by Tony Attwood
Y
ou may have come across the Football Observatory, not least because its reports are often quoted in the media as a source of independent, quality analysis of what is going on.
And quite reasonably too, because The Football Observatory is a research group within the International Centre for Sports Studies (CIES), an independent study centre located in Neuchâtel, Switzerland which specialises in the statistical analysis of football. It was created in 2005 by Dr. Raffaele Poli (pictured left) and Dr. Loïc Ravenel.
As such it is statistical in its work and analysis and its work is overseen by recognised academics. Which is to say not your regular newspaper warriors who think that interviewing three fans in the pub is a survey or a dozen ranters on AFTV is a survey.
The work is CIES is quite rightly commented upon because it is serious stuff, but because it is up to the media which of the CIES press releases and reports are picked up and turned into news, the bias of the media is always there.
And indeed this affects what we read since blogs as well as professional (I use the word in its money-taking sense, not to imply any standards or authority) journalists choose which of the findings they want to highlight.
Find one that says something positive about football and which follows the national line laid down by the media and everyone picks it up. But when they come up with something a bit unnerving, well, they’re just a bunch of Swiss academics. What do they know about English football?
So not surprisingly, CIES Report number 61, “What football fans think about the professional game” hasn’t had much coverage. Which is a shame because its findings are dynamite.
To give a flavour of it, the report tells us something about those completing the survey:
This survey was completed by 2,061 individuals which is very different from the daily pap that we are fed in which one journalist or ex-footballer being interviewed, gives an opinion on one player. It is also getting on for double the size of the interview base of most opinion polls.
But here it is not so much the depth of the research that is of interest here but the findings.
Too many internationals
61.5% of respondents said, “There are too many matches between national teams.” It is an issue that Untold has been commenting on since we started publishing in January 2008. I very rarely hear any mention of this on radio or TV, where they pump out internationals as if they were a good thing (which of course they have to say since they have just paid lots of money for the rights to broadcast).
Untold has always been with the majority here, because it is the clubs that pay the wages, but also pick up the tab when the player comes back injured. Mr Wenger’s comment that international managers are like car thieves who nick your car, wreck it, and give it back to you saying “get it fixed ready for the next game” was accurate then, and is accurate now. He got a massive fine for that one.
53.2% said “There are too many refereeing errors”.
Not that there are “some” errors but “too many”. Have you heard that replicated in TV or radio reports? I haven’t – because in the English media beyond the occasional, “I think the ref got that one wrong,” you don’t hear comments about referees at all. The days when Alan Greene could say on the BBC, “The referee was absolutely appalling,” are just a dim and distant memory. He was initially banned from saying that, and then dropped all together.
PGMO, the referees’ organisation, creates the rules of what may be said, rather like some tin-pot dictator lording it over his cowed poverty-stricken people.
So let’s remember this, the majority think there are too many referee errors. The PGMO who run refereeing in the PL say they are over 98% accurate. Just the fact of that difference of opinion is worthy of debate even without considering which side is right.
30.3% said, “There is too much match fixing.” And this is never ever mentioned in the UK media even as a possibility.
Now I am not saying there is match fixing, and as I always try to say, I have proof of match fixing, but I am very dubious about PGMO because of the way it runs refereeing the PL, which as I have so often pointed out, is very different from the way refereeing is run in other countries (one ref overseeing the same team time and again, no geographical balance among refs etc etc).
And now I can see it wasn’t just me. Here we have fractionally under one third of respondents saying there is not just some match fixing but, “too much match fixing.”
And this is consistently my point; not that we should take it that there is match fixing, but it should be a topic we can discuss – but we can’t, because the media won’t even mention it when it turns up in a survey from an organisation they often report.
The conclusion therefore is….
The dominant factor in terms of everything you read and hear from the media relating to football is “does it fit our vision of what football is?” And if the answer is no, it is not reported.
The Arsenal video collection: free
140 historic Arsenal videos with more being added each day
Gaslighting: how refereeing in the Premier League is manipulated, and why the media never speak about it.
- 1: Are the referees and the media really out to get Arsenal, or am I just imagining it?
- 2: How discussions about refereeing are deliberately stifled by the media
- 3: Referees: the odd statistics that are simply never revealed or discussed
- 4: How we have been utterly misled about football: part 4
- 5: Hiding the problem of refereeing is destroying the credibility of the Premier League
- 6: Revealed: PL referees are not 98% accurate but actually just 75% accurate
Does the report say the countries from which the opinions were obtained, Tony, and/or give any breakdown for this?
Tony
Re Match Fixing.
Paragraph 2 you say: “…I have proof of match fixing…”
I think this a typo.
Apologies if not.
Otherwise spot on as usual. Always nice to know you’re not a lone voice.
two thousand and we dont know where from -seems hardly representative and who is a football fan and who are these people who produced the report
I haven’t read the report but haven’t you just done what you accuse the mainstream media of doing – picking key points from the report to support “what you’ve been saying all along”?
ron
Fair enough questions but I think the point being made is that the media will take the views of 2 pissed vagrants in the park and call it a survey if it suits their agenda.
