You there!! Can you help me?

[tentacles99]

I'm trying to make a simple blackjack simulation using winning and losing probabilities. Does anybody know where I can find these numbers? I need to know the probability of: losing, tying, winning (with and without bj), doubling, splitting. And, in the case of a double or a split, I would need to know the above probabilities. It's not too important what the rules are or how many decks are used, I'm just trying to get a ballpark estimate. I would greatly appreciate any help and advice you can offer. I thought it would be easy to get these numbers but I can't seem to find them anywhere.

[Sonny]

Originally Posted by tentacles99

I'm trying to make a simple blackjack simulation using winning and losing probabilities. Does anybody know where I can find these numbers?

If you’re making a BJ simulator then it will be able to tell you all those things.

In general you will win/lose/push about 43/49/8% of the hands played. You can get more specific numbers from these websites:

http://www.blackjackinfo.com/bb/showthread.php?t=7446
http://www.bjmath.com/bjmath/conseq/streak.htm
http://wizardofodds.com/blackjack/appendix2.html
http://wizardofodds.com/blackjack/appendix2b.html
http://www.bjmath.com/bjmath/ev/6dh17.htm
http://www.blackjackinfo.com/bjtourn-dealercharts.php
http://www.blackjackinfo.com/bjtourn-doublechart.php

-Sonny-

[Canceler]

Originally Posted by Sonny

If you’re making a BJ simulator then it will be able to tell you all those things.

It looks to me like he wants to simulate the results of playing blackjack using probabilities, rather than having the program actually play simulated blackjack.

[Automatic Monkey]

Maybe this will help you. This is from some research I did to optimize online blackjack play with sticky bonuses. The rules were 8D, S17, DAS, LS. Expressed as absolute probabilities.

Effect of splits are not included because I didn't have the ability to segregate that data out properly in my sims.

Straight win: 0.3263
Straight loss: 0.4089
Straight push 0.07393
Soft DD win: 0.008772
Soft DD loss: 0.007284
Soft DD push: 0.001002
Hard DD win: 0.04697
Hard DD loss: 0.03114
Hard DD push: 0.006030
Surrender: 0.04344
Natural win (3:2): 0.04391
Natural push: 0.002188

Hopefully this is what you were looking for.

[tentacles99]

Thanks for your replies. Canceler is right on, I am looking for the kind of numbers that Automatic posted. Without splits, however, I don't think it would be possible to have an accurate simulation. I appreciate your response nonetheless. I'm curious Automatic, what program did you use to write your bj sim? With my very limited experience, I've been using Excel coupled with Visual Basic. Like yourself, I'm trying to optimize sticky bonuses - particularly ones with max cashouts.

[Automatic Monkey]

Originally Posted by tentacles99

Thanks for your replies. Canceler is right on, I am looking for the kind of numbers that Automatic posted. Without splits, however, I don't think it would be possible to have an accurate simulation. I appreciate your response nonetheless. I'm curious Automatic, what program did you use to write your bj sim? With my very limited experience, I've been using Excel coupled with Visual Basic. Like yourself, I'm trying to optimize sticky bonuses - particularly ones with max cashouts.

I used CVData to generate the results for the different kinds of hands. Because I had no way to separate the doubles after splits from the splits themselves, I assumed the win rates for split hands would mirror those of other hands and apparently it's very close, because my numbers match the win/lose/push ratio that Sonny posted.

Then I wrote a RNG generator in Liberty BASIC that would deal random hands in proportion to the way they would occur in a real game and give me target vs. bustout data for different bank sizes and targets. Turns out, for almost any kind of sticky bonus the ideal fraction of one's starting bank to bet is around 20%.