by Walter Broeckx
Next in our series as we cover each and every Premier League team we have Fulham.
Now before any Fulham supporters comes on here and suggest that it is not the averages but the actual decisions that count (as some fans of other clubs have done) may we suggest you take a look at both the introductory articles which you can find here and here. You may also enjoy our other site: Referee Decisions.
And if you did you can come back and have a look at the statistical pointers we have found when we compare the overall results of Fulham with the results of each ref when he does Fulham games.
Of course you might have bad experiences with some refs in some games but who, after seeing the statistics, you find are 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 |
|
|
Fulham |
443 |
137 |
128 |
178 |
30,93% |
28,89% |
40,18% |
|
| Atkinson |
22 |
11 |
3 |
8 |
50,00% |
13,64% |
36,36% |
8,56% |
| Clattenburg |
22 |
6 |
10 |
6 |
27,27% |
45,45% |
27,27% |
8,56% |
| Dean |
27 |
8 |
11 |
8 |
29,63% |
40,74% |
29,63% |
10,51% |
| Dowd |
26 |
10 |
7 |
9 |
38,46% |
26,92% |
34,62% |
10,12% |
| Foy |
22 |
12 |
3 |
7 |
54,55% |
13,64% |
31,82% |
8,56% |
| Friend |
10 |
2 |
2 |
6 |
20,00% |
20,00% |
60,00% |
3,89% |
| Jones |
12 |
3 |
3 |
6 |
25,00% |
25,00% |
50,00% |
4,67% |
| Halsey |
24 |
7 |
4 |
13 |
29,17% |
16,67% |
54,17% |
9,34% |
| Marriner |
17 |
5 |
4 |
8 |
29,41% |
23,53% |
47,06% |
6,61% |
| Mason |
20 |
7 |
5 |
8 |
35,00% |
25,00% |
40,00% |
7,78% |
| Moss |
3 |
0 |
0 |
3 |
0,00% |
0,00% |
100,00% |
1,17% |
| Oliver |
6 |
1 |
2 |
3 |
16,67% |
33,33% |
50,00% |
2,33% |
| Probert |
11 |
4 |
2 |
5 |
36,36% |
18,18% |
45,45% |
4,28% |
| Swarbrick |
2 |
0 |
2 |
0 |
0,00% |
100,00% |
0,00% |
0,78% |
| Taylor |
10 |
4 |
4 |
2 |
40,00% |
40,00% |
20,00% |
3,89% |
| Webb |
23 |
6 |
8 |
9 |
26,09% |
34,78% |
39,13% |
8,95% |
In general in their seasons in the PL Fulham has a win percentage of 30,93%. Now let us see how this number is reached if we look at the current referees who are active in the PL.
The first remark is that all the refs seem to be available to do Fulham games and so we have 16 refs to do their games. I do like to point at the fact that Moss and Swarbrick who are relatively new to the PL will not be counted in this. They only did 3 and 2 games so far so not a real good measurement possible. The statistic from Moss doesn’t look too good to be honest with 3 defeats so we must keep an eye on this in the future of course. The result under Swarbrick is better as they managed to get 2 draws when he was in charge. But now let us move on to the refs we have a bit more numbers in Fulham games.
Who is a good ref for Fulham? (And by good I mean that he is a ref under whom they win much more than their average league number of wins).
The best ref for Fulham is Foy. He has a win percentage of 54% when he does Fulham games. He is closely followed by Atkinson under whom Fulham also win half of their games. The next best ref is Anthony Taylor. With 40% still a big difference with the overall league average. So we could conclude we have three “good” refs for Fulham.
Who can we classify in the average category when doing Fulham games? Clattenburg, Dean, Dowd, Jones, Halsey, Marriner, Mason, Probert and Webb all have a win percentage that is not far off what Fulham manages to win in the PL. This means that 9 out of the 14 refs we counted are close to the average result. So we have 12 refs out of 14 who are either good or average for Fulham.
Now let us find out who are those two Dean-refs (a local joke for bogey refs because of Arsenal’s results) for Fulham? In this case we see that Friend and Oliver have the lowest win percentage for Fulham and can be considered as referees Fulham rather not see appointed for their games.
