So what is wrong in the PL? Case study 2: Chelsea

by Walter Broeckx

Next in our series as we will cover each and every team we have Chelsea.

Now before any Chelsea supporters comes on here and start wondering what this is about I would like it if you would read the other articles on this first to understand what we are trying to do.

You can find them here and here

And if you did you can come back and have a look at the statistical things we have found when we compare the overall results of Chelsea with the results of each ref when he does Chelsea games.

Of course you might have bad experiences with some refs in some games but who, after seeing the statistics, appear to be not really that bad in general.

You can share your experiences of course. But this article is not really about those games in particular. It is more about the total picture of the referee and this team.

Under the table I will try to give a short explanation on what you see.

 

Total

won

draw

lost

won

draw

lost

% games

Chelsea

797

414

206

177

51,94%

25,85%

22,21%

Atkinson

24

18

4

2

75,00%

16,67%

8,33%

8,14%

Clattenburg

19

12

4

3

63,16%

21,05%

15,79%

6,44%

Dean

48

33

7

8

68,75%

14,58%

16,67%

16,27%

Dowd

33

21

5

7

63,64%

15,15%

21,21%

11,19%

Foy

27

20

5

2

74,07%

18,52%

7,41%

9,15%

Friend

8

2

3

3

25,00%

37,50%

37,50%

2,71%

Jones

8

7

0

1

87,50%

0,00%

12,50%

2,71%

Halsey

39

22

8

9

56,41%

20,51%

23,08%

13,22%

Marriner

19

11

4

4

57,89%

21,05%

21,05%

6,44%

Mason

15

10

4

1

66,67%

26,67%

6,67%

5,08%

Moss

2

1

1

0

50,00%

50,00%

0,00%

0,68%

Oliver

4

2

1

1

50,00%

25,00%

25,00%

1,36%

Probert

9

7

1

1

77,78%

11,11%

11,11%

3,05%

Swarbrick

2

2

0

0

100,00%

0,00%

0,00%

0,68%

Taylor

5

3

1

1

60,00%

20,00%

20,00%

1,69%

Webb

33

14

11

8

42,42%

33,33%

24,24%

11,19%

The first line is the overall win/draw/lose % of Chelsea in the PL era. So you can see that Chelsea has won around 52% of their games in the PL.

The first thing we now do is to see if the refs who are in the PL for the moment are in line with this win percentage.

Chelsea is one of the teams who can have had all the refs in their games. So apparently no ref ever declared to be a fan of Chelsea or to have any interest in this team.

First of all I want to start and say that the numbers of Swarbrick, Moss and Oliver will not be really counted, even thought the score of Moss and Oliver is close to the Chelsea league average. But when you only have done two or four games I don’t think this is enough to give us real insight in how Chelsea do under these refs. So that leaves us with 13 refs to have a closer look at.

Let us start with the refs under whom Chelsea has more wins than their average PL numbers. And by more wins I mean a lot more wins. And then we see a few remarkable numbers. The ref under whom Chelsea wins the most games is Jones. With a win % of 87,5 this is almost the double of their usual win %. Only one defeat in 8 games so far. Eight games is not really that much but still an important trend we should be keeping an eye on in the future.  We also have Probert with an very high win percentage but this is also based on only nine games but still something to keep in mind.

But if we move on to the refs who do more of the games we see that their best ref is Martin Atkinson. For the regular Untold readers this is no real surprise. We have highlighted this before. A win percentage that is almost 50% better than normal after that many games looks very strange. And very out of line.

The same goes for Foy.  Beyond this some refs under whom Chelsea grab more points than average are: Dean, Mason, and Dowd. And this might come as a surprise after this seasons game but also under Clattenburg Chelsea have a high win perecentage. Certainly higher than the league average win percentage.

So all in all of the 13 refs we take in to account we find 8 refs under whom Chelsea win more games than average.

Talking about average. We have a few refs who come close to the overall win percentage of Chelsea. Those refs are Halsey and Marriner. And we could maybe add Taylor to that list. That is three refs under whom Chelsea come close to the average win %.

But now move on to the refs who are bad for Chelsea. And there we find two names. That we find the name of Howard Webb in this category is not a real surprise for most Chelsea supporters I guess.  But the team under whom Chelsea has the worst performances is Friend. As he doesn’t have a lot of Chelsea games until now we don’t know how this will develop. But certainly Chelsea fans should keep an eye on this. So will Untold Arsenal of course. But that makes two bad refs for Chelsea. From a Chelsea point of view not that bad I think.

