By Walter Broeckx
As we were talking about my ref review of the year I mentioned in passing that Old Trafford idiom which seeks to explain why they usually get more penalties than other teams: “Because when you have more possession you obviously get more penalties”. People rightfully challenged me on this and asked me if there is any evidence for this.
So I went on a search on the internet to see if I could find something. Finding the penalties given for and against each team was not a big deal. Finding the time of possession was a bit more difficult. But I found some numbers and I just hope they are right. But at first sight they do seem to be possible.
And so I tried to do some calculations based on the Old Trafford principle: the more possession you have the more chance to get penalties.
| 2010-2011 | ||||||||||
| Team | Poss | Not poss | Pen for | Pen ag | Statistic for | Compare | Statistic ag | compare | ||
| Chelsea | 58% | 42% | 8 | 3 | 6 | 2 | 4 | -1 | ||
| Manchester United | 56% | 44% | 5 | 6 | 5 | 0 | 4 | 2 | ||
| Arsenal | 60% | 40% | 7 | 9 | 6 | 1 | 4 | 5 | ||
| Manchester City | 53% | 47% | 9 | 4 | 5 | 4 | 5 | -1 | ||
| Tottenham | 53% | 47% | 7 | 6 | 5 | 2 | 5 | 1 | ||
| Liverpool | 52% | 48% | 7 | 7 | 5 | 2 | 5 | 2 | ||
| Fulham | 50% | 50% | 3 | 1 | 5 | -2 | 5 | -4 | ||
| Bolton | 46% | 54% | 5 | 6 | 5 | 0 | 5 | 1 | ||
| Aston Villa | 48% | 52% | 5 | 4 | 5 | 0 | 5 | -1 | ||
| Everton | 50% | 50% | 4 | 2 | 5 | -1 | 5 | -3 | ||
| Newcastle | 50% | 50% | 4 | 5 | 5 | -1 | 5 | 0 | ||
| Wigan | 50% | 50% | 3 | 6 | 5 | -2 | 5 | 1 | ||
| West Ham | 44% | 56% | 6 | 6 | 4 | 2 | 5 | 1 | ||
| Sunderland | 48% | 52% | 5 | 6 | 5 | 0 | 5 | 1 | ||
| Blackpool | 49% | 51% | 8 | 3 | 5 | 3 | 5 | -2 | ||
| West Bromwich | 48% | 52% | 5 | 6 | 5 | 0 | 5 | 1 | ||
| Stoke | 38% | 62% | 3 | 5 | 4 | -1 | 6 | -1 | ||
| Blackburn | 41% | 59% | 2 | 7 | 4 | -2 | 6 | 1 | ||
| Wolverhampton | 50% | 50% | 3 | 4 | 5 | -2 | 5 | -1 | ||
| Birmingham | 47% | 53% | 2 | 5 | 5 | -3 | 5 | 0 | ||
| Average/pos/penalty | ||||||||||
| Average | 49,55% | 50,45% | 5,05 | 5,05 | 10,19 | |||||
So this table gives the link between penalties for and against and compare it to possession. The most important columns are the COMPARE ones If you see a score of 0 (zero) in this table this points at the fact that the team got as many penalties for or against as you could expect from their possession. Because for every 10,19% of possession you can expect to get one penalty a season. Thus meaning a team with a possession of around 60% should be getting 6 penalties. A team with around 40% should only get 4 penalties. And the other way round when it comes to getting penalties against you.
So the next step is to see if there could be a link between this and to try to do this I had to look at the data and see if I can see some kind of pattern. And I tried to put this in a table.
| For | Against | |||||||
| Range -4 | 1 | Fulh | ||||||
| Range -3 | 1 | Birm | 1 | Everton | ||||
| Range -2 | 4 | Fulh,Wig,Blackb,Wolves | 1 | Blackpool | ||||
| Range -1 | 3 | Ev, New,Sto | 5 | Chel,MC, AV,Stoke, Wolves | ||||
| Range = 0 | 5 | MU,Bol,AV,Sun,WBA | 2 | Newc, Birm | ||||
| Range +1 | 1 | Arsenal | 7 | Tot,Bol,Wig,WH,Sun,WBA,Blackb | ||||
| Range +2 | 4 | Chels,Tot,Liv,WH, | 2 | MU, Liv | ||||
| Range +3 | 1 | Blackp | ||||||
| Range +4 | 1 | Mancity | ||||||
| Range +5 | 1 | Arsenal | ||||||
| 20 | 20 | |||||||
And so I divided it in ranges. But because you cannot work with half a penalty I had to stick to the numbers before the comma.
And then I had to take as normal the range from -1 over 0 to +1. Because this can just be the result of the fact I had to get to the number before the comma. If you are still with me?
As a result we could say that when it comes to penalties for we can see that 9 teams are totally normal in this. And then we have 8 teams who have a slightly difference but only just one penalty away from being normal. So that makes it 17 teams that are very close to making it look normal that there is a link between possession and penalties.
There are only 3 teams with an abnormal score in this. Birmingham who has had 3 penalties less than could have been expected. And Blackpool and Manchester City who have had 3 and even 4 penalties more than you could have expected when it comes to comparing possession to penalties.
Let us turn to penalties against.
And then again we have the same range as normal and this time we have 14 teams in the normal range. 3 teams who have a slight out of order range. So again we make it at 17 teams in the range or close to the normal range.
We have two teams who hardly make fouls or who hardly get penalties against them like Fulham and Everton. And that brings us at a total of 19 teams.
And then we have the major exception to all the rules this season….Arsenal. We have had 5 penalties given against us more than you would have expected if you compare the possession and the penalty decisions. You can see in the table there is a huge gap between us and the rest of the league.
So could I prove that there is a link between possession and penalties? Not 100% but for most teams the numbers are more or less closely linked at first sight. Some 17 teams both in terms of penalties for and against who are only 1 penalty away from what could be expected. That is 85% of the teams.
But I certainly could show that Arsenal had a lot of penalties given against us and totally out of line with the rest of the league. In fact Arsenal had the honour of almost having 10% of all the penalties given in the EPL against teams was given against Arsenal. The exact number was 8,9% of the penalties. That sure is a lot.
So if the Old Trafford rule is correct we surely are the exception to the rule.
————-
How every book and article gets the name of Arsenal’s second manager wrong.
Untold Arsenal on Twitter is in the top 1% of all Twitter sites for followers @UntoldArsenal
Untold Arsenal on Facebook here
—————————
History of Arsenal: how every history book gets the name of Arsenal’s second manager wrong.
Making the Arsenal – the book of Arsenal death and rebirth
Part of the ‘Old Trafford factor’ is that the team playing at home gets more penalties because (presumably) you have more possession at home – not least because the opposition retreats and allows it and your home fans demand it.
One of the great things about Wenger is that he demands that his team plays the same away as they do at home and that we usually have more possession of the ball away than does the home team. It worked very well in the early years of his management (we went two seasons unbeaten away from home) but less well in recent years as opponents have got wise to it.
It’s also key to understand how possession is used.
Stoke for instance use there’s to set up set pieces (including vast numbers of throw-ins) so that their defenders can come forward and use a crowded penalty area to work their ‘magic’. This rarely wins them penalties – more likely a free kick to the other team.
A penalty won by dribbling the ball into the area and being brought down(more the Arsenal style) is very much reliant on the referee spotting the split second contact/lack of contact and that (as we’ve seen so many times this season)is not something that they do that well.
interesting statistic,
do you have this kind of data for the European Champion League?
perhaps it can prove if Arsenal were unbiasedly treated by the Primer League referees.
Walter, I bet sometimes you wish you had pursued a life in the field of statistics :-)… you seem to enjoy the numerical analysis!
This is such a cool analysis … thanks Walter !
Just for clarification – I think that the “OT rule” means “More possession in the opponents box” rather than just possession.
What about possession in the offensive 1/3 of the pitch, or possession in the box…after all that is where the penalties are called. Possession in midfield means nothing except time wasting.
If anyone has the numbers of possession in the opposite box I would love to have a go at them. 🙂
Phoenix gunner,
give me numbers and I start looking for something.
And to tell you the truth I was really really bad at maths at school… 😎 Still am I think… 🙂
@Walter: I didn’t want to be a one-note pain in the arse, so I’d like to ask again if you are/might still be considering going forward this summer with the companion video to the important calls/non-calls from your report? Cheers!
Here are stats for possession in the final third, I’m not sure it will be possible to find numbers for penalty box possession. Using the drop-down tab you can find the numbers for all PL clubs.
http://www.whoscored.com/Teams/13/Analysis/England-Arsenal
Bob, I am still gathering the data on my computer. But it is a lot of work to get all the games downloading… But if I can find the time in the next month I will sure try to put them together. But to put them in one video I will need some help. I hope one of my sons can help me out on this.
Wrenny,
Thanks a lot. I will be going back to work with these numbers.
@Walter: many thanks for considering the video. It’s easy to make requests on this end, and it’s your time that’s in play. Perhaps some of us can make technical suggestions for Walter and Son(s)that could speed up the downloads and the editing? Maybe others already have downloaded matches and can compress the files and share them with Walter? I don’t know, but there would be great value is such a video on many levels and any help to streamline the needed effort would be of help to anyone who wants (or doesn’t object to) the analysis and education that UA/Walter provides to expand and be more effectual in its influence on behalf of next season’s prospects. All of us could learn a lot by then reviewing Walter’s analysis of key fouls and having the video evidence (perhaps linked to from your table(s), Walter) to check against. Surely there’s some help that UA fans could muster or suggest?
@walter, if there’s anything you need just put a shout out, people here will do anything within their power to help. Maybe list things you’d like to do with things that are holding you back so everyone can see where you need help.
One of the problems is that I only use the minutes indications since a few months. And so I have to see the game to get to the key incidents.
But I just get to it when I get some spare time. But for some reason I just cant find the time for the moment. And that is a bit crazy as I have not ref reviews to do…
But other and more urgent work keeps coming between it for the moment. But we will get there I hope.
Walter,
I can do all the voodoo with regard to video editing/encoding etc (I used to provide IT support/consultany to the Broadcast Department of BBC Television). If you were to provide me with a list of incidents (with their time) for a specific match, I would be happy to create a edited “Higlight reel”. I am sure that I can find any matches that are needed for this.
If we have several people doing this part (and I am sure that there are plenty who visit that are capable), it will be very easy for next season and would free you up to review the earlier games for editing.
Additionally, I am happy to (initially – as long as the bandwidth usage isn’t silly!) host the video files, if that would help.
You have my email, if you are interested in any help, just drop me a line.
The issue of penalties awarded versus penalties merited is a very subjective one. Since the referee is the final judge and jury on intent,timing,degree of danger to opponent or absence thereof, who touched the ball,if at all, etc. this kind of statistical analysis really can’t ever hope to prove Arsenal were ill done by. That said, there is certainly a questionable tendency for EPL officials to award more penalties against AFC, and that merits a more in-depth review!