Friday, 13 January 2012

Maffs and I - uneasy bedfellows

Cassini kindly pointed me in the direction of a post on his blog (here)  in which he discusses mathematically assessing a match using the aforementioned Poisson method. This I have read. Four times, now. I've also tried to make head or tail of the wiki article on the same subject.  Microsoft's help files on the Poisson function in Excel might just as well be written in ancient Mandarin for all the sense they made to me. This is in no way due to the quality of the writing in any of the pieces, by the way. I'm  sure that to the mathematically adept if all makes perfect sense.

Then it occurred to me that I don't NEED to understand the bloody thing, just to know how to use it! So a quick google search later saw me reading a page at Pinnacle's web site explaining one way to calculate the number of goals a team is likely to score in a match against a known opponent. A video on Youtube  then showed me how to use the Poisson function in Excel. A mixture of both techniques would enable me to price
up a Correct Score market for myself. Marvellous.

A little bit of data crunching of stats from the Premiership so far this season, downloaded from that wonderful stats site football data into a workable spreadsheet,  I entered my first Poisson calculation with baited breath.

Thinking it made sense to use real data to try to model a real match I've come up with the following for tomorrow's clash between WBA and Norwich :-

 A nil nil draw has a percentage chance of 5.27, equating to 18.98 decimal. With 0-0 priced at the time of writing at 14.5 to lay -  that looks a value lay to me.

My spreadsheet also calculated AU (or at least, it calculated 4-0) as having a Poisson of .15 - decimal 666.66. Therefore laying AU at 10.5 must be the value bet of the century!

So my intention therefore is to do a double headed lay of those two scores for £100 stakes each leaving a £1300 liability on 0-0 and £850 on AU. Any scores in between land me £200 less BF's commission.

You're more than welcome to join me, but if I were you I'd seriously ask myself whether the author is maybe dabbling in dark arts that he hasn't fully fathomed out yet.... your choice!

5 comments:

  1. Glad you're having fun with this Gun, I've been meaning to look at this myself. I can spot a dangerous mistake here though

    "My spreadsheet also calculated AU (or at least, it calculated 4-0)" You have calculated the probability of 4-0 here not >3 goals, the probability of 4-0 will be significantly higher than over 3 goals to any team!

    ReplyDelete
    Replies
    1. Sorry, it's late. Clearly you haven't put over 2k liability on a maths experiment! All the same it looks like fun.

      Delete
  2. AU is not the same as > 4 goals. It is a bet on either team (or both teams) to score 4 or more goals.

    ReplyDelete
  3. This strategy is actually being sold right now as Assured Soccer Profits.

    ReplyDelete
  4. Yes, is "Assured Soccer profits" and i have been long term profits with

    ReplyDelete