The good news for Fulham is that both these referees don’t get many Fulham games. In fact if we look at the long run they only got an influence of nine points on average. So that is rather good news for Fulham.
And also if we look at the numbers of the refs who do the most Fulham games over the years we see that Dean and Dowd (that somehow sounds almost like a comical duo to me) are two refs who are in the average category. In fact both these refs are at most responsible for 12 points in a season for Fulham. And as they are in line with the win % it means that they don’t suffer any major losses because of their appointments.
I also would like to point at the fact that the refs are rather evenly spread for Fulham. By this I mean that the refs in general have no big influence and that there is no real +15% ref for them who does more games than average.
So the only worry for Fulham is how the new refs Moss and Swarbrick will develop over the years. That is still an unknown factor for the boys from Craven Cottage. Only time will tell us how this will develop.
One could say that this is an almost perfect referee report. And I would even feel like complimenting the refs and even the PGMOL for such a nice and even spread of referees. But then when I think of other teams I wonder why this cannot be the case for all the teams.
And then I remembered the last year review. And then I realised that we found lot of bad refereeing in the Fulham games. But the bad refereeing didn’t affect the total win percentage of Fulham one could say. A rather strange thing. Something that leaves me with something to think about. At first one could say that the bad referee decisions were almost more in favour and helped them a bit? If I have the time something to dig deeper in the future.
The referee analysis
- Case study 3: Everton
- Case study 2: Chelsea
- What’s wrong with the refs in the PL: Case study Aston Villa
- Is it right to have refs doing two PL games over a weekend?
- So what’s wrong with the Premier League – the opening part of the story
The books…
- Woolwich Arsenal: The club that changed football – Arsenal’s early years
- Making the Arsenal – how the modern Arsenal was born in 1910
- The Crowd at Woolwich Arsenal FC: crowd behaviour at the early matches
- Royal Arsenal: from the Common to the Manor. Coming next.
The sites from the same team…
- Referee Decisions – just what are the refs up to this season?
- The Arsenal History Blog from the AISA Arsenal History Society
- School sports minibuses: providing your child’s school with a bus
- The Direct Marketing blog: direct mail, email, websites, blogs
- Good Ad: reviews of the best and worst direct advertising there is
- The Massage Centre, serving Northants and Rutland
so what is this saying?
I would like to draw your attention Walter, to the amount of games Mark Halsey has done for Fulham.
Mr Atkinson does seem to like his west London teams, will be interested to see how he does with QPR.
Another team, another topic and an unbelievable decision today. Wigan started getting a lot of decisions in their favour at the end of last season…..it is happening again?
Mandy Dodd, keep your eyes open for the Wigan numbers 😉
but not going to give it away right now but something that is rather amazing. I can tell you that.
Will do Walter,
http://www.mirror.co.uk/sport/rugby-union/rugby-england-complain-ref-steve-1773531
Looks like the England rugby team have their own Mike Dean, they have won only three out of eleven with this guy and are complaining about the weekend game.
There seem to be a lot of them out there!
3 wins out of eleven? We would give an arm for that record under Dean 😉
Maybe Fulham prove that it DOES all even out in the end 🙂
On the rugby, it really looked like the ref was looking for things to penalise the English scrum on. They were already under pressure and the psychological impact really didn’t help.
That said, look at the difference in approach between the sports. In football, everyone who commented in the article would be about to get a ‘XXXX is fined XXXXX and warned about their future behaviour, rather than some actual debate.
Not to say England would have won under any circumstances.
On Wigan, again look at the football–Rugby differences. The FA cop out and point to the rules to show their helplessness. If it had happened in Rugby, Newcastle could have cited the player after the game, and he could have been punished, whether the ref saw it or not.
Excellent, interesting work Walter. I am enjoying this series much more than I expected. It is a good way of looking for the triggers that promote the tendency to bend! Two teams are being ‘controlled’. Are the refs ever given two of their ‘favourites’ together– or the opposite? And what are the gambling implications, bearing in mind the said ref’s proclivities? (Oh look here–a team he hates is playing one he loves–better get on sharpish!!!) PGMOL must know all this stuff. How do they select the officials bearing all this in mind?
