A perfect and total example of how our perception of football is being manipulated

By Tony Attwood

Paul Wilson writing in the Guardian recently made this comment…

 Referees do not always make the right decision on the spot, and sometimes they make the wrong decision for the right reasons. But with time, hindsight and replays, the disciplinary commission should be capable of sorting all that out. 

Now on the face of it that seems fair.  Really, nothing to argue with is there?

Well actually there is one rather clever omission from this commentary – an omission that twists the whole argument about referee errors down one particular alleyway without giving the reader a chance to see what has happened, or indeed consider another approach.

It is one of those rather dastardly techniques that have become the centrepiece of contemporary football journalism and which has led to the manipulation of the way in which we see football, as I will try and explain.

So the Guardian article ignores totally the issue of the level of referee accuracy and instead talks just about this one incident in isolation.   Of course the appeals procedure itself helps keeps the myth alive for as the Guardian reports, Vardy, “argued the referee had been wrong to send him off, the commission found the official had sufficient reason. End of appeal, end of story.”

Except that if the journalist was doing his job he would ask, “how many major decisions do referees get wrong each week?” and look around for a source of information.   But the journalist didn’t and indeed even if he had, I suspect the Guardian would have felt the story too hot to handle.

I reach this conclusion because of the way the Guardian ends its piece…

That’s football – it can never be perfectly consistent and often the best plan is to simply shrug and accept its ups and downs. But after the past couple of weeks surely someone within the FA ought to be at least reflecting on what really constitutes a reckless challenge, violent conduct and dangerous play.

The difference is that I would write

But after the past five years of detailed analysis surely someone within the media ought to be at least reflecting on just how many errors referees make, and why PGMO withdrew their claim that their employees are 98% accurate.


15 Replies to “A perfect and total example of how our perception of football is being manipulated”

  1. I would find genuine errors by referees tolerable (speed of game, they are human after all).

    The real concern should not be about “errors” ( a term which implies innocence), but calculated wrong decisions which are the result of bias on the part of match officials.

    The reviews on this site have produced abundant evidence that this is the reaality

  2. Following from John L – they must take a few bob here & there, they are human after all….!!!
    Nothing better than an open recruitment of officials while on field radio communications are broadcast to public & appointments are made by draws from a bag to ensure open selection of officials witha maximum of 2 matches per team per season with an extra one for cups.

  3. Interesting one this week was Mason shaking his head for a pen claim then pointing to the spot a second later. Seems about 99% certain the head shake was him deciding no pen, and then he changed his mind because someone told him otherwise.

    Now, nothing at all wrong with assistant referees offering advice, but wasn’t the old way much better, with a flag and a word, consultation people could see? More importantly, shouldn’t such a massive change have been officially introduced somehow?

    If that’s what it was, a linesman saying penalty to a ref who, with a perfect view, decided no, and the ref then going with that decision without a moment’s thought…such a change in the game needs to be discussed honestly and weighed up, surely.

    What’s more, can the fourth official now perform this role as well, can they say yellow or red, and are refs liable to accept what they say as easily as Mason appeared to? Especially when the ref might be a young pretender and the fourth official very experienced.

    If we turn to rugby, it’s easy to forget that these officials are all miked up and communicating and, incredibly, you get to hear it all…how decisions are made, when they get things wrong. Just this week a ref managed to get a huge decision completely wrong which changed the result of a game. He got a rule completely wrong about whether there was time for a lineout- there wasn’t- which led to a match-winning try.

    Embarrassing for the ref, but mistakes happen.

    Once more though you’re asking why is rugby mature enough to let us see all that, why can players who are trying to smash each other to bits be trusted to behave with a microphone around them? The sad truth is football is probably deemed to big and too profitable to allow the product to be spoiled with bad language and, well, bad refereeing being displayed to all.

    It may be that this plays into the hands of those with much more sinister motives, who want refereeing to stay just as it is, out of sync with virtually every other major sport. Closed off, secretive, and well designed to cover mistakes and worse.

  4. Other thing to grab my attention this week came in the top 2 clash between Bayern and Leipzig.

    Bayern already a goal or two up when, around 30, their player, chasing Lahm, lunged and studded his achilles, with no hope of reaching the ball. Ref ran over brandishing a yellow but his linesman told him to change it to red. Justice was probably done.

    What made my heart sink was how closely the foul resembled that of Carroll’s on Kos. Only difference being Carroll’s was far more brutal and dangerous.

