question about sims and betting amounts

[michael990]

we all know the house edge in blakcjack given AC rules and basic strategy played to perfection is about .05 %, meaning we can expect to win about 49+% of the hands in the long run. Given this fact, I came up with a betting system that seems to me to be a guaranteed profit in the long run, especially on a computer simulation of say a million hands.

start with a flat bet of lets say $200 on the first 10 hands, win or lose. Then starting with the 11th hand, bet $500 if your current win % up to that point is less then 45% and bet $50 per hand on any hand where your current win% up to that hand is greater than 55%. If your win % to that point is between 46 and 54 % then stay with the $200 bet.

So this basically takes advantage of that the long run will approach that 49.5% average win rate. when we are below that we are bound to get closer to it with more wins, when we are above it we are bound to see losses. This wouldnt be practical in an hour session in a casino, but I think in one of those million hand computer simulations it would come out ahead. Thoughts?

[AlexD30]

we all know the house edge in blakcjack given AC rules and basic strategy played to perfection is about .05 %, meaning we can expect to win about 49+% of the hands in the long run. Given this fact, I came up with a betting system that seems to me to be a guaranteed profit in the long run, especially on a computer simulation of say a million hands.

start with a flat bet of lets say $200 on the first 10 hands, win or lose. Then starting with the 11th hand, bet $500 if your current win % up to that point is less then 45% and bet $50 per hand on any hand where your current win% up to that hand is greater than 55%. If your win % to that point is between 46 and 54 % then stay with the $200 bet.

So this basically takes advantage of that the long run will approach that 49.5% average win rate. when we are below that we are bound to get closer to it with more wins, when we are above it we are bound to see losses. This wouldnt be practical in an hour session in a casino, but I think in one of those million hand computer simulations it would come out ahead. Thoughts?

Very interesting. This deserve further studies. But for short run on hourly basis you can expect big fluctuations. I guess.

Las Vegas downtown Fremont Street guys will definitely study this and come up with an answer in couple days.

[Nickels_n_Bullets]

Blackjack for Ac rules (no Surrender)

Win 43.5 Lose 47.9 Tie 8.6 Total 100
Ignore the ties 43.5/(43.5+47.9) = 47.6% win 52.4% Lose
House advantage using Basic Strategy is 0.4%

Due to DD and splits the 43.5 wins generates 47.5 units won

Therefore, you win 99.2 units and you lose 100 units... Of the 199 units bet the house keeps less than 1.

BUT in 199 HANDS you will win about 95 times.

This is much different than what you thought was happening.

[michael990]

i see what your saying nickels, but my guess at the expected number of wins per 100 hands isnt important. what im saying is..when your win rate is well below what you can expect bet higher, when its above bet lower.

in other words, lets say after 100 hands i have only won 30. well the math says that over the next 100 hands i should win about 65 of them. so i would bet higher. conversely, if i win 65 out of my first 100 hands, then the odds say over the next 100 i should only win 30. if i base my bet amounts on this then i should come out ahead. again this theory may be useful only in sims of 1 million hands or greater, but if i could come up with something that works for that, then maybe i can beat a FAIR online blackjack game.

[Kain]

The problem is that the odds for every hand don't take into account your previous hands.

You could theoretically go into a casino and lose every single time, or win every single time.

[Ray]

kain- You are correct....History don't matter in BJ. When you reach the
30 level, what will happen in the future is unknown. You can't say that
you will move closer to anything. Same is true for the 65 level.

Ray

[Nickels_n_Bullets]

I agree with Ray and kain here... no matter what happens over the previous 2,10,50, or 50,000 hands, your odds of winning the next one is still 3 in 7. The only thing that is different is the score of the method of choice (the count). That means you have an edge, but are not guaranteed a better chance of winning, although its more likely you get a better hand.... but thats true for the dealer also.

[AlexD30]

I agree with Ray and kain here... no matter what happens over the previous 2,10,50, or 50,000 hands, your odds of winning the next one is still 3 in 7. The only thing that is different is the score of the method of choice (the count). That means you have an edge, but are not guaranteed a better chance of winning, although its more likely you get a better hand.... but thats true for the dealer also.

Counting gives you a minuscule edge in theory over the long run but you still need to win that specific hand when you max bet out on positive. When you plunk that max bet, the true count has got to match the winning hand for you to get the “rewards” of betting by the count, else is no value whatsoever. Playing card counting is like trying to grab a winning white ball from a container that has 51 whites and 49 blacks and those are divided in 27 losing whites, 24 winning whites, 25 losing blacks and 24 winning blacks. You should expect to lose big frequently in the process because you have to match a winning hand to be white in order to have the big bet won. The big bets are the key for the card counter. No matter what, you still lose more hands than win regardless what the count is.

Card counting is an academic subject and a fancy way to talk about BJ but is not the way to make money. If the purpose of playing this game is to make serious money than card counting is not the way to go about. I played card counting full time for many years and made very little money with much work in between. The rewards are not worth the effort.