Last season a group of qualified referees reviewed more than 40% of the EPL games. The reviews themselves were based on full match video footage with the advantage of video technology features such as slow motion and pause.
By reviewing those 155 games we have made a database of more than 7000 decisions that have been judged by our panel of dedicated and qualified referees.
The numbers you will see are based on those decisions and those reviewed games.
Below is the review we published in the summer on Swansea and the refs. We are publishing it again because we hope to attract more referees who are willing to monitor what the refs in the Premier League get up to. If you would like to help us please have a look at our earlier post on our desire to recruit Swansea fans who are qualified refs to help us in this venture.
We managed to review 8 games of Swansea and this represents 21,05% of the total games played last season. We could agree if you said this is not enough, because we would have loved to do more games. But as we didn’t have more games that have been shown live on TV it was all we could do. And the little fact that our team of referee reviewers could not do more games because we could use a few more helping hands.
So if you are a qualified referee and want to help us: just get in touch and join our team for next season. Let us see how the refs did in those games.
And with still the bad memory of our previous report in mind it is difficult to keep my feet on the ground. From the utterly depths of despair surrounding the Stoke City numbers we go to referee dreamland in this report.
In the Swansea games the ref made a very good score of more than 81% correct calls. This is just under 10% better than the league average. After yesterday’s depressing numbers this is what the doctor ordered. And if we put weight on to it we can see that the % of correct decisions is almost the same. And this means that in the Swansea games the refs have done 10% better than compared to the league average.
SO IT CAN BE DONE!!! In fact this is at least how it should be done for the whole league! So the refs can do it. Let us have a look at the decisions to see how which type of decisions was better than the league average.
The goal decisions are coming close to what it should be. Almost 4% higher than the league average so Swansea City games was blessed with some good decisions when the goals were scored. Only 1 wrong decision. We demand more of course and a score of 99% should be the ultimate goal but after yesterday I would take this for the whole league in an instant.
The offside decisions are very much in line with the league average, a bit better but not by much.
In the other decisions we see that they have a score higher than 81%. So that is very well done by the refs. Everything can improve but this is a nice start.More than 9% better than the league average.
Also the penalty decisions are much better (+17%) than what was the league average.
And look at the red cards! 45% better than the league average. Still not good in total but this is very well compared to the rest. A remark is that Swansea play a game of possession and keeping the ball in the team and on the ground. And as a result you seem to get not a lot of red cards in their games. In fact you had very few fouls in total in their games.
And also the yellow cards handed out are almost 8% better than the league average. Well done referees!
So few mistakes, but how was it divided? Let us check this in the next graphic.
If we look at the unweighted numbers first, as you might remember the normal negative away bias a team meets is -1.826 and Swansea getting a negative away bias from -1.750 is as close to normal as you can get.
The normal positive home bias is +1.826 and Swansea got a score of 3.000. So there is a slightly higher than normal home bias in their favour. And this leads in the unweighted score to a small total bias in their favour. Now let us put some weight in to the decisions.
The normal negative away bias in the league was -2.619 and for Swansea this was -2.500. So almost normal.
The normal positive home bias in the league was + 2.619 and for Swanseay this was +2.500. So again as close to normal as it can get.
And the final result is again starting to make me drool over these numbers. Because if we put weight on the decisions we see that the final score in total is 0 (zero).
So looking at that number I can say that if we look at Swansea City (and just at their numbers alone) : IT EVENS OUT!!!
So let us check the referees who are responsible for these numbers.
From the 7 refs in total who did their game there are two refs that step out of line one could say. Chris Foy with a very negative score, and Michael Oliver with a very positive score. The 5 other refs are very close to the zero score one could say. 5 of them having a positive bias and this is normal if we look at the unweighted final score in the previous graphic. But this is all rather marginal.
However if we put weight on the decisions we see that now 3 refs had a negative bias against Swansea. But apart from Foy who had a bad score it is rather close. The only one with again a rather out of line positive bias is Michael Oliver. But they almost cancel themselves out a bit and so we get the magical zero score in bias at the end of the season in the games we reviewed.
Can we see some individual scores reflected in the points? Well the first game they got more going against them and lost. But as there is no big decisions going against them it doesn’t seem to have affected the final score.
In game 21 they had some decisions going their was and won 3 points. In game 28 they had some things going against them but could win. In game 37 they didn’t have much in their favour and dropped points.
So maybe they had a small benefit left or right along the way but it all seems very respectable.
IT CAN BE DONE!
This is an example on how it should be. The refs in these games have shown that they can produce good games and acceptable scores. The overall score was good, the different type of decisions were good. So the refs can do it. It just seems to be problem to do it on a constant basis.
There was a very small bias in favour but it all was so close to the league average that it stayed well in line. A remark I must make is that the home/away bias should not exist in a perfect world.
If we look at this team alone we could say that it can even out for a team. The big question we should raise now is : why doesn’t it happen to the rest of the teams ? If it can be done for this team, why not for the others?