Where as a ‘scientific survey’ of 2000 people, no matter where they’re from, will be completely ignored if it doesn’t. I think that’s the point being made.
wikipaedia says it mainly looks at transfer values and team profile
i did look at the report and to be fair most of it is pretty standard stuff -people dislike the playacting etc of players , think there is too many internationals , would prefer more diversity in leagues , 69.7% think there is not too much match fixing-so as there is nothing particularly revealing this might explain the media not taking much notice. it does have a positive view of womens football which might have appealed to the media .
ron
Again I think you’re missing the point.
You say “there is nothing particularly revealing this may explain the media not taking much notice l” but that isn’t the point and in actuality unlikely to be the reason.
As I said if it was 2 tramps in the park slagging off our back 4 they’d take an interest simply on the basis it’s slagging off Arsenal.
If it was 2000 people saying how great our referees are and how the FA are doing a wonderful job I’m pretty sure they’d take notice, and they wouldn’t give a toss where they came from.
@ Ron
If your population size was 10 million a survey of 1,100 would give you results which had a margin of error of +/- 3%. Even a survey of just 100 people would still give a margin of error of just +/- 10%. So a sample of 2,000 is significant.
Without knowing the cohort in question it is, of course, impossible to know what level of accuracy we are talking about. But when you consider that the large the population involved, the smaller the percentage of that population that is needed for the same level of accuracy. For example, if you were looking at a population of just 1,000 you would need a sample of over half (525) to achieve an accuracy level of +/- 3%.
Well Jim, I have to say I have tried to see it from your point of view, but really I can’t. The Observatory conduct loads of surveys, and the media pick out certain ones to maintain their specific view – for example that international football is wonderful and much loved.
I have pointed out that here is a report from an organisation that the media does like to quote which says that a lot of people don’t like international matches and it has had not commentary in the press.
Hence I report what the media in general do, and then what the Observatory does. It seems straightforward to me. But maybe I’m just being blinkered. I don’t think so, but I’d be silly to deny that was possible.
Nitram, your ability to read my mind is getting worrying. Can you shut down your telepathic powers after 9pm please.
Mikey, I think there is also the secondary point, that the media will quote the Observatory’s figures (and thus clearly approve of their numbers and their methodology) when the findings support the media’s vision of football, but not otherwise. If the number of replies is inadequate for this survey, then looking back at other surveys it is inadequate for them – but when the point suits the agenda, the media take it.
But I agree with you, this number of replies is perfectly reasonable. I’ve seen general election surveys of voting intention done on smaller numbers.
@. Tony
I think in the context of your article, it’s actually the primary point. I couldn’t agree more.
I can’t pretend I’m mathematically minded enough to explain the theory behind it but I understand there is a cut off point whereby the probability means that you don’t need to increase the sample size whether it’s based on a population of 1 billion or 100 billion. I believe that number is actually about 2,000. The random nature of the sample is obviously critical though. Unlike the media who are selective!
Mikey
As you say the random nature of the survey is critical, and as this survey was done by a credible scientific body I would be fairly confident it was.
Unlike the utterly predictable and non scientific way the media go to AFTV every time they do a survey on Arsenal, knowing full well they’ll get the critical responses they desire.
@ Nitram
Absolutely hilarious. I read your first sentence and immediately thought I would comment back about AFTV…..LMFAO!!!
They might as well do a survey on Tottenham High Road about us…………but the media would still publish it as a survey of “people from north London” 🤣🤣
I know.
We get all these dramatic headlines such as:
‘Arsenal fans shocked by potential transfer’
‘Fan poll suggests all is not well despite late comeback’
Our source says big player ready to leave over training ground bust up’
Then when you look behind the headlines you discover the ‘fans’ they’ve spoken to are 2 AFCTV numbnuts on a weekend bender in Brighton, and the ‘source’ is Stewart Robson, who’s full time job since being sacked by Arsenal is in fact slagging off Arsenal.
Still, much more credible than a bonified scientific study of 2000 people don’t you know.
they invited over 27,000 people ,their subscribers, to take part from many countries and in the end only just over 2,000 replies were validated and used -largest country responding was france – so it ends up being a pretty small survey – so it does not seem all that relevant to the average football supporter in the uk – they had not done this type of survey before so i imagine they will be looking for improvements. the report does however conclude that although there is need for changes to the game faith in the beautiful game remains intact. sounds the sort of thing you say the media only print
ron
It’s been explained to you in detail how a 2000 sample is not a small survey.
From
https://www.surveymonkey.co.uk/curiosity/how-many-people-do-i-need-to-take-my-survey/
Respondents Needed at Error of ±3%, ±5% and ±10%
Population first then respondents required for relevant % error as indicated above.
10,000—–1,000—-385—-100
100,000—-1,100—-400—-100
So 2000 responders from 27,000 people is actually way above the amount needed for a + or – 3% error. In other words it is not a ‘small’ sample at all, but actually it is a massive sample being as it is twice as many as required to achieve the scientifically accepted levels error.