BlackJack Simulator/Analyst

OK, I looked at your code. What a mess.

Instead of trying to ferret out the problems, I ran it, and looked at the decks post-shuffle.

After shuffling, the deck was composed of the following cards:
18 10
26 2
24 3
24 4
24 5
24 6
24 7
24 8
24 9
28 A
24 J
24 K
24 Q

The missing 6 tens are easy to account for. Your text string for the deck is missing one ten. Multiply that by 6 decks and you get 6 missing tens.
But, your shuffle also mangles other card counts, adding 2 twos and 4 Aces to the deck. I'm sure the rest of your code is similarly bug-laden.

Next time you come up with an "Advanced" strategy that disagrees with the experts, consider that your coding might be the culprit.

P.S. Now I see why you called the shuffling subroutine "suffer". :laugh:

Corrected missing 6 tens

Ken Smith,
When I sent my source code to you, I found the error to define a card $deck:
I changed from
$deck = "A A A A 2 2 2 2 3 3 3 3 4 4 4 4 5 5 5 5 6 6 6 6 7 7 7 7 8 8 8 8 9 9 9 9 10 10 10 J J J J Q Q Q Q K K K K" to
$deck = "A A A A 2 2 2 2 3 3 3 3 4 4 4 4 5 5 5 5 6 6 6 6 7 7 7 7 8 8 8 8 9 9 9 9 10 10 10 10 J J J J Q Q Q Q K K K K";
That is why 6 decks missing 6 tens and I needed to retest my program. I am happy you found the problem and tested my program.
My strategy may not agree with you as an expert, but please do not make fun of me as a retired programmer doing this just for fun.
You pointed out my bug and I will correct it. I was confused when my post disappeared and appeared back on your web site. I am wondering if this will post to your web. I hope it will not make you upset. Please let me know what you think. Blackjack simulator/analyst is an interesting topic I started.

Your barring had nothing to do with #1 and everything to do with #2. zg

cjjssc, Your post wasn't removed, and you weren't barred. I moved a post you made from News/Announcements into the more appropriate forum: Blackjack Card Counting. I left a temporary 3-day redirect so you could find it. Glad to see you've corrected the decks issue. I still think the shuffle needs some attention in your code.

My test proved your Hi-Lo card count work

Ken Smith:

Thanks, I updated my program to define one deck as
$deck = "A A A A 2 2 2 2 3 3 3 3 4 4 4 4 5 5 5 5 6 6 6 6 7 7 7 7 8 8 8 8 9 9 9 9 10 10 10 10 J J J J Q Q Q Q K K K K";
I used only one deck to test and display the shuffled cards, only display first 8 X 6 = 48 card as :

shuffle 1 decks: 106582AJ3 64Q692510 92AJ7996 4K8332Q3 874AAQJK 107QKK7J10


It has 4 tens, but I will modify program, when I have time :
1. Display all card to verity each card after shuffle.
2. Change advanced strategy to My strategy.
3. Users can defines each different strategy as my strategy from database by email name
4. This is just for fun.

It is on internet has limit speed only up to 10000 hands, but seat 4 use added $4 per true count,
And seat 7 uses $2 per true count.
My test proved your Hi-Lo card count work even on limit 10000 hands already.
(Dead link: http://bj21.tearsjoys.org)

Your "test" "proves" nothing. In a strip game using HiLo with a 1:15 spread and optimal betting, you have a 30% chance of losing in 10,000 hands even though you have an advantage. 10,000 hands is completely meaningless.

Wow . . . a software developer claiming there are no bugs in a piece of software he's written! That'll be a first !