Specificity and Equity

For at least three and a half years now, I've cautioned that in-play betting is unlikely to be a long-term profitable exercise for the majority of home traders, and the esteemed expert Joseph Buchdahl had this to say on the subject recently:

Supporting another opinion of mine, that one needs to specialise in markets less followed, Joseph had this to say:

When I came of betting age, options for betting were very much limited. 

As an outsider to the sport, I long ago learned from an insider that the idea of winning long-term at horse racing, whether flat of jumps, was impossible, and decided greyhounds was likely to be similarly fixable and thus also to be avoided. 

Which pretty much left football, although back in the early 90s, there was nowhere near the level of interest in the game that there is now. As I wrote here in 2009:

Back in 1991-92, I used a pencil, a pocket calculator, reams of paper, hours of time and Elo ratings on British football to make a nice profit. Ladbrokes ultimately closed my credit account, which I was quite proud of at the time

It does surprise me that football almost totally dominates discussions on betting these days. Simple logic should suggest that the chances of one person being consistently better than 'the vast majority of others' are remote. 

There are lots of sports out there, with plenty of data, but heretofore apparently without the same amount of serious interest. If anyone might have taken an interest, it would of course be Joseph, but a recent Tweet supports my opinion:

Those who have read about my interest in the underdog in a few sports will perhaps know why, but my point here is that the competition you are up against in football means that it is one of the toughest sports at which to be long-term profitable.

Some of you might have noticed that while researching this post, I found Joseph's Tweets to be a rich vein of nuggets. On June 5th, he asked:

I liked the one reply to this:

Joseph didn't respond unfortunately, but apparently there are still people out there believing the unbelievable. "Idiots" might be a little harsh, but clearly these people aren't critical thinkers, logical in approach, or understand basic probability. 

The 60% / 2.8 reference was of course to Mel whose pablum has recently drawn some attention. When asked for his thoughts on my posts pointing out several inconsistencies in his claims received this reply:

So much for civil discourse. Blocked for asking a few questions? How convenient, but I can't say I'm hugely surprised by this tactic:

Also worth noting that a volatile temperament is not one well suited to the world of in-play trading.

As with most charlatans, Mel has his ill-informed followers, seemingly in thrall at the idea that someone can consistently hit 2.8 winners significantly more frequently than 35.7% of the time. I would be too, but remember the old adage - "If it seems too good to be true, it probably is".

To hopefully make it clear, although I suspect many people believe what they want to believe despite the evidence, full Kelly suggests that 37.8% of your bank would be optimal for an edge of the size Mel claims. 

Even if the win rate were 'only' 50%, the bet size would still be a huge 22.2%. 

At a pathetic 45% strike rate the optimal bet is still a large 14.4%, and at a feeble 40%, a not insignificant 6.7%.

It's only at 38.3% that we get around 4%, with half-Kelly making this the 2% that is generally accepted as a realistic stake. 

My final comment is to highlight Mel's curious use of the word 'equity', rather than 'bank'. This rang a bell with one reader, who pointed me to this well known 'equity curve simulator' product:

I'm sure it's just a coincidence, but as I am in the dark now, I shall have to rely on others to let me know of Mel's progress, unless I have another Twitter account of course... 

As Carl Sagan said, "extraordinary claims require extraordinary evidence". Winning close to 60% of the time at 2.8 is an extraordinary claim, while this is not extraordinary evidence: