From the Premier League referee who almost never sees a home win, to a referee who hardly ever sees an away win…
The Premier League has appointed Peter Bankes as the referee for the trip to Brentford on Wednesday. He has overseen ten games this season – and he is known as the referee who sees fouls very, very rarely. In fact, in seeing 18.5 fouls per game he sees around 20% fewer fouls than referees such as Anthony Taylor. Indeed among the top referees in the PL only Anthony Taylor sees fewer tackles as fouls (all figures from WhoScored).
But rather curiously he is a bit of a penalty giver, having awarded four penalties in ten games. But when it comes to giving out yellow cards he is mid-range with four yellow cards per game and just two red cards across all ten of his PL matches this season.
And rather curiously he gives out more yellow cards to home teams than away teams – 18% more this season. That could offer a little help to Arsenal in the forthcoming game, as long as he stays with this bias.
But to see what we might call the “strange” or perhaps “eccentric” nature of PGMO at work yet again, we may note that just as in the last match we had a referee who had not overseen a single home win all season until he got Arsenal at home (when the sheer inability of Ipswich Town to mount an attack meant that all the help in the world couldn’t get them a goal) so now in this forthcoming game with Arsenal away, we have a referee who has overseen eight home wins out of ten games this season.
Let’s be clear. In the last game, played at home, we had a referee who had not overseen a single home win before the Arsenal game. Now in this game away we have a referee who has overseen eight out of ten home wins.
And yes of course you could argue this is just a coincidence. One of those things. And besides you can prove anything with statistics.
But actually you can’t. There is no getting away from the fact that when at home Arsenal got a referee who until that point had never seen a home win this season. Now Arsenal are away from home and the referee has the second-highest level of home wins in games of any referee in the league this season.
There are 15 referees who have overseen seven or more Premier League games this season, and we could have had any of them. But when Arsenal are away we get the referee who dishes out home wins as if they were going out ot fashion, and when Arsenal are at home, we get the referee who sees more away wins than any other.
Now these are absolute facts, not opinions, and you can see them for yourself on the WhoScored Website.
The answer given to statistics like these is usually one that says that it all evens out in the end. But in fact there is no evidence that it does. If the referees were not biased we would not see the range of home wins or away wins per referee between the top ten referees on the PGMO list, yet we do.
The range of home wins per referee, among the most used referees, goes from 7% to 80%. The range of away wins per referee goes from 64.7% down to 20%.
In the Premier League, there have been 188 games. 40% have been home wins, 32% away wins and and 28% draws. Now obviously there must be some variations between referees, depending on which games they see, but we might expect overall to see the level of results per referee somewhere around the average. Figures are rounded to the nearest whole number.
| Referees | Games 24/25 | Home wins | Away Wins | Draws |
| Total games | 188 games | 40% | 32% | 28% |
| Ref Kavanagh | 13 | 8% | 23% | 69% |
| Ref Bankes | 10 | 80% | 20% | 0% |
| Ref Taylor | 17 | 17% | 65% | 18% |
.
By any analysis these figures are outrageous. Of course we don’t expect to have all referees seeing 40% homes wins, 32% away wins and 28% draws. But to have one referee overseeing just 8% home wins while another gets 80% home wins is, to put it in the neutral of terms, “unexpected” and “worthy of investigation”. Likewise to have one referee see 65% of his matches end as away wins while another is running at a third of this rate, is, what shall we say? “Unexpected” is perhaps the best neutral word.
As for draws running between zero and 69% of all the matches seen – well. If you are trying to explain this by pure chance, then your understanding of pure chance is different from mine.
And yet absolutely NONE of this is mentioned in the media. So we might ask not only how can such variance happen, but also WHY IS THE MEDIA NOT REPORTING IT?
In earlier years, my postgrad research was not in maths, but in the social sciences, an area of work that, you may know, uses a lot of maths, and I had to be pretty nifty with numbers to be awarded the research degree. And I can tell you from that experience, even though it was some years ago, if figures like these ever turned up in research the immediate reactions would be a) you’ve made a mistake and b) if you haven’t made a mistake then there is something very fish going on here.
And I’ve checked – and you can check my numbers. I don’t think there is a mistake.
So if you agree these results are at least “curious” if nothing else, you might join with me in asking why the media never mention them.
