By Tony Attwood
“Lies, damn lies and statistics,” is, I guess, the ultimate anti-analytic statement. The total way of saying, “I know what I think, and you can’t tell me no different.” The end of detailed examination as a way of understanding the world.
So trying to analyse what happens over time to a football club, how it can go through a run of games without scoring, and what this is likely to mean for the future is bound to bring more of the same in terms of response. Arguing against the “lies and damn lies” group gets “clutching at straws” as a response, and so it goes round and round.
And yet Untold deliberately has the name “Untold” because it covers issues and undertakes analyses that seemingly few if any other sources of information and comment on football are willing to take. Even when they are fairly obvious. We’re here because the media tell us that Arsenal have the worst injury record of any club. We’re here because most of the rest of football commentary is about simplicity.
By way of example take this run of results
- Arsenal 0 WHU 0
- Tottenham 0 Arsenal 0
- Arsenal 0 Sunderland 0
- Arsenal 0 Fulham 0
That was four successive league games in 2008/9. So what happened next? Or in fact for the purposes of the lies and damn lies brigade what do you think was most likely to have happened next? (The answer is at the end).
There were the usual comments about Arsenal were lucky to get nought on radio and TV and in the press, and oh how they loved it, not least because the games before that run were a 1-1 draw and another 0-0, away in the Cup at Cardiff.
Doom and gloom were in the air, this was definitively the year we would finish outside the top four. We would never win anything again under Wenger. Tottenham would of course finish above us etc etc.
The fact is that evolution has made our brains irrational (if you want to know exactly how and why in detail read New Scientist, 12 December 2015, page 31 onwards.) We are irrational because being irrational helps us make quick decisions with limited information, and difficult decisions with large amounts of information that we can’t process all at once.
So for primitive man, the unexpected sound made him jump for cover, and then made his companions laugh at him when it turns out not to be something that is liable to eat him, but nothing at all (although actually it was an earth tremor 150 miles down and 200 miles east).
So for the primitive football supporter looking at the 1000 events that happen in 90 minutes of football, a suggestion that anything other than “more of the same” will occur gets sniggered at, and out come the simplistic sayings once again.
Today we need more information to live far more complex lives, so we use statistics to draw information from large amounts of data. The “get rid of Wenger now” approach to a match that has not gone right, is, in essence a Stone Age approach from people who benefit from the statistical analyses that goes into each and every medicine they take for each illness they get but don’t realise that their cure is based on medicine which is based to a large degree on statistics. Except they probably decide not to finish the course of tablets like it says on the packet because they feel better and anyway they just say that on the box to sell more pills.
And anyway, the Stone Age ended because mankind ran out of stones. Everyone knows that.
The reality in football as anywhere else is there is no such thing as certainty, but there are indicators. And the last time I looked Mr Wenger’s record at Arsenal was managed 1101 competitive games, won 633, drawn 255, lost 213. That is a percentage win rate of 57.5% across a huge number of games. Far better than any other permanent Arsenal manager. Better than Herbert Chapman with 49.88%. Better than George Graham who had 48.91%.
Oh but you can prove anything with stats – Wenger was lucky to inherit that back five. Except that back five delivered unto Arsenal league finishes of 4th, 5th, 10th and 12th in the four seasons before Mr Wenger took over. What he did was take what had become a rather ordinary despondent and failing back five, and made them Double winners.
Now there is a point in all this, and the point is “what happens next?” and “what is likely to happen next?”
Our win percentage in the league this season is 54.16%, below the norm – and we have failed to win any of our last four games. So on the basis of what normally happens under Mr Wenger the most obvious prediction to make is not that we will have more of the same, but that we will recover. That, based on the past 1101 games is the most likely outcome from this position.
But of course for those of a negative disposition, who maybe learned algebra, trig and geometry at school, and who like to believe in clear fixed answers, we won’t recover. But I am sorry to say that the maths they taught in schools that is not maths at all. Because maths, as they learned it, was all about certainty. A triangle has 180 degrees in its three corners – FACT.
But actually maths isn’t about fact – it is about making sense out of chaos, and looking at the most likely outcomes, which is why we have statistics – those numbers that people who don’t understand them decry so vigorously.
Of course I would have been much happier this morning, watching the sea batter the Norfolk coast if we had won these recent games that we have drawn, and were still top of the league. Because being top of the league is nice. It gives a warming feeling when walking along the beach in 60mph winds. But I know from the history of Mr Wenger that after a run like this comes the victories.
It is the teams who are on the long run of wins who need to worry. Last time Tottenham had a couple of high scoring matches (beating Bournemouth 5-1 and Villa 3-1) they then won one of the next five. Last time Man City won two in a row they won three of their next eight. Last time Leicester won three in a row they then won one of the next five.
Of course you can prove anything with statistics. So, next time you are feeling a trifle unwell, best not go to the doctors and then take the pills you are prescribed. There will probably be a 95% chance that they will actually remove the infection, make you feel a lot better and (if you take the course) stop you getting it again. But no, just say, “you can prove anything with statistics”, so better just to be ill. The infection might then take a deep hold on your body, and could be with you for years to come, but that’s a small price to pay for the chance to repeat a meaningless mantra.
(After that run in 2008/9 we won seven of the next eight).
- Arsenal – Southampton 0-0: another example that it doesn’t even out
- Arsenal managers – the complete analysis
- 3 February 2001 Dennis Bergkamp scored the only goal at Coventry City to give Arsenal’s first away win since November. It was a moment the footballing world changed. Arsenal lost seven times that season but within a year Arsène Wenger was talking about an Unbeaten Season.
- 3 February 2004: Middlesbrough 2 Arsenal 1 (league cup) – despite the brilliant league run Arsenal went out. Keown was sent off on 45 minutes, Reyes scored an own goal after Edu had pulled it back to 1-1.