    This is the thing: the demand is there for us to become as good in our league as Bayern are in theirs, an extremely tough ask. But Bayern will get those decisions, they will get the protection of their players, they will get the pens. If someone makes a red card challenge on their players, top of the table clash or not, first half or not, they’ll get the red. We absolutely will not.

    Handily, BT sport instantly put out a clip of the incident on their twitter feed, for all to view, replete with Hargreaves commentary- as clear a red as you get, disgraceful challenge, etc.

    Whereas if you want to look again at what went down with Carroll and Kos, you’d struggle to find it anywhere on twitter (I couldn’t), let alone on BT’s official feed. This is important, in my opinion, in allowing the same shit to go on and on.

    An alternate world almost. If you could find the real time commentary for that Carroll foul, I’ve no doubt they wouldn’t be saying anything about certain red, etc, and it may even be the same man, Hargreaves, commentating on the game.

  5. Rich
    Yes i have noticed that in UK tv the replays are very rarely shown. Why? is what i would like to know. Very fishy.

  6. Hear, hear to all the previous comments.

    The bottom line in England is that officialdom doesn’t want fairness because then it cannot pursue its unwritten agenda of helping certain clubs at the expense of others. Its that simple.

  7. I think there are assumptions beyond what Tony has pointed out.

    But how many of us have seen images or video clips (few seconds at most), which show at least one foul? Maybe the “image” comes from a TV camera that was distributed to TV (and actually broadcast), or maybe it comes from some newspaper photographer at the game.

    Just looking through News, here is a link to an article about a game in the Arabian Gulf League, with an image of two players in a game. The player in yellow has one hand in contact with the player in white, and the other hand unseen as to contact. Neither hand is an a position that is “fair”.


    I don’t think there would be many referees who would call a foul on the incident, but the contact is contrary to the laws of the game.

    There are about 100 minutes in a game (including time added on for stoppages). We probably can think of players who could foul as often as once every 2 seconds (maybe not continuously for 100 minutes). But that gives us an upper limit of something like 3000 incidents in a game. Some of these incidents could be called as fouls (which would require the ball be in play), some of these incidents could be called as fouls _IF_ the ball was in play, and some probably would require the issuing of a card (unsporting conduct, etc.). We’ll assume that it is possible for players to go through the entire game doing nothing that is contrary to the laws, but instead of saying 0 is possible, we’ll call that 1.

    Question is, what is the average number of incidents by players during a game? The geometric mean of 1 and 3000 is about 55, maybe that gives us a hint? I will guess that the average number of incidents per player in a game, is something like 30. Which means that in the course of a game, there are (on average) about 660 incidents which the referee should have acted on. In a rough game, a referee may call 30 fouls, which is about 5% (of 660, but if it is a rough game, the average number of incidents is probably well above 660). And let’s say the variance is over-dispersed wrt a Poisson, such that we have a standard deviation of 10 (a variance of 100).

    It is almost a situation that not only is the statement that the PGMO staff are 98% incorrect, but that PGMO staff fail to act on 98% of the incidents which happen in a game.

    The big difference between lower leagues and youth leagues, and the EPL is visibilty. In the lower leagues and youth leagues, there may not (probably isn’t) enough cameras recording the game to catch even most events, let alone have cameras equipped properly. In the EPL, there are (or could be). The EPL should have analysis of all the TV cameras output (only 1 can be used at a time for broadcast) for every game. Some analysis can be done algorithmically (players in offside positions, officials calling players in offside positions) and some needs to be done manually. And then there are images which come from the generic media which become published after the game in some respect. Many of those images will show incidents (after the fact).

    Let’s say it is possible to analyse and catalogue the TV camera coverage within 3 days of a game. Before the next game occurs, and before the interviews of the manager for the next game occur.

    Two categories, officials and players.
    _Officials: it is a matter of record how many decisions they made, we now have a measure of how many incidents happened (some incidents may not be published, but we should have most). We can assign a grade to the officials.
    _Players: We have an average number of incidents (and its standard deviation)m ad for demonstration we’ll assume an average of 30 with a standard deviation of 10. We add 1 to both, and we want to work with numbers greater than 3 times the adjusted values.

    _(30+1)+(10+1)*3 = 64

    If we had a player with an exceptionally high number of incidents, they need to be dealt with. And how they are dealt with, depends on that number we just calculated (64). Except that if one player had 3000 incidents by themselves, the average could not be as low as 30. A detail I will ignore for now.

