need advice..computer sims and real world

#1
hey. Man, I'm so glad you decided to change software because now I can participate on the forum. I am a newbie when it comes to card counting. I haven't completely decided on a system and am experimenting right now. I made a computer program that will simulate a blackjack game. Right now, I know you, Mayor, say to go more with betting efficiency but I am leaning toward a system with good playing efficiency. It seems on my program that a system with good playing efficiency has less fluctuations in profit gain. I live in Indiana have been to a few casinos but am not that experienced yet in that either since I am rather protective of my money and a little paranoid that someone in the casino is going to know what I'm up to. I plan on playing single Deck and Double Deck as much as possible. Anyone in Indiana or Illinois know where the good double deck games are? I just want to know from those of you who test on computer programs, how accurate do you think a computer program can be? I am taking into consideration penetration % but my shuffling is one thing that I'm not sure about. I just don't want to take a grand to a casino and lose it. Thanks...justlearning
 

The Mayor

Well-Known Member
#2
You said: I just want to know from those of you who test on computer programs, how accurate do you think a computer program can be?

The answer is that they can be as accurate as you like. It depends on the question, and the type of algorithms developed to answer that question. In many cases, they are 100% accurate (combinatorial analysis). In other cases, where you are getting approximations, the longer they run, the more accurate the result.

I think your question really is: "do the results computers predictallow us to gain insight on real casino results?"

The answer is YES. Real life in casinos has a variety of other factors that we learn to deal with, but if a computer tells you something is good, then it is also good in a casino -- there is nothing about "the real world" that makes computer simulations invalid. Quite the opposite, it has never been shown that real world conditions produce outcomes that contradict computer simulations.

Let us know what type of programs you are writing and the readers herewill do what they can to give you good advice.

--Mayor
 
#3
Well, I have had a lot of time on my hands lately due to some personal problems of mine that will blow over in the next few months. With the time, I wrote a program in basic. It relies around the fact that there are 13 cards in a suit and 52 cards in a deck. 0 is 2, 1 is 3, 2 is 4, 3 is 5 up until 9, 10,11, and 12. 8 through 11 are 10's and 12 is the ace. So 27 is a 3 card since 27/13 has a remainder of 1 and 1 is 3. Anyone can write one of these and it will help you all that much more when you enter the casino. Or if you have money, buy one. I don't have any money. The good thing about writing one is that you can change anything you want. I'm hoping to pursue a career as a professional gambler. That's what brought me to the site. I majored in mathematics in college and after 5 years of it, I realized I didn't have anything to do with my knowledge. Is anyone just playing bj as a career?
 

The Mayor

Well-Known Member
#4
Greetings fellow programmer,

I have some C++ code I wrote for a few classes. I'll post it here when it is much further along (all my Blackjack sims were written in Java up until now).

What you described is one way to have the computer create the deck data structure. In my programs, I use a "Card" object, then a "Deck" is an array of "Card" objects. A "Hand" is an array of "Cards" as well, but dealt from a "Deck" object. I also have classes for a "Round" for "Dealer_Strategy" for "Player_Strategy" and for a few other objects. Altogether there are about 10 classes interacting.

Object oriented programming really rocks when it comes to programming blackjack!

Have far have you gone in actually getting your program to play/sim blackjack?

--Mayor
 
Top