The same goes for betting on doubles. They may be attractive because of the combined odds. But if you double up 2 evens shots are you really saying the second outcome is worth twice as much as the first? What if you back a 5/1 shot and then a 3/1 shot in a double. Does the 3/1 shot really warrant a stake 5 times the size of the 5/1 shot?

Also if the first leg of the double loses did you really want nothing on the second?. If so what the f*** where you doing including it in the first place?

c) Price Related Variable Unit Plans. 
This is a plan where the stakes are escalated according to the odds with the aim of achieving a roughly similar return with each winner. E.g 3 units on at evs or below - 2 units 5/4 - 2.1, 1 unit 3/1 above. This isn't a bad technique as it requires discipline and also recognises that all bets aren't equal. However by focussing on the end result your are applying rigid rules that do not necessarily reflect the true value of the bet. E.g the 2/1 shot may be an even better bet than the evs shot (in terms of pure chance) yet you are having twice as much on the evs shot.

d) Confidence Related Variable Unit Plans 
Similar to the above but you're using a gut feel to set your stakes. Has the benefit that perhaps you are loading according to the true worth of the bet, but who really knows given you are applying no discipline or technique to the setting of this. However if you have a good gut feel it may be racking them up for you. But wouldn't it be better actually understanding the decision and having method behind it.

e) Value Geared Bets
This is the method I am currently looking into. I have come to believe there should be one guiding principle when it comes to staking. The more the odds are in your favour the more you should have on. Basically this means if you back a 2/1 shot whose true odds should be 4/5 you should have more on that a 2/1 shot whose true odds should be 7/4. Although both good sound bets (we are assuming your measure of the true odds is accurate) surely a level staking approach doesn't reflect the exciting opportunity of the 4/5 shot priced at 2/1.

To use the blackjack example. You can be the worlds best card counter but won't make any money UNLESS you are able to increase the size of your bet when the deck is in your favour (i.e. on a high count). This is effectively levering value / striking when the %'s are in your favour and is entirely independent of the result of the last hand (count aside). This is how you turn the %'s in your favour.

I think it should be the same in other sports. The disadvantage of this method is, of course, that if you can't accurately assess true odds you're in for a nasty shafting. But if you can't do that I hope you're only betting for fun anyway!

So here's a suggested appraoch which brings together the value element and the weighted staking element. It requires a comparison of available odds with true odds to calculate the bet strength. 

Determining this is very personal. One way would be start by calculating the difference in true odds (this is subjective) over available odds as a % (a % is more accurate than points as it takes into account the length of the odds i.e 2/1 for an evs 1/1 is a bigger "rick than 10/1 for an 8/1).

E.g. Lets say for me to place a maximum bet I would require a 350% "value" premium. So if the "true odds" (determined by my form analysis and pricing up of the market) of my selection was evens for me to place a maximum bet the odds offered would have to be 7/2. My maximum is 10 pts, my minimum is 1pt. My bet strike is a premium of 125%. So on an evs shot I'm only interested in betting at 5/4 or better (otherwise I don't deem it value). I would then have a sliding stake scale according to bet strength (defined by true odds vs available). So I may come up with the following:

True odds EVS.
Avail Odds Stake
5/4 1 (this is min bet as it is at 25% premium)
6/4 2
7/4 3
2/1 4
5/2 6 
3/1 8 
7/2 10 (a once a year, monster price rick)

Here the stake units would be within my comfort zone. If my minimum bet is £100 and my maximum is £1000 then each unit - £100. Obviously this maximum bet should be placed in the context of the bank. It's ludicrous to have £1000 on a selection when your bank is £2000.

In that case it would make sense to increase your bank, decrease the unit range (say 3 point not 10 point) or bet smaller amounts. For me 10 pts is far too volatile but I think a range up to about 4 is a realistic starting point. E.g. £50 min, £200 max. (2 pts = £100, 3pts = £150).

Anyone any opinions on any of these methods? 
Do you think the last technique would work?
Are there any others that people operate?

Irrespective of the above it would be useful if we could instigate a standardised site wide unit plan. It's very difficult to gauge confidence regarding bets when people give figures (which are relative) or pluck units out of the air (both of which I think I've been guilty of). Axe - any ideas?.

Also, I think spread betting requires a whole different article….

Andrew Moraghan
(Comments on this article to Moz via the forum)

Click here for the world famous Punters Forum