Frequency of number of cards per hand.

[Skull]

Hey guys!

I am designing a very simple program in MATLAB to analyze some aspects of the game without the need to make lengthy simulations in CV.
So, in this program, I don't actually play the hands, but I define a random number of cards to each hand in order to compute the RC, TC, etc...

I know we get an average of 2.7 cards per hand, however, I don't know the correct frequency of occurrence of each number of cards per hand. I am using:

2 cards: 0.65;
3 cards: 0.15;
4 cards: 0.10;
5 cards: 0.06;
6 cards: 0.03;
7 cards: 0.01;

This distribution has an average of 2.7, but it is only a VERY WILD WILD guess.

Does anyone have the correct numbers?

I thank you very much in advance.

Best!

[Norm]

Cards per hand: Modern Blackjack page 430
Hand lengths: Modern Blackjack page 432

[Skull]

I don't know why I keep forgetting Modern BJ and BJ in Color have it all!!

Thanks a lot Norm!!

Best!

[Skull]

but it is only a VERY WILD WILD guess.

A TERRIBLE guess by the way...

[DSchles]

All right; I'll bite. Why do the percentages in all the columns of the yellow boxes, on p. 432, add up to considerably more than 100%? (Rounding seems very much NOT to be an option!)

Don

[Three]

Norm always gives 110% effort.

[Norm]

Norm always gives 110% effort.

This'll take a while to figure out.

[Three]

Splitting? Split hands counted as one hand for total number of hands and multiple hands for number of cards.

[Norm]

Possible.

[Norm]

Yep, that's it. To get the right numbers, scale each column to sum to 100%.

[Norm]

2	38.04%	51.48%	66.09%
3	45.36%	36.96%	27.74%
4	13.27%	9.58%	5.33%
5	2.92%	1.76%	0.75%
6	0.38%	0.20%	0.08%
7	0.02%	0.01%	0.00%

[Skull]

AAHH that's why my program was all around with its results!

Yeah alright... just kidding

Thanks Norm!

[Skull]

Hey Norm!

I have a different question now, but I decided it was not worth it to open a new thread.
Still with respect to my program, now I need to calculate the probability of a given hand yielding a split, a double, a BJ, an insurance or a surrender play. Then I need to calculate the probability of that play being a win, a loss or a tie.
I know CVData gives me all the information I need, I just want to make sure I am correct in my assumptions, because, as you know, there are A LOT of tables in CVData.
This is how I want to go about it:

- run my base sim and then go to Statistics/percents/True Count percentages. Columns 6, 7, 8 and 9 give me the probabilities of making each play per TC;
- then go to Statistics/WonLost/True Count Contributions and, for insurance, DDs and splits, divide the Win/Lose column by the Bet column in order to find the probability of winning or losing by TC;
- of course BJs are automatic wins and surrenders automatic losses.

Is this correct?

I am a little confused because of the TBA% column which, in my utmost ignorance, should equal the result of the aforementioned division. But, of course, my assumption is not correct...

Thanks in advance for any help!