[GoldenBootIRL]

I have a question about the Kelly Criterion and how it relates to current bank betting strategy.

I would appreciate it if someone would double check my maths.

----------


There are 2 bets: 4/6 and 10/1.

You have an edge in both bets of 20%.

----------

20% edge on 4/6 bet: win 72% of the time.

20% edge on 10/1 bet: win 10.9% of the time.

----------


The Kelly Criterion states you should bet:

30% of your current bankroll on the 4/6 bet (Your bankroll will grow, on average, by about 3.14% on each bet).

1.99% of your bankroll on the 10/1 bet (Your bankroll will grow, on average, by about 0.19% on each bet).


----------

Most professional punters do not achieve better than 20% ROI (+EV) on their bets on the long run so this is an optimistic forecast but we can just focus on the theory for now.


When someone tips they get more credit for tipping the 10/1 shot but if good bankroll management skills are used the 4/6 shot should produce more profit over the course of 100 bets even though the expected ROI (edge) is 20% on both bets.


----------

I was asking in a different thread about expected losing streaks for each of the bets above over the course of 100 or 1000 bets to get a better idea of optimal current bank betting strategy or at least to be able to understand how to adjust downwards the Kelly Criterion to fit in with an individuals tolerance of risk as the 4/6 shot can be expected to encounter a losing streak of 5 bets in a row which would result in a loss of 83% of the bankroll.

----------

When deciding how much of your bankroll to stake using a set staking plan based on the EV of your bet, what staking plan do you use and why do you feel this suits your risk/reward comfort level?

When using a current bank staking strategy instead of a level stakes betting strategy do you feel than if two bets have the same expected ROI in a vacuum, then the bet that should win more frequently will be the more profitable bet over 1000 bets whereas in a level stakes betting strategy each bets will simply provide the same ROI and profit?





Thanks.

[GoldenBootIRL]

In a nutshell what I'm asking is:





Using a full kelly criterion current bank betting strategy...


assuming a 20% ROI on all bets placed...


do shorter priced/more frequent winning bets have a better return over 1000 bets than bets with larger odds and proportionate winning frequencies.

[u21c3f6]

In a nutshell what I'm asking is:





Using a full kelly criterion current bank betting strategy...


assuming a 20% ROI on all bets placed...


do shorter priced/more frequent winning bets have a better return over 1000 bets than bets with larger odds and proportionate winning frequencies.

Yes, with the same ROI, using Kelly, the growth rate is greater for wagers with the greater win rate (which you indicated in your first post).

Using your figures there is no question that one should concentrate on the 4/6 72% rather than the 10/1 10.9%. Assuming that one was capable of those results and starting with a $1,000 bankroll, after 1,000 wagers you would have the following bankrolls:

10/1 10.9% $6,504.41 (a very nice increase) and
.
.
.
.
.
.
.
.
.
.
.
4/6 72% $43,445,855,987,131,100

Just a small difference.

Joe.

[kpoutlaw]

Well, I can't give you a mathematical answer to your question, but as a newbie to the Kelly betting system, I'll tell you the results I've had over the past 3 months. I live in Korea, where the betting sites are all illegal, highly juiced, and rarely move their lines. Some sites even deliberately manipulate the odds for two opposing teams in soccer in order to trick bettors to bet the other side! For example, Manchested United is 1.4 Westbrom 3.2 ..........so they make Manchester United 2.3 Westbrom 1.7, and once they entice all the big bettors to pound MU and MU wins, they kick you off and jack your money! lol (I found a good way to take advantage of their deception is to bet only like 20-30 dollars, and they won't jack you even if you win.)

Also, they will lower the odds on highly juiced favorites for soccer and give you really high odds on the underdogs..higher than what other foreign sites offer, games where MU is like 1.2 and the other team is like 8.0.....so obviously, you have a distinct edge betting the underdog if they give you such high odds, but how often do these games hit? Maybe I haven't bet long enough with the kelly to give you the answer you want and I may be wrong, but over the course of at least 300 bets using the kelly, I have rarely seen those 10 to 1 shots hit and they simply aren't worth waiting for to hit. I don't care how you justify it mathematically, but the former rather than the latter have always had a better return for me. Bet big on the more frequent winning bets with an edge and yield a much better return in a short time! Yes, there is the possibility of getting your bankroll wiped out should you hit a bad streak, but by starting with a bankroll of $500 and fully kelly betting simply by finding a weak line on the korean sportsbooks and comparing it to the true winning probability or fair odds on the foreign books, I have seen my bankroll shoot up in no time. But then again, I do this knowing that if I lose $500, I won't be devastated, and also, I know that I do in fact have a distinct edge with pinpoint accuracy because all I am doing is exploiting weak lines that I compare to lines at 10 different foreign sites instead of actually having to think of what the true probability of winning might in fact be.

I guess once the Korean sites figure out what I'm doing and start to jack my money, I'll have to turn to the foreign sites and figure out this stuff too, but for now, it's all gravy when everyone else is doing the math for you and all you gotta do is crunch in the numbers and full kelly that ass...lol