VAR: An Unconscious Bias? (Updated)

[Joshref]

This is a topic I made 6 months ago, but following the reigniting of this debate yesterday in regards to the Burnley vs Manchester City red card, I thought we could revisit these stats again. I can link the original thread for more info on how I got the stats and everything but this is just a quick overview to see a pattern. Pre VAR is based on the average of the data for the 3 seasons before VAR was introduced (16-17, 17-18, 18-19). We ask the question

Is there any evidence that the introduction of VAR has seen referees reduce the amount of penalties and red cards given due to an over reliance on the technology?

Red Cards
Pre VAR: 42
19-20: 44
20-21: 48
21-22: 43
22-23: 30
VAR: 41

Penalties Awarded
Pre Var: 96
19-20: 92
20-21: 124
21-22: 103
22-23: 99
VAR Average: 105

Our stats can’t really tell us too much here. The number of red cards on average has slightly decreased and the number of penalties has increased. However, we can remove VAR decisions and look at on field decisions.

Red Cards
Pre VAR: 42
19-20: 38
20-21: 32
21-22: 32
22-23: 27
VAR: 32

Penalties
Pre Var: 96
19-20: 77
20-21: 117
21-22: 74
22-23: 85
VAR Average: 88

As I believe Chas stated last time I posted stats like this, do these stats prove anything other than the fact I have too much time on my hands. Stats can be manipulated by anyone to paint a picture, but I will point to some findings.

The first, the number of red cards and penalties awarded on field has been less than those awarded in total every single season. To phrase that better, the number of penalties and red cards awarded by VAR is always more than the number overturned. This is of little surprise. Red cards and penalties won’t be given by refs unless they’re 100% certain. It makes sense that a referee is more likely to miss a penalty than to incorrectly give one.

The second finding is one that has no explanation to it, and will be given one depending on where you stand on the debate. There were an average of 42 red cards shown in the 3 seasons Pre VAR, and 41 in the 4 seasons since VAR, so very little change in the number of red cards shown. Despite this there’s a 24% decrease in red card shown on field by the referee. The number of penalties has gone up since the introduction of VAR but there’s still an 8% decrease in penalties shown on field by the referee.

The stats will show what you want them to show, but I just thought this is an interesting update on VAR stats that I’ve updated as we come into a fun season.

 

[RustyRef]

Something else has changed in recent years though. Across the world referees have been encouraged to manage situations rather than necessarily use cards. We've seen this in the Euros, the World Cup, and in most domestic competitions, and that means reductions in red cards could well be down to a raised bar.

 

[Viridis1886]

This is a topic I made 6 months ago, but following the reigniting of this debate yesterday in regards to the Burnley vs Manchester City red card, I thought we could revisit these stats again. I can link the original thread for more info on how I got the stats and everything but this is just a quick overview to see a pattern. Pre VAR is based on the average of the data for the 3 seasons before VAR was introduced (16-17, 17-18, 18-19). We ask the question

Is there any evidence that the introduction of VAR has seen referees reduce the amount of penalties and red cards given due to an over reliance on the technology?

Red Cards
Pre VAR: 42
19-20: 44
20-21: 48
21-22: 43
22-23: 30
VAR: 41

Penalties Awarded
Pre Var: 96
19-20: 92
20-21: 124
21-22: 103
22-23: 99
VAR Average: 105

Our stats can’t really tell us too much here. The number of red cards on average has slightly decreased and the number of penalties has increased. However, we can remove VAR decisions and look at on field decisions.

Red Cards
Pre VAR: 42
19-20: 38
20-21: 32
21-22: 32
22-23: 27
VAR: 32

Penalties
Pre Var: 96
19-20: 77
20-21: 117
21-22: 74
22-23: 85
VAR Average: 88

As I believe Chas stated last time I posted stats like this, do these stats prove anything other than the fact I have too much time on my hands. Stats can be manipulated by anyone to paint a picture, but I will point to some findings.

The first, the number of red cards and penalties awarded on field has been less than those awarded in total every single season. To phrase that better, the number of penalties and red cards awarded by VAR is always more than the number overturned. This is of little surprise. Red cards and penalties won’t be given by refs unless they’re 100% certain. It makes sense that a referee is more likely to miss a penalty than to incorrectly give one.

The second finding is one that has no explanation to it, and will be given one depending on where you stand on the debate. There were an average of 42 red cards shown in the 3 seasons Pre VAR, and 41 in the 4 seasons since VAR, so very little change in the number of red cards shown. Despite this there’s a 24% decrease in red card shown on field by the referee. The number of penalties has gone up since the introduction of VAR but there’s still an 8% decrease in penalties shown on field by the referee.

The stats will show what you want them to show, but I just thought this is an interesting update on VAR stats that I’ve updated as we come into a fun season.

If you really want to get into the analysis, I think the stats you need is a Wilcoxon Matched Pairs analysis. That will tell you if there is anything significant.

 

[Joshref]

Something else has changed in recent years though. Across the world referees have been encouraged to manage situations rather than necessarily use cards. We've seen this in the Euros, the World Cup, and in most domestic competitions, and that means reductions in red cards could well be down to a raised bar.

Of course, but it is interesting the number of red cards is very similar. All that’s decreased is the number of red cards shown on field.

I’m not saying there’s anything to these stats btw I’m trying to remain unbiased. I’m relatively certain that there has not been a directive where they go “wait for VAR before making a decision” but just presenting the findings

 

[Viridis1886]

If you really want to get into the analysis, I think the stats you need is a Wilcoxon Matched Pairs analysis. That will tell you if there is anything significant.

As a rather sad aside, one season I did some statistical analysis on scores given to referees on a club forum in relationship to the result.

You will never guess what? It turns out that to high degree of statsistical significance (95%) that there is a relationship. When the team won, the ref scored highly, when the team lost, the ref scored badly. Whoda thunk it eh?