Knock-Out Blackjack

I have been reading and trying to understand Knock-Out Blackjack by Dr. Vancura and Ken Fuchs.

On page 48, (Chapter 3), while still using the gumball analogy, they state with an IRC equal to 0 the expectation is 0 when the running count reaches +4 (pivot point). The "key count" is reached when the running count is at +5. This makes total sense and is empirically demonstrable using objects such as gumballs, coins, marbles, cards, etc.

However, the table on page 75 (Chapter 5) states that starting with an IRC of 0 the key count is reached at a running count of +2. I can't make this observation, and the theory is apparently eluding me. It appears to me that there are still two more lower cards in a single deck (when I reach a count of +2) than high cards.

Where am I going wrong? I can't seem to find the answers in the text.

Thanks,
Cotton Bob

KO is an unbalanced count. In single deck, there are 24 low cards and 20 high cards. To even things out you multiply the count by 20/24, so at +2 you really have 2 * 20 / 24 = 1.7 more high cards remaining than low cards.

From what I remember, the gumball example used a balanced count.

Actually, the gumball example is a direct analogy. 20 negative items 8 neutral and 24 positive, which is the same as 20 cards, A - 10, 8s&9s are neutral and 2 - 7 are positive. And, as you point out, when the count in this particular systems is at +2, there are still more low cards than high cards left in a single deck.

I am a newcomer to card counting, but I have a pretty good foundation in statistics and stochastic probablity. Using the formulations given in the table for "KO Rookie" I can't seem to come up with any positive expectations for the player.

Is there someone using this system that can tell me if those are indeed the counts they use, or is it a misprint? My title page tells me this is the first edition published in 1998 and with reprints in 2000 and 2001.

Thanks again for any help.
CottonBob

I am a newcomer to card counting, but I have a pretty good foundation in statistics and stochastic probablity. Using the formulations given in the table for "KO Rookie" I can't seem to come up with any positive expectations for the player.

According to an article in www.bjmath.com, the average single deck effect of removing a single low card on betting is +0.46%. Therefore your advantage if you remove two low cards in a SD,H17 game is -0.18% + 2*0.46% = 0.74%.

Is there someone using this system that can tell me if those are indeed the counts they use, or is it a misprint? My title page tells me this is the first edition published in 1998 and with reprints in 2000 and 2001.

KO is an extremely popular system, due to its combination of power and simplicity. It is well regarded amongst BJ authors and experts. Although I use HiLo, I recommend KO to anybody, especially if you are just starting.

Just as a comparison, according to www.bjstats.com, the SCORE for KO in SD,H17,31 card penetration is 73.25, while for HiLo it is 69.65. The N0s are 13,719 and 14,412, respectively.

SCORE is the gold-standard for comparing strategies (higher is better). N0 is also used for comparison and equals the number of hands you need to play to have your expected value = 1 standard deviation (lower is better).