Generally Accepted Lay %'ages

[thunderfoot]

What are the generally accepted percentages to increase Bookmakers SP odds by, to calculate Betfair LAY odds? I think it's about 10% for odds up to about 5.00 (which is as far as I'm looking to go on a new 'system'), BUT how would you then apply that 10%? i.e. Bookmakers SP Odds = 4.00, would Betfair LAY odds be calculated by :- 4.00 +10% = 4.40 LAY Odds or ((4.00-1) +10%)+1 = 4.30 LAY Odds.

[maria]

Re: Generally Accepted Lay %'ages Hi Thunderfoot, I promise I don't mean to sound dismissive of your question, but it honestly doesn't matter whether you use 4.3 or 4.4 to represent a starting-price of 3/1, because whichever one you use is fictional, and you can't measure whether or not your system is a profitable one from theoretical calculations of "generally accepted percentages". You can do this only by consistently putting real bets on with real money at live exchange prices after an extensive period of paper-trading from observing live exchange prices without actually submitting bets. Trying to use these "approximate averages" is the precise reason why so many racing forums are packed with laying systems which make dramatic profits on paper yet nearly all mysteriously and unaccountably fail the very minute any reader tries to duplicate them with real money. Again, I promise I'm not trying to be impolite, but there just isn't a clearer way to make this point: real bets for real money (after extended observation of live exchange prices, of course) are the only way you can learn anything. I think it should surprise nobody that what's "generally accepted" leads to losses, when you look at the proportion of punters who make losses ... :eek

[thunderfoot]

Re: Generally Accepted Lay %'ages Cheers 'M', thought you might come up with a sensible and logical reply.;)

[itooam]

Re: Generally Accepted Lay %'ages I would agree with Maria - but don't forget you do have free access to betfair data if you can be bothered downloading it all - I had to buy a 300 gig hardisk to help store it in sqlserver. Probably will take 120+ gig to store in any reasonable state (though I downloaded "other" sport too). Anyway, that will give you 2 years to play with.

[itooam]

Re: Generally Accepted Lay %'ages I did ask the same question a while back on another forum: http://www.flatstats.co.uk/ppp/viewtopic.php?t=6439&postdays=0&postorder=asc&start=30 If that link gets removed go to flatstats messageboard and click on Systems Testing Forum then click "Using SP Odds for Forecasting" (page 2 at time of writing) some useful formulae in there. Note I tend to test a system for last 15 years with rp data and I have created a lookup table (rather) than a formula that has been created using BF SP odds versus rp SP odds. I know it isn't an exact science but as close as you will ever get... Once I am happy that it is producing consistent profit I then test it seperately with bf data.

[slapdash]

Re: Generally Accepted Lay %'ages I know maria has far more practical experience of laying horses than I do, but just to agree with her and maybe say the same thing she's saying from a more theoretical point of view: If you look at some "average" of how much difference there is between the lay odds you can get and SP, and use this to test a laying system, the danger is that you may just have a system that finds horses where the difference between exchange prices and SP is greater than the average, not one that finds profitable lays!

[itooam]

Re: Generally Accepted Lay %'ages This is the exact reason for using the betfair data for testing the final 2 years - though some would argue 2 years data is of little value (depends what type of systems you write - I prefer small yield but high turnover). Ultimately it could be argued that using the exact odds as given free from betfair is again subjective in its use as they can change on a second by second basis. With this in mind I would still be happier using the estimates than going in blind? But I understand both your views - I believe you are giving good advice - that of being aware of the pitfalls in using avg odds prices for testing purposes.

[DC]

Re: Generally Accepted Lay %'ages To give you some encouragement Thunderfoot, my Profitable Lay System was based on "assumed exchange odds". I hadn't put a penny down before it went live on here. It turned out ok. :ok

[thunderfoot]

Re: Generally Accepted Lay %'ages

...... you do have free access to betfair data if you can be bothered downloading it all

Yup!!! Been there -Done that!! For some reason Betfair has stopped providing the Datafiles from March (I think). DC, Thanks for the encouragement. 'Spose if we're testing LAY systems then it's always best to err on the side of caution and calculate greater loses than what might actually happen ;)

[itooam]

Re: Generally Accepted Lay %'ages The betfair data releases did stop for a while - but have now continued again. Though now they are downloadable weekly - so a bit of a pain to import if you haven't done it for a while! After reading the comments from Maria and Slapdash I ran a little test to see how Maria's system would have performed for her first 140 races using betfair starting prices. I was suprised to find that it was a hefty loss (whereas Maria had a nice profit). I then checked betfair prices (at Maria's post times) to ensure no trickery was going on. I should never have doubted you Maria :) - all was fine. This revelation emphasised the importance of getting on at the right odds. I wonder if the market becomes more "efficient" the closer to the off you get? I would be extremely interested if anyone has done any studies into this - or can point me in the right direction? (Sorry Maria I was trying to reverse engineer your system :$. Looks like you were using more than one system as I couldn't find any patterns to your selections!?)