By Tony Attwood
This article originally included a quote of the number of VAR events per game in the PL, with the wrong figure quoted. The error (totally mine – I blame sticky hands on the calculator) doesn’t affect the key argument that the PL games there is a much higher level of VAR corrections than in the Swiss League which has a near identical number of games played each season. Thus I’ve removed the stat as it has no bearing on the piece. Apologies for the error, and thanks to those who pointed it out).
The VAR figures have been published for the 2019/20 season and we can now see that all told, this past season 109 events on the pitch were overturned by VAR in 380 games. This is interesting because prior to VAR being introduced the League wrote (and it is still on their website) “There will be a high bar for VAR intervention on subjective decisions to maintain the pace and intensity of Premier League matches.”
Yet it turns out that VAR overturned nearly three time as many decisions in the Premier League as VAR did in the Swiss league – which played a similar number of matches in the season. Suggesting referees in the PL made or experienced three times as many mistakes and uncertainties as their colleagues in Switzerland.
But first let’s understand how this system is working by looking at Arsenal’s figures in particular and the PL overall.
Arsenal had nine changed events as a result of VAR – five against the club and four in the club’s favour. So in the end “it all balanced out” as the old phrase has it.
Except that it wasn’t this way for all clubs. For two clubs benefited enormously from VAR. Top of the charts were Brighton who had a net score of eight major events in their favour in the season, including seven goals against them being disallowed!
Manchester United were second with seven goals against them disallowed!
So when we look at all the benefits the club had (such as a goal against being not allowed or a goal they scored that was not allowed by the referee but allowed by VAR), we do find two clubs benefited tremendously. This table shows the number of goals each club gained or lost as a result of VAR through the season.
- Brighton & Hove Albion +8
- Manchester United +7
- Crystal Palace +4
- Burnley +3
- Newcastle +3
- Southampton +3
- Liverpool +2
- Leicester City +1
- Tottenham Hotspur +1
- Manchester City 0
- Arsenal -1
- Everton -1
- AFC Bournemouth -2
- Chelsea -2
- Watford -2
- Aston Villa -3
- West Ham -4
- Sheffield United -5
- Wolves -5
- Norwich City -7
What the figure of -1 means for Arsenal is that Arsenal were overall worse off in terms of goals because of VAR than they would have been without VAR.
Here are Arsenal’s details through the season… Data is from ESPN.
Game: Manchester United (A; Sept. 30)
Incident: Pierre-Emerick Aubameyang goal, originally ruled out for offside, overturned on review, 58th minute – FOR
Game: Crystal Palace (H; Oct. 27)
Incident: Penalty awarded for foul by Calum Chambers on Wilfried Zaha (initially booked for diving), 32nd minute – AGAINST
Incident: Sokratis goal disallowed for foul by Calum Chambers, 83rd minute – AGAINST
Game: Norwich City (A; Dec. 1)
Incident: Pierre-Emerick Aubameyang missed penalty retaken (and scored) following encroachment by a defender, 29th minute – FOR
Game: Brighton and Hove Albion (H; Dec. 5)
Incident: David Luiz goal disallowed for offside, 63rd minute – AGAINST
Game: Crystal Palace (A; Jan. 11)
Incident: Pierre-Emerick Aubameyang sent off for foul on Max Meyer, 65th minute – AGAINST
Game: West Ham United (H; March 7)
Incident: Alexandre Lacazette goal allowed after incorrect offside against Mesut Ozil, 78th minute – FOR
Game: Leicester City (H; July 7)
Incident: Edward Nketiah sent off for foul on James Justin, 75th minute – AGAINST
Game: Watford (H; July 26)
Incident: Penalty (scored by Pierre-Emerick Aubameyang) awarded for foul by Craig Dawson on Alexandre Lacazette, 2nd minute – FOR
In terms of goal scoring, the extremes there show that Brighton and Hove got seven more goals than they would have done without VAR, while Manchester United and Southampton got six each. At the other end, West Ham suffered the most with a final tally of losing five goals.
Putting their data together we can see that VAR overturned 104 decisions during the course of the season. Without VAR Manchester United would have slipped out of the top four and Brighton might have been relegated.
But there is more, for what is also interesting, and what I think no one has noticed until now is that this is very different from VAR in Switzerland, the only other country for which I have comparable data.
Clubs in the Swiss Super League play two games fewer than Premier League clubs (36 rather than 38), but whereas in the Premier League 104 decisions were overturned in the season, only 36 were overturned in the Swiss League. There were almost three times as many VAR correction in the PL as there were in the Swiss League.
Now that is a huge difference. It suggests that either the VAR was working in utterly different ways in each country (which is not how it is supposed to be), or the employees of the PGMO were almost three times as likely to make mistakes as Swiss referees, and VAR has found them out.
