...adjusting at aprox. 10% +/- of my total bankroll. I read that betting this way reduces win rate by about 50%. Currently I am playing with about a 1% lifetime ROR, should I double my bets to try to compensate for this loss in win rate?
...adjusting at aprox. 10% +/- of my total bankroll. I read that betting this way reduces win rate by about 50%. Currently I am playing with about a 1% lifetime ROR, should I double my bets to try to compensate for this loss in win rate?
> ...adjusting at aprox. 10% +/- of my total
> bankroll.
Are you considering betting 10% of your total bankroll during a session or on one hand? I was unclear as to your question when you wrote it, so maybe others are as well. Could you clarify what you are asking with more precision or an example so we understand what it is you are after? Thanks.
To answer your question as *I* understood your question to be written: If you are considering betting 10% of your bankroll on ANY single hand of blackjack, you have seriously misunderstood some of the concepts of Kelly betting, optimal bet ramping and units, etc... As I understand Kelly betting, you should NEVER exceed 2% of your entire CURRENT bankroll on a single hand no matter how large of an advantage you perceive to exist. Even this is considered risky by some conservative players and they play "half-Kelly" where 1% of their bankroll is the maximum they will ever wager on any single hand. Theoretically, anytime you play less than full Kelly, you give up some "win rate" in order to decrease your RoR chances. Is this what you mean by reducing your win rate by 50%, reducing your bet by 50%? But then you contradict that with the last line in your post asking if you should increase your bet to compensate for the loss of win rate. If you increase your bet, you are actually increasing your win rate (provided it is in + counts) but also increasing your variance and your RoR.
Your question is very confusing to me, perhaps to others as well. Or maybe I'm just an idiot that doesn't understand what problem you are trying to solve (which has a strong possibility). As I started out this reply: could you clarify your question for me (us)? Thanks.
> ...adjusting at aprox. 10% +/- of my total
> bankroll. I read that betting this way
> reduces win rate by about 50%.
Where did you read that?
> Currently I
> am playing with about a 1% lifetime ROR,
Seems fairly safe. :-)
> should I double my bets to try to compensate
> for this loss in win rate?
No, certainly not. But why do you think you've lost 50% of your win rate?
Don
> ...adjusting at aprox. 10% +/- of my total
> bankroll. I read that betting this way
> reduces win rate by about 50%. Currently I
> am playing with about a 1% lifetime ROR,
> should I double my bets to try to compensate
> for this loss in win rate?
I meant re-sizing my bets when I win or lose 10% of total bankroll ( as opposed to re-sizing after every session or "cutting to half stakes when half bankroll lost". )
I read the "re-sizing before each bet" article somewhere, but I can't find it now, I have a copy of it saved on my work computer, I will have to read it again tommorrow. It said in the article that this re-sizing eats up half of your win rate in the long run because of win-loss and loss-win sequences.
When I suggested doubling my bets, I should have said that I am trying to figure out if I should bet enough extra to offset the 50% loss in long-run win rate.
Hope this makes more sense.
Dewayne
I found that article on bjmath.com, it is called "What Proportional Betting Does to Your Win Rate
by Stanford Wong"
In reading it again, I think that this is a method of betting full kelly, without the full 13.5% ror , but at the price of 1/2 of the win rate, by re-sizing before every bet.
I think I figured this out, it seems that if I increase my bets, my ror will increase and I will have to resize more often, or increase the amount of bankroll change before re-sizing.
This article, unfortunately, was later lambasted by Peter Griffin. I would forget about it, if I were you.
The problem is that Wong's conclusions are for the average case and don't consider all the extreme possibilities. When all are factored in, you win at the rate you ought to win at -- not at half the rate.
Don
Posting Permissions
-*-
Bookmarks