Been thinking of this topic off and on for the last few days.
I think's it's helped me calrify my own approach to punting and with a little tweaking using some of the concepts Muze discusses I think I could be happy with my staking next season.
For me I want to get a mix of probablitly to win / value in the odds.
By now everyone is probably aware of the much quoted "If I gave you 2-1 for a coin flip how much would you bet?" scenario whereby the bet can only be made once and must be made with a very large portion of your bank.
So, for me following on from this my gambling concept differs slightly from this precept:
"My stake size depends on the % value difference im getting between my price and the bookie price."
Now, the perecntage unit is an importnat point. It may just be a phscological issue with me (and I suspect it is) but lets say my bank is 50 pts and my maximum bet is 15 pts.
I price two games up,
Tits @ 1.8
Ass @ 3.0
The bookmakers open up Tits @ 2.0 and Ass at 4.0. Would it be correct to assume that you'd have more stake on Ass Muze as this provides a variance of 8%, as opposed to Tits which offers a 5.6% variance?
In thinking about this assumption in order for it to be successful I think you'd need to find 3 other bets of equal or higher probability. I suppose that this is easy enough but the frequency might not be high (i.e. 6 bets per season) and your bank may have diminished by this time due to a few loosers. In short if you have a stand alone bank and are mixing the variance percentages you are gambling on you may not have enough to cover the loosers.
Also there's the amount of time you're actually taken to turn over your inital investment. For example a few winners is able to provide a 30 to 40 pt cushion on your maximum weekly bets and this amount is able to be invested in other areas (shares / outright bets / odds arbitrage, etc,etc). Also for me confidence comes into betting and sometimes I ignore matches when my bank is low as I've just plain bottled it.
I guess the point I'm making is that I first try to identify games that I believe have a probability of >50% and then look to see what the bookies are offering.
I believe that this is quite common sub concious thing. It's easy to scan the coupon and look for evens / odds againsts and then to think if it's better than a 50/50. In non-league punting in particular the speed of punting after the prices have been published is not to be discounted.
For example, Leek were 6/5 when they came out on Sunday at
Bet365 against Burscough. Unfortunately I was in the middle of doing something else and had not yet read that Leek had recently made some shrewd aqusitions to their squad. Within a 24 hour period the game is now priced at 5/6 I think.
Again, I suppose the time issue with implementing a system based on percentages could be improved with a spreadsheet but I still think this still take awhile if you took the same time over every game (or maybe you use a bit of "gut feel", selectively price a dozen and see what the prices are?).
Apologies if this seems a bit rambling in places as I'm half thinking in type. I'm fairly certain of the way I want to go now.
With
Paddy Power joining the non-league game now (I think the number of online books covering these matches have doubled this season from 3 to 6 or so) then they'll be more chances I'm sure to catch them with their pants down.
If anybody has any comments that I coulod do with considering / factoring it I'd love to hear them.
AMP