Beginner's Math/Blackjack page
Re: KO always shows advantage on the end
Posted By: Leroy Nimka In Response To: KO always shows advantage on the end (Mika)
Date: Friday, 24 May 2002, at 4:26 p.m.
In Response To: KO always shows advantage on the end (Mika)
"I always have a problem with KO. If the cards are evenly shuffled, you always can think on the end of shoe that you have an advantage. Is it right?"
I think I know what you're saying, although your message title seems a bit misleading. By "if the cards are evenly shuffled", you really mean "if the deck is neutral throughout". That is, there are no excesses of high or low cards concentrated anywhere. You're complaining that under this condition, when fewer than half of the pack remains in the shoe, the K-O running count can be above the key count indicating an advantage when that advantage doesn't actually exist. That's true, and it's also true that at the beginning of the shoe, the K-O running count can fail to indicate an advantage when an advantage actually does exist.
Both effects are due to the lack of true count conversion, but simulations show that both effects are far less important than you seem to think they are. K-O is generally optimized for the point where roughly half the cards have been played, and it tends to lose accuracy as you stray from that point in either direction, making the pack look less advantageous than it really is at the beginning and more advantageous than it really is toward the end, if the running count is below pivot. (Above pivot the situation is reversed, but that's of little importance.)
Keep in mind that this loss of accuracy from the absence of true count conversion does not apply at pivot, where the most important decisions are made, and this is one of the factors that restores the strength to K-O, making it competitive with Hi-Lo. As you consider playing and betting decisions that are made farther and farther from pivot, the lack of true count conversion becomes more and more important, but the decisions that are made at pivot, of which there are so many important and profitable ones with K-O, can be made without any regard to the number of decks remaining. Remember, even a balanced count like Hi-Lo has a pivot, but that pivot, which is zero, doesn't indicate a very exciting and useful condition. The Hi-Lo pivot tells you whether to hit or stand on 16 vs. 10 or 12 vs. 4, but that's about it. The pivots of unbalanced counts like K-O (and Red Seven) are located at much more useful points, and tend to "focus", as I like to say, on more important and useful conditions. That's why the pivot of a count system is so important, and it's a factor that gives K-O a basic edge over Hi-Lo.
If you're really concerned about true count conversion, you may be better off sticking with K-O and adding true count capability to that system instead of switching to Hi-Lo. You can do true count conversions with unbalanced counts just as you can with balanced counts. You just have to select an initial running count (IRC) such that the pivot is zero instead of the Knock-Out Blackjack standard of +4. The true-count version of K-O is popularly known as "TKO" and the details can be derived from the information in Knock-Out Blackjack without much trouble, or, better yet, with a good simulator program like SBA.
I've also developed a novel variation of TKO that you might consider, which I call the "Color K-O System". You can jump to my description any time by clicking on one of my gophers. My web page also features some tables that will help illustrate some of the concepts about unbalanced counts that I've described above.
Messages In This Thread