Now the good thing about this for Chelsea fans is that Friend only does a small number of Chelsea games so his total influence on the league table is rather small. But the influence of Webb is bigger. He has on average a potential influence of 12 points for Chelsea in a season.

But the other side of the coin is the fact that the ref whom they get most (Dean) is a ref under whom they win more games than usual.  Dean is responsible for 18 points in a season on average. Way too high to be healthy but I don’t think Chelsea fans will mind Dean that much as he brings them more points than average.

So the overall report about Chelsea is rather a good one. Most refs have a better than expected win percentage in general. And only two refs under whom they perform not good at all. I think I have seen worse reports in my life.

 Recent posts

 

The books…

The sites from the same team…

Fill in

10 Replies to “So what is wrong in the PL? Case study 2: Chelsea”

  1. I think all is not bad for Chelsea, because if Mr Chelsea invites you on his 2billion pound yacht trust me you will be a different ref the next time you officiate his team

  2. Is there a way for factoring the Russian Cash into these calculations? Given the financial doping which began midway through the EPL era, with the related spike in league performance, the more recent referees will obviously oversee higher-than-average performance by Chelsea, despite their bias; indeed, given the scale of the Russian Cash, it’s entirely plausible that a referee with a strong anti-Chelsea bias actually had better-than-average numbers. Perhaps you could divide these figures into pre- and post-financial doping groups? Just a thought…

  3. Walter this is a very revealing and informative analysis.

    At some point the weight of evidence confirming ref bias must become overwhelming – when that happens those who protect the present system will start thinking in terms of abandoning Riley et al and will concentrate on protecting themselves. That will be very interesting.

    Does the PGMOL have to publish annual accounts?

  4. As the most refs apart from the older refs like Dowd, Foy are doing Chelsea games from after the Abramovich arrival it should be in line a bit I think.

    And Babakdrdaemi, no linesmen are not covered and unless I can turn this in to a full time job will not be covered…

  5. Excellent work once again Walter but here are a few questions. By the way I am NOT a statistician but did study it in my psychology training:

    1)Chelsea have been in the top 4 for the last 11 years so perhaps they should be reviewed for those refs who did top 4 games against Arsenal,United,Liverpool, or City during that time?
    2)How have they done against London clubs home and away?
    3)What are their overall results both home and away with refs?
    4)I would love to know their results since Riley’s promotion to the PGMOL.

  6. These analyses are unfortunately lacking crucial points. For each game it would be prudent to check Chelsea’s expected result and compare to the actual result, rather than the win rate over 5 years.
    For example, who were the teams in Jones’ games? If they were all relegation candidates/mid table fodder you’d expect that kind of win ratio.
    Finally you have to show that the difference from the expected results are significantly different from the actual ones, I think you’d used something like a chi-squared test to do that.

  7. I do feel this is a difficult one as the average of Chelsea of 51% is not fair to judge refs on. I believe this may have been fairer if the average was taken once Roman took over.

    Not surprised Webb is coming up as a common negative ref for the top clubs. No surprise if you compared games V you would see Webb refs most of UTD’s big games. Meaning he effectively has twice the impact. giving UTD points while taking them from other teams.

    Cant wait for UTD. Plus I hope you send these findings on an easy to read PDF across to the PGMOL, FA, the Premier League and every major newspaper and sports media out there.

    This is a fantastic site

    Keep up the good work

  8. Hi Walter,

    I’ve been away for a while, but fair enough if the refs have started since then. But are the older refs still enough to skew the figures?

    I can’t wait til these studies and the other work you’ve done makes the mainstream. We need accurate, informed analysis!

  9. We must know that average is not a realistic yardstick for measurement. Eg, 100 and 1 ve an av of 55.5 just look at wat av can do. Now chelsea av of 51% may be a balance of poor results b4 ab and excellent result after ab. The refs reviewed were from current matches so shd be naturally better than the av bc of after ab factor. However, i feel the refs ve a secret responsibility of assisting big club world wide. The problem with arsenal is that the ref ve de-listed arsenal from this category bc we are not buying stars. If arsenal assemble a team of stars, the officiating may change.

Leave a Reply

Your email address will not be published. Required fields are marked *