Wednesday, May 19, 2010

Bullshit's coming home

Reasons to believe England will not win the World Cup (part 99). A bunch of geeks employed by a load of chancers predict they will.

Analysts Matthew Burgess and Marco Dion took data on Fifa rankings, previous football results and bookmakers' odds, then used them in a quantitative analysis model - or Quant Model - designed for assessing the value of stocks.

The JP Morgan pair freely admit that their predictions should be taken "with a pinch of salt", but viewed the World Cup as "an ideal opportunity to light-heartedly explain quantitative techniques and demystify the typical Quant framework".

Now I am not really qualified to discuss JP Morgan's quantative model in any great detail; it may well be an excellent predictive tool and not at all something that has filtered out the fact that England are a one-man team with no goalkeeper worth speaking of in order to pander to a the national propensity towards over-optimism for the sake of a bit of cheap publicity.

Then again, my instincts may well be justified.

A serious point lurks here, however: the belief that you can construct a mathematical model that can account for the vagaries and falibilities that govern human behaviour was one of the causes of the recent financial crisis. Whether some banks' quantitative models ignored, or chose to ignore, the fact that lending money to people who couldn't pay it back to buy overpriced houses was a bad idea doesn't matter in this context. It went wrong. So it will be with the football.

Besides, you just know the Germans are going to win.

Labels: , , ,


Anonymous Anonymous said...

Way back in an entry on Wednesday, September 26, 2007, you posted a raunchy nun picture.

Any clue on the origins of this fine piece of art? I note you had a far finer copy than the BBC.

7:54 am  
Blogger bill said...

I'm awfully sorry anon, but I can't now recall where I got that from. (The hard drive on which it was stored has since died on me, so I can't really trace it).

I believe it dates back to some thoroughly disreputable 18th century work. I might revisit this important area of historical research soon.

10:37 am  
Anonymous Anonymous said...

oh please do. i just obtained a perfectly awesome enormous copy of aforesaid thoroughly disreputable work. it has a very very vague signature that seems rather long & unreadable. i will keep checking back.

10:55 am  

Post a Comment

Links to this post:

Create a Link

<< Home