Open Archive 2
[ Read Responses | Return to the Index ]
[ Previous | Previous in Thread | Next ]
Re: Is it possible to improve PE of KO count?
Posted by Pete Moss on 16 October 1997, at 3:58 a.m., in response to Re: Is it possible to improve PE of KO count?, posted by twind on 15 October 1997, at 9:14 a.m.
I made my own true count indices. I use 54 of them, a bit more than the standard 16 or 18. I used SBA. You can make approximate indices by taking the double deck indices and subtracting 4 from all of them. Ken Fuchs has also suggested that you will not be far off if you use Wong's Hi-Lo indices and subtract 4 from all of them. To do true count conversion, you start your count with IRC = -4*D, where D is the number of decks, not -4*(D-1) as recommended in the book.
I don't remember if the standard deviation of money results per round is lower when optimally betting 1-3 with my system than with optimally betting 1-5 with plain K-O. The optimal unit size is larger in the first case, because of the higher EV. However all that is factored in when calculating the expected number of rounds to double the bank. I assumed proportional betting (fraction of bankroll), with the bankroll size adjusted after every round. That is obviously a simplifying assumption that cannot be done in practice.
Pete
Responses
The Open Archive 2 is maintained with WebBBS 2.24.060398.