performance- scores are more consistent. 
 

The picture is different when we lose - that’s when extreme scores creep in and ranges become greater. 
 

As you’ll know @sage it’s a 2 hour job to prepare the output so not easy when you’ve already spent a whole working day in front of the computer; but I will eventually put out some additional analyses. 

[Carnero]

@Ellafella some people do seem to have a strange bias against some players, their chosen boo boy of the moment as it were (e.g. Collins early season, Bird now), and mark them far more harshly than the majority of other posters do. Wouldn't it be worth dropping out say the highest 3 and lowest 3 ratings for each player to smooth out these biases, or would this have added too much more hassle?

[Ellafella]

50 minutes ago, Carnero said:

@Ellafella some people do seem to have a strange bias against some players, their chosen boo boy of the moment as it were (e.g. Collins early season, Bird now), and mark them far more harshly than the majority of other posters do. Wouldn't it be worth dropping out say the highest 3 and lowest 3 ratings for each player to smooth out these biases, or would this have added too much more hassle?

I can do that @Carnero - but it may prove controversial. It's not a problem to exclude extreme scores and indeed it is something that research statisticians do from time to time. @Sage has mentioned looking at ranges and this is something that I will get my teeth into especially if it shows that some players {eg Tom Lawrence} have a "marmite" following.