How can that possibly be? The Premier League is a much bigger League than the Swiss League in terms of earnings, and therefore presumably in terms of its ability to have the best referees. So how can they be making three times as many errors as their Swiss counterparts?
It is just another unanswered question about PGMO referees, and one that will remain unanswered as long as the wall of silence around PGMO is allowed to go unchallenged by our puppydog football journalists who seem utterly unable to do anything except praise one player and criticise another.
But here are some suggestions:
a) Because PL refs including VAR refs each undertake many more matches than their colleagues in Switzerland, they get tired and don’t see everything clearly.
b) Because of the culture of secrecy surrounding PL refs they know that they don’t have to worry about being found out and so are less careful.
c) Because issues such as the way that crowds influence referee decisions (see links below for articles where this is analysed) are ignored in English reporting of games, and because PGMO has put out statistics that claim their referees are over 98% accurate (which the no crowd noise research contradicts) there is no desire to show that referees are improving, as this would undermine all previous press releases and their mock statistics about referee accuracy.
But there is one other point. The PL website contains a huge amount of statistical data but it has nothing concerning changes relating to VAR and how they compare with other countries or other leagues.
And this is not the only oddity, as we found when we were analysing the clearly linked topics of tackles, fouls and yellow cards. Lots of data, but the one crucial piece we needed which ultimately (when found from another source) revealed the utterly weird and unique figures for Leicester City in terms of tackles per foul and fouls per yellow card.
Of course this might be another coincidence, but these sudden gaps in the statistics do make me wonder.
And one last point: why is it left to Untold Arsenal to do all this work? Why won’t our magnificent football media tackle topics like this? They are, after all, paid to do the work. We struggle to raise enough money to keep the site running.
But we do keep going and here are some of our articles on referee bias.
- The relationship between tackles fouls and cards
- How a club can commit the most fouls but get the fewest yellow cards
- The shocking figures when fouls, penalties and cards are compared.
- Those who can make you believe absurdities: how the media affects refereeing
- The shocking statistics on how referees have affected Arsenal’s league position
- Crowdless Stadia Stats.
- How referees are affected by crowds
Great work again Tony. However, this makes no mention of penalties awarded which may have been dubious at best. I know there’s a level of subjectivity here but Man U’s 21 this season have included some real howlers for which other clubs might possibly see a player carded for diving.
Further, you have, of course, mentioned the Nketiah VAR sending off against Leicester but we all know that many fouls of a similar (and much worse) nature go unpunished. I’m not in anyway criticising your work or conclusions here though but it’s as much (if not more) about what VAR chooses not to see as what it officially picks up. What I’m actually suggesting is that this level of “error” is just the tip of the iceberg in my view. I wouldn’t be at all surprised if our refs are far more than three times as incompetent (or whatever) as the referees in most countries in Europe.
Great research. If truth must be told & which might be bitter, most of the PL referees are below average. They seem to go into the field with mindsets for or against some teams. They do certainly have their favorite teams. Some decisions of the VAR too are questionable due to the same factors. Swiss referees are better not b’cos of the size of their league, but b’cos of the integrity of the persons involved. How many of the Pl referees are involved in World cup matches compared to those that feature from other leagues? Pl referees need to wake up, b’cos, week out, they keep wrecking people’s joy for their controversial decisions, which in some cases get the support of VAR. The use of VAR therefore needs to be standardized, as well.
The maths in the first paragraph is wrong. 0.3 events per game would be a more accurate representation.
3.5 events per game would mean 1330 decisions overturned.
James is correct. 3.5 events occur every game week.
Thank you – that’s what comes of trying to get the article researched and written before I go to work. Better to be fully awake. I’ll see what I can do in terms of re-writing.
49, thanks, as per more note, I’ve taken out that stat as it is not relevant to the issue within the piece, and just a bit of background.
Interesting article here: https://www.skysports.com/football/news/11095/12047992/premier-league-referees-set-to-receive-bonus
Paid even more for being rubbish! Mike gets to choose I reckon.
The biggest issue in the officiating shortfall is the incidence of simulation. The simulation of officials is probably the most unnoticed by the simulating media.
I suppose it all goes round and comes round and that’s why we call it a ball!
There is anpother element of why EPL referees make so many mistakes as compared to their ¨foreign¨ colleagues. From what I can determine, there is very little followup by the PIGMOB after an official handles a game, unless there were serious and highly visible errors noticed by everyone. That is simply not the case with the PIGMOB or the EPL for that matter. Basically the PIGMOB, the organisation mandated to oversee officials in the EPL simply doesn’t care.
There is another element of why EPL referees make so many mistakes as compared to their ¨foreign¨ colleagues. From what I can determine, there is very little followup by the PIGMOB after an official handles a game, unless there were serious and highly visible errors noticed by everyone. That is simply not the case with the PIGMOB or the EPL for that matter. Basically the PIGMOB, the organisation mandated to oversee officials in the EPL simply doesn’t care.