    Our mythical infringer could have 48 times as many incidents as this flag value of 64. The object is to issue a banning order. It doesn’t make sense to ban a player for more games than occur in one season based on what is nominally persistent infringement. So, we look instead at the square root of this number (which is under 7, so we call it 6). and we subtract 1, which gives us 5. And you guessed it, the perpetrator gets banned for 5 games.

    Let’s take a player with 65 incidents (just over 64). Divide by 64 (a hair over 1), take the square root (even closer to 1, so we are calling it 1), and subtract 1 and we get 0. No banning order.

    Calculated this way, it is only players with more than 255 incidents, who will be issued banning orders. This is almost 10 times our guess at an average number of incidents per player. Maybe a player is slightly under this 255 count at 3 days post game, but in subsequent days more incidents become known through the press. A late banning order could be issued.

    All stadia are to have available at the game, differential GPS capable of 20 Hz corrections (or better), and all “professional” cameras are to make use of that differential GPS (or better) and a 9DOF IMU (so that we know the orientation of the camera in 3-space), and each frame shall be labelled with that metadata. We hence know exactly where that camera was and in what direction it was pointing at the time the image was taken. (Exact to within a few cm or so.)

    People like to protest. Players shall be allowed to protest, as many incidents as they want. The fee to file a protest is one Pound Stirling multiplied by the square root of the weekly salary divided by one Pound Stirling. If a player makes 10,000 per week; to file a protest on _one_ incident is 100 pounds. Non-refundable. If they want to protest 10 incidents, it is 10 times as much. All protests must be made in person, no sending a lawyer as a proxy. All protest hearings only consider a single incident. If a player wants to protest 10 incidents, they need to make the time to appear at 10 hearings. The hearings are to take place before the next scheduled game.

  8. Will the video assistant referees – VARs not be the answer to the Pgmol match officiating officials stubbornness to change from gagging up against teams in the PL in their continuous manipulating of results in PL games in particular against Arsenal?

    Since the Pgmol referees are well protected by the oversight body of the FA from reversing the controversial decisions they’ve been taking in PL games mostly to keep blocking Arsenal from winning the title for 12 years on the trot. Whereas, Arsenal would have won the title at least on four occasions in those past 12 years.

    The VRAs when officially introduced to the PL games will go a long way to eliminate some of the bias major decisions the Pgmol referees have been getting away with against Arsenal or at least have them minimized to the barest minimum.

    The VARs are said will come on stream for official application in the major leagues in the world as from 2018-19 seasons. But what happens between now and that set date? Will the Pgmol referees continue in their bias match officiating against Arsenal for one to two more seasons?

    Since Fifa have successfully tested this VARs at the just concluded Club World Cup games in Japan and are said will be refining the application before the 2018 World Cup takes place in Qatar where the application is expected to be fully deployed for use. I think if the FA are sincere to see a cleaner referring in the PL matches by the Pgmol match officiating referees, the FA can order for the VARs be officially introduced in the Premier League matches as from the beginning of the 2017-18 season.

    Hiding under the guise of the system has not yet been fully tested for implementation will be unacceptable as the Pgmol referees can start to test the application in PL matches as from the second half of this current season with every intention to start using at the beginning of the 2017/18 season. It can’t take an age to learn how to use this VARs application just as it doesn’t take an age to learn how to drive a car, does it?

  9. Para

    I barely watch sky sportsnews anymore but I know they used to show some incidents on repeat, but even then they felt like editorial decisions which didn’t obey any particular groundrules.

    Also, it could be that tv regulations prevent sky, BT or anyone from showing incidents in a game immediately.

    Nonetheless it all feels very unbalanced.

    On this one, Bayern got the decision, got the commentator insisting vociferously it was a bad tackle that deserved red, then got the incident smoothly packaged rolled out for people to look at on the internet and social media

    On our one with Kos, a worse foul than the one against the Bayern player, we got the wrong colour card, commentators looking to excuse the challenge [a guess, but a fairly safe one] and, I expect, no similar clip from BT or any other broadcaster at any point.

    To cap it off, I strongly suspect the premier league patrol twitter and the internet and get any recordings removed of clips showing moments and decisions they’d rather disappear. Clips like that Carroll foul are always pretty hard to find weeks and months down the line.

    They have easily got the power to do that, and I remember them going hard after the website 101 goals to try make them stop showing any prem clips whatsoever.

    That Carroll/Kos one pissed me off so much at the time and this week’s action brought that injustice right back to me.


    The kos one again, with crappy music, no commentary, poor picture quality and a distinct lack of official broadcast status. Could be removed at any moment


  10. Rich

    Unfortunately there is no rational or reasonable explanation that can be given to explain why football is decades behind it’s sister sports. Not one.

