Systems Archive 1

[ Read Responses | Return to the Index ]
[ Previous | Previous in Thread | Next ]

Re: KO.. 12 VS. 4

Posted by Pete Moss on 15 May 1998, at 6:05 p.m., in response to KO.. 12 VS. 4, posted by mike on 21 April 1998, at 4:52 p.m.

Point 1: Twelve vs. 4 is one of those plays that is so close off the top that it doesn't really matter much what you do. It is so close that the number of decks in play and the composition of your 12 can affect the theoretically correct decision. Playing one or two decks, you hit 10-2 vs. 4. Playing less that eight decks (which I trust you are!), you hit 10-2 if the dealer stands on soft 17. In fact, a card-counter who uses the Kelly criterion should actually hit all 12 vs 4 off the top, unless his bet is less than maybe about .4% or .5% of bankroll. I don't know the exact number. (Hitting reduces variance by making pushes possible.) Even that would vary by the composition of the 12. There are nits on nits.

Point 2: Because the index is not the pivot, one shouldn't be too surprised if the system does not give the theoretically correct play off the top. Running count systems are best for indexes at the pivot ("A" indexes in K-O Speak). To get accuracy for indices removed from the pivot, one must use true count conversion. "Accuracy" is relative to what the tags are capable of discriminating. The K-O tags with no side counts are not great for 12 vs. 4, even with true-count conversion.




The Systems Archive 1 is maintained with WebBBS 2.24.060398.