Posts Tagged odds
Book%
Posted by Peter Webb in Bet Angel on October 13th, 2011
I fielded a question from somebody recently about the book% and what information does that contain?
I’ve seen quite a few theories about what the book% can actually tell you, but here is what it tells me.
At it’s most basic level the book percentage is merely a broad measure of the market. A big number of selections will equate to a big book% and small field a small one. This is due to the spread between back and lay prices. For example two selections priced at 1.99 would create a book of 100.5% on the back side and 100% on the lay side. If you expanded the selections to ten runners priced at 10′s, the best book you could get on the lay side would be 95.2%.
Early on in a market the book% will be higher than near the start of an event and this contraction is caused by more and more competitive prices being offered to the market. This process produces a subtle drift in prices on the back side of the market.
Back in the early days the book% was higher than it is now. Overtime it has shrunk and that has meant that outright backing or laying strategies have become a lot more viable. But whatever you do, you can’t avoid giving something to the house in the form of commission. However, in highly variable markets its quite possible to get a back or lay price matched above or below the VWAP price. This makes backing or laying for value quite possible. A side effect of the shrinking of the book% over the years is that less money has been lost and this, I suspect, is one of the reasons that Betfair have felt uncomfortable with the lack of losers. But it’s an exchanges biggest USP as well.
From a trading perspective the book% doesn’t hold much value for me as it is a re-active measure. When the price on a highly liquid runner is shortening, this will increase the % and when it is drifting it will decrease it. You could argue it pinpoints the turn of a trend but that is very reliant on what is happening elsewhere in the book. You may as well look at a chart at that point!
Book% isn't an open book
Cheaters never prosper
Posted by Peter Webb in Bet Angel on September 24th, 2011
I saw this post on another blog recently and it shows up how statistics can actually reveal very little at all. This is obviously unintended in this case, but is a trap a lot of people fall into very often. Especially when gambling / trading.
http://www.soccerbythenumbers.com/2011/09/cheaters-never-prosper-connection.html
I first looked into the relationship with fouls and goals some time ago and the link is the attacking strength of one team over the defence of another. If your defence is under pressure then they are more likely to commit a foul. It’s as simple as that, there is little more to be read into things.
For some time I sought to understand how to price odds on a football match. This was the obvious key to finding value. Bit by bit I improved my forecasting skills to the nth degree. In that quest I discovered many things, but one of them was how a lot of stats that are proffered to the market are in fact completely useless in determining value.
If a team has scored two goals a match for ten matches in a row, then you may think this has some bearing on the next match. But ultimately it doesn’t. But that doesn’t stop every pundit and his son offering this as a useful stat. In fact the source of true odds actually lies much deeper in individual statistics and how you plug two disparate teams together. Sort that out and you have the key to successful forecasting.
Feast or farce
Posted by Peter Webb in Bet Angel on June 27th, 2011
Not difficult to choose what to do this afternoon. Yes there was some racing on but when end up with 15 minute gaps and one race at Pontefract had just two runners, though curiously this was the best race of the day for me.
Over at Wimbledon there were quite a few decent matches and opportunties so it was this that held the main interest for me today. As a point of interest, if you want evidence that people are using models like Tennis Trader, have a look at the following implied probability graph. The odds map perfectly onto the the points won or lost on serve, you can clearly see the ranges.
There are 351 tick increments on Betfair, an average of 36.7 games at a Wimbledon match and hundreds of points in a match. If you can pinpoint where the odds should without an aid given those variables, you are a better man than I!
Crianza
Posted by Peter Webb in Bet Angel on June 19th, 2010
I’ve had a few beers on Crianza this evening after it romped home at Redcar today. Nothing remarkable about a horse winning a race, but Crianza came in at odds of 1000 and I had money on.
It’s the second time I have managed this, the last time was on Bermondsey Bob at Salisbury in August 2008. Therefore I assume the mean return time of this sort of thing happening is around once every two years. The funny thing was I didn’t even realise at first as I was already looking at the next race and hadn’t paid any attention to the finish of the previous race. It was only when one of the team piped up all excited that I realised what had happened. A major bonus on what has been a very tricky and difficult week.
I am already looking forward to the next one in 2012!