    And I would not ask nor expect the friends and colleagues of Stuart Hall to report this or other important stories from football. After all these are the same people that spent decades ignoring the actual bona fide journalist Andrew Jennings’ work. And much much worse.

    I’m not going to ask the turkey who or what it voted for, I’m just going to eat it.

  11. I think it is wonderful that video may become a formal part of the game. But, I do not think that the video that we may think we know from years of watching TV, is what we are looking for.

    If the frame rate for current TV display of games is 24 frames per second, that is slow enough it just on the threshold of flickering. In modern camera sensors, we have an array of “counters” sitting behind different colour filters. At the start of a frame, all the counters are reset to 0. When the “shutter closes”, the counts in each cell are read. Usually, some kind of processing is done on the image just captured, and when that is finished, we are then ready to process a new frame.

    If I have a 3 bit counter, and I start having it count it will go:

    _0, 1, 2, 3, 4, 5, 6, 7, 0, 1, 2, 3, 4, 5, 6, 7, 0, …

    In other words, 7+1 = 0.

    Our counters in the camera are doing the same thing. If in the course of counting, a counter reaches the highest number it can count to; it can do a couple of different things. It can just “wrap” back to 0 and continue counting. And you get a warped pixel. Or it can set a flag to indicate that overflow happened. And it may be that part of overflow, is to stop all counting in all cells.

    If in recording a football match, we find that none of our cell counters get over half of the maximum count value; we can take advantage of that by doubling our frame rate. We will acquire 2 complete frames in the time previously allowed to acquire 1. If we do this once, we get to 48 fps, if we do it twice we get to 96 fps, and so on. The video used for assisting referees and for analysing the game, must be of a higher frame rate than what is necessary for TV.

    If the recorded from rate is N times the base 24 fps (where N is an integer), for TV all that is needed is to send every N’th frame to the TV broadcast stream. And the amount of data stored for the actual game is N times what we would need to record the TV version at home.

    Except that every frame, we want the GPS position and timestamp, as well as the 9 DOF inertial management unit (IMU) readings. Oh DOF, is degrees of freedom. An IMU has gyroscopes and accelerometers that allow it to track angles and accelerations, and for generic 3D purposes you want 9 DOF.

  12. “Recipes” for Christmas Cheer

    1/2 shot eggnog
    1/2 shot peppermint schnapps

    Mix thoroughly

    1.5 ounce green apple vodka
    0.5 ounce Triple Sec
    1 t grenadine
    3 ounce cranberry juice

    2 passionfruit, halved, seeded and mushed
    60 ml peach schnapps
    60 ml vodka
    2 cups cranberry juice

    serve with ice

    Oh, Santa’s postal code is H0H 0H0. And apparently he does answer. That is what Canada Post says anyway.

    You might want to change the formatting on the postal code:
    _ H0 H0 H0

  13. EggNog Based on CookingForEngineers recipe (2 parts)
    6 large, Omega enriched eggs that have been washed and checked for cracks
    1/4 cup sugar
    2 C whole milk (grass fed?)
    Cook (stirring constantly) over low heat until it reaches 160F and coats back of spoon. Remove from heat but place on insulated cutting board for 5 minutes. This is _not_ pasteruizing, but goes part ways.
    2 C whole milk (grass fed?)
    1 t (good) vanilla extract
    1/8 t fresh ground nutmeg
    Stir in this remaining whole milk and spices. Chill at least 4 hours. Serve within 24 hours.

    A typical eggnog recipe would be 1C of sugar, not 1/4C. If you wanted to lower sugar some more, maybe try partially replacing some sugar with Splenda, stevia, fructose. But not for next part.

  14. Many of you may have noticed that the price of eggnog goes crazy low after Christmas. What the heck can one make from eggnog? The milk and eggs are wonderful ingredients. I don’t think there are too many recipes that vanilla will negatively affect. We don’t need the sugar in our diet.

    Okay, what you can do with eggnog, is make egg pasta (noodles). You can make it for yourself, possibly storing it in the freezer after it has dried. Or perhaps a local charity will take dried egg noodles?

    So, get some durum wheat flour and a pasta maker, and make your own noodles. You are going to want and make your own dry noodles. Jamie Oliver is English, he has a recipe for egg pasta. But basically you find a good wheat flour for making pasta, and you add it to your liquids until you have a dough to work with. And then you knead it, knead it and knead it some more. You want to develop the gluten. But this is where a pasta maker helps, is in the lots of kneading.

Comments are closed.