For the last three years the sports stats guys OPTA Pro have organised a football analytics conference in London in the spring. It’s mostly for bloggers to present new analytics ideas to the professional football clubs in attendance. I thought it would be cool to present something, although I wasn’t entirely sure my employers would have been happy for me to share any of our stuff so publicly, but I sent off an abstract submission on spec anyway.
As it happened, there was never any chance of an issue with proprietary knowledge as my email to OPTA got lost in the ether – because I’d used an apostrophe in the title of the word doc I sent (no kidding). My mail client didn’t alert me it hadn’t been delivered it or anything, it just sent it to some email elephant’s graveyard and left it at that. There is some irony, I guess, in the idea that I was proposing to tell football clubs how they could run their clubs more efficiently using technology, and I couldn’t even manage to send an email properly.
Never mind, it probably wouldn’t have been chosen anyway. I’m not sure OPTA would have been too keen on a ‘betting guy’ getting a spot (although they had the excellent Ted Knutson of Smartodds, and formerly of Pinnacle Sports as a judge of the abstracts). My scepticism about how well football clubs have used analytics to date probably wouldn’t exactly endear me to OPTA or their audience, so there’s probably a good chance I’d have been hounded off stage under a hail of rotten fruit in any case.
I thought I’d post the abstract here anyway in case it sounded of interest to anyone……
Title: Using the Asian betting market to rate football teams & players (is ‘the market’ the best football judge in the room?).
Can the ‘opinion’ of the betting market be used by professional football clubs for practical uses?
- Interpret the Asian betting markets to extract objective ratings for football teams & players.
- Combine an interpretation of betting market data with OPTA data to chart a specific performance pattern; namely the decline in recent years of English Premier League teams in the UEFA Champions League.
A. Demonstrate a practical way in which ratings can be extracted from betting market data.
Show how they are objective and accurate – they benefit from the ‘wisdom of crowds’ principle – and can be used to answer questions like….
- Where does the market rate our team in our league?
- Does the market rate us as an improving/declining team?
- How does the market rate our upcoming opponents?
- How does the market rate our chances in the 2nd half here?
- How valuable is player X to his team (Goals/Points/£Value)?
- Does the market think we’ve improved for our new signings?
- Does the market like our new manager?
B. England has only had 3 Champions League semi-finalists in the last 6 seasons. They had 11 in the 6 seasons prior to that. By using data from Asian closing lines we can get the market’s opinion; have English teams just been unlucky? Are their results simply reverting to a mean? Or does the market infer a genuine decline in EPL teams relative to other leagues, and the other 4 major European leagues in particular?
Does any Asian betting market trend match/follow/anticipate the underlying (OPTA) performance data of English teams in the Champions League?
Assuming the betting market shows a real downward trend – which would go against the greater EPL investment in playing talent relative to other leagues – propose a theory to explain;
Why are England’s top teams getting worse?