There is a thesis or two here- and perhaps the DPP should be sponsoring them! Facts regarding a person spending twice the amount received on benefit is counted as evidence of benefit fraud. How about the evidence of a ref, who for more than one season, oversees a run of results against a top team out of all statistical likelihood? Hmmm!
Take it easy Walter and thanks again.
Doanythingformoney,
After we have finished the teams we will take on the refs. And then it could get really interesting because then you will have the stats for each ref with all the teams they do. I think that anyone who likes to gamble a bit might find some interesting things hidden in those numbers. I had been thinking of using that information to try and do some betting on certain teams in combination of refs. But I live in a bet-free zone a bit. 😉 But I sure think that there are some interesting things that can be found. Of course nothing is 100% sure but still I think it can be very useful and interesting information for those who like to take on a bet.
Hi Walter
I had a small suggestion which might make your analysis a bit more sharp. If you look at the numbers of Dean for Fulham, the win numbers are pretty normal but the draws are very high and the losses very low.
What this tells me is that Dean is still favouring Fulham so can you please have something like average points per game. Lets say if Fulham has 30% win, 30% draw and 40% loss then the points per game is 3*30%+1*30%+0*40%=1.2. Now if we look that for Dean it would be 3*30%+1*40%+0*27%=1.3. Now this certainly is a big gap for me as they get 8% more points per game under dean.
Abhishek Kumar,
thanks for the suggestion. I will continue as we are doing now but will keep this in mind and maybe try to rewrite this like you suggested in the future.
To sharpen this up the actual game outcome % should be compared or weighted with the pre-game outcome % (use bookmakers odds to provide these as they have massive amounts of data and statistical models to predict the likely outcome of any given game). Using the epxected points per game you can compare this to the points per game actually achieved.
Extreme example for illustration: if Fulham had played Man U 10 times under the same ref (unlikely but that’s why it’s extreme). Fulham’s expected points per game would be low e.g. 0.4 pts per game.
However their actual points per game may have been 0.8pts per game By your analysis the ref may have only overseen 1 win so would have a win% of 10% and 5 draws 40%.
Although this ref only sees a win in 10% of Fulham’s games you would say he was a dean-Ref. However he would actual average 0.8pts per game where they expected only 0.4pts per game so actually doubles their expected return. i.e. he is favourable to Fulham overall and they can expect a better result than normal.
Walter,
four case studies, four almost identical spreads – all teams seem to have one or two “Dean’s”, and one or two “Anti-Dean’s”, the rest sit bunched in the middle, where you would hope and expect them to be.
This to me is perfectly predictable, and if any statisticians by trade happen to be reading this, I think they would probably agree.
Performances of any band of individuals will always have this kind of spread given a large enough sample – a small number at either extreme, and most gathered in the middle.
This to me indicates the PGMOL have broadly got their assessments and personnel management about right, over a season there will always be stars, flops, and mostly average performers.
Professional football in this country is a multi billion pound industry, and any hint of corruption would certainly damage the key players in their pockets, where it counts – I’m just not buying the assertion that there is any wide spread conspiracy, or even a small one, merely human error due to crowd and managerial pressures, usually in the favour of big clubs with devious, Scottish managers.
Again, I will be shocked if any of your studies deviates from the pattern established from the first four cases.
However, the work is valuable, in so far as it establishes there is no statistical bias favouring any one team over the course of a season, and maybe the tin foil hat loonies will give it a rest 🙂
It’s grim up north
be patient and be shocked. Not all the teams have such a spread 😉
1 day, maybe 2 days patience. Man C, Man U, Newcastle and Norwich all look very interesting. Even QPR as far as they can look interesting 😉
Still digging up the rest….
Great stuff. Keep it up please.
And its grim up north,
you must keep an eye on the % of games a ref gets to do of a club. There you can find the key to how the PGMOL is working