Forums

General Betting

Welcome to Live View – Take the tour to learn more
Start Tour
There is currently 1 person viewing this thread.
prestbury road
09 Jan 12 15:55
Joined:
Date Joined: 06 Feb 06
| Topic/replies: 280 | Blogger: prestbury road's blog
Can anyone recommend a source for learning the mathematical theory behind price derivation for major sports?

The source can be a text or a website.

Many thanks in advance
Pause Switch to Standard View Mathematical Derivation
Show More
Loading...
Report Eat_your_greens January 9, 2012 6:04 PM GMT
I am subscribed to patrickMT's, YouTube Channel. His video's have helped me brush up on my maths skills. Although it's not sports related, he explains maths very succinctly.

http://www.youtube.com/user/patrickJMT?feature=g-all-u
Report DivideByZeroError January 9, 2012 6:35 PM GMT
Probably best to brush up on your measure theory first, I can recommend Noel Vaillant's site:

probability.net
Report kenilworth January 10, 2012 11:44 PM GMT
It comes with experience.
Report henok January 11, 2012 11:30 AM GMT
most of the mathematics associated with football markets is simple mathematics. if u had good probability and arithmetic skills u will find how
Report prestbury road January 11, 2012 1:16 PM GMT
Gentlemen, thanks for the replies.

Greens/Divide, thanks for the advice, I will check out these sources

Kenilworth, I agree experience helps, but I want a reliable source for the theory.  If you dont have the theory, it will take a hell of a lot longer to figure this out!

Henok, I have excellent mathematics skills but I would like to know for example, which distribution curve should be used for football and which for basketball? 

I can derive tennis prices, correct score markets for other sports etc, not such great  issues when it comes to sports or markets with lots of definite increments (points in tennis, frames in snooker) and I can use stats and a bit of my own nouse (sp?) to create the base variables to begin with to do this, but I would like to know the exact models to use for each sport to be able to derive prices both pre-live and live. 

If a football team starts at 1.60 and is 1-0 up at half time I know that their price will be roughly 1.17 - 1.20, but this is just from experience and a decent understanding of prices, I would like to know how the price arrives here only using supremacy and expectancy values.

Any further help would be very welcome.
Report Contrarian January 11, 2012 1:41 PM GMT
but I would like to know the exact models to use for each sport to be able to derive prices both pre-live and live.

Well, there is not necessarily a great deal of consensus on what the correct model is for each sport.

Eg. in football, some argue for a (non-homogenous) Poisson distribution, and others for a negative binomial.
Report prestbury road January 11, 2012 3:50 PM GMT
Contrarian (aptly named!), you are absolutely right, but I would like to find the models for each sport that there is most concensus about, or even any models that are applicable that I could take an opinion between myself. 

For example, where would I learn the negative binomial and the Poisson?  Have you any good sources?
Report dunlaying January 12, 2012 9:53 AM GMT
For the Poisson distribution there is a very good and concise explanation in
FIFTY CHALLENGING PROBLEMS IN PROBABILITY,F Mosteller,Dover Books,available from Amazon for about £5.It is to be found in the solution to the Moldy Gelatin(sic) problem.
The Poisson Process is dealt with very well by Prof.Henk Tijms in
UNDERSTANDING PROBABILITY,Cambridge Press.(Be wary of his answers as he is a little slap dash, to say the least.)
You will also find the negative binomial distribution dealt with in Tijm's book.
..........................................................................
PROBABILITY MODELS ,Prof.J Haigh,Springer Undergraduate Mathematics Series,is very good too.He covers most things in this book and he is,as far as I can tell,accurate.Buy it from America it is cheaper over there,a lot cheaper.
...........................................................................
Report prestbury road January 23, 2012 5:26 PM GMT
dunlaying:  thank you very much for posting, this is exactly what I was looking for
Report JPL66 January 23, 2012 7:35 PM GMT
By now there's a big academic literature on modelling football scores, but one of the earliest papers that kicked it all off (and explains the basic idea of Poisson regression) is 'Modelling Association Football Scores' by M.J. Maher (1982).  Google Scholar will turn up a pdf.
Post Your Reply
<CTRL+Enter> to submit
Please login to post a reply.

Wonder

Instance ID: 13539
www.betfair.com