Ali Baba: KO

[Ali Baba]

I have questions about how in the world the KO can perform even close to the other counting systems. For instance, if you're playing Hi Lo in a six deck game, and six cards numbered 1-6 come out right off the top of the deck, then the running count will be +6, and the true count +1 - time to raise your bets or wong in. But if you're using KO and six cards numbered one throught six come out, the count will only be -14, and the key count will still be 10 low cards away. (The KO does count the seven, whereas the Hi Lo, does not, but for simplicity I used only 1-6 cards for this example). This lack of accurate information to size your bets would seem totally unacceptable and useless to the counter. Also, if there is 2 decks left in the game, and all the cards have come out in "perfect" distribution, then a count of negative 4 would actually mean a true count of zero in Hi Lo, but according to KO it's time to raise your bets. To take this example even further, with 1 and 1/2 decks left to be dealt, a KO running count of negative 4 could actually be a Hi Lo true count of somewhere between -1 and -2. Can anyone help me out with how the KO can still compare to Hi Lo and other counting systems. The lack of "true count" accuracy would seem completely completely unacceptable in many situations, especially off the top of the deck. Thank you.

Ali Baba

[Parker]

It is true that KO (and all unbalanced counts used in running count mode) tends to underestimate the advantage early in the shoe, and overestimate it very late in the shoe.

All counting systems are compromises, with various strengths and weaknesses. KO counts the 7, and is very accurate at the pivot, which is where we make our big bets. Hi-lo does not count the 7, a fairly important card.

Many people reacted as you when KO first came out. Now, 8 years, millions of actual hands, and probably trillions of simulated hands later, we know for a fact that, overall, KO performs nearly as well as Hi-lo without the complication of a true count conversion.

[Sun Runner]

> KO counts the 7, and is very accurate at the pivot, which is where we make our big bets.

I assume you mean if playing with KO?

Pivot point in a balanced count is TC of -0-, right, advantage gained .. about zero.

KO's pivot is approx 'Hilo TC of 4', right, advantage gained about 2%.

It just seems to me to be a disconnect in the definition of pivot point.

Please help me see where I'm missing it.

Thanks.

[Parker]

I was talking about KO -- that was the point.

All balanced counts have a pivot point of zero, which is equivalent of Hi-lo TC 0. All counts, balanced and unbalanced, are most accurate at the pivot point. Other than the 16 vs 10 index, not much of interest is happening at TC 0.

The pivot point of unbalanced counts is determined by the degree of imbalance. KO is unbalanced by 4, so the pivot will be determined by the formula: IRC + (4 x #decks). Since KO is a level one count like Hi-lo, the pivot will be roughly equivalent to Hi-lo TC +4.

This is a good place to be accurate. We have big bets out, and we are near several important indices, most importantly insurance (KO insurance is taken at (pivot point -1) or above). This is one of the main reasons that KO works as well as it does in running count mode.

Red 7, because it only counts the red sevens, is unbalanced by 2, so its pivot is equivalent to Hi-lo TC +2. Some feel that this is a more useful point than +4, given the mediocre pen so common in today's games, and sims seem to support this. Red 7 will perform as well or better than KO in many games with average pen, but in well-penetrated games KO usually edges out Red 7.

This is actually a simplified explanation of how unbalanced counts work in running count mode. If we were to use Hi-lo in running count mode, we would have accurate information at RC 0. However, we have no edge at this point. By the time the RC gets to, say, +4, our information is greatly affected by where we are in the deck, that is, RC +4 with 5 decks left is very different from RC +4 with one deck left. By unbalancing the count, we move the pivot point (where we have accurate information in running count mode) to somewhere more useful.

When we true-count KO, we adjust the pivot point to 0, but we must keep in mind that this is still equivalent to Hi-lo TC +4, so we will start ramping up our bets around TC -3, and our indices will be adjusted accordingly.

(Parts of this post are excerpted from my upcoming book.)

[Sun Runner]

> All balanced counts have a pivot point of
> zero, which is equivalent of Hi-lo TC 0.

So, HiLo is 'most accurate' at TC 0? Interesting.

> (Parts of this post are excerpted from my
> upcoming book.)

Put me on the list!

Thanks again.

[MJ]

How much better does TKO perform when compared to KO and Hi-Lo? Lets just assume a 6D game with AC rules with a 1-20 spread. Is there a significant difference in performance to justify true counting KO? Thanks for any help.

-MJ

> I was talking about KO -- that was the
> point.

> All balanced counts have a pivot point of
> zero, which is equivalent of Hi-lo TC 0. All
> counts, balanced and unbalanced, are most
> accurate at the pivot point. Other than the
> 16 vs 10 index, not much of interest is
> happening at TC 0.

> The pivot point of unbalanced counts is
> determined by the degree of imbalance. KO is
> unbalanced by 4, so the pivot will be
> determined by the formula: IRC + (4 x
> #decks). Since KO is a level one count like
> Hi-lo, the pivot will be roughly equivalent
> to Hi-lo TC +4.

> This is a good place to be accurate. We have
> big bets out, and we are near several
> important indices, most importantly
> insurance (KO insurance is taken at (pivot
> point -1) or above). This is one of the main
> reasons that KO works as well as it does in
> running count mode.

> Red 7, because it only counts the red
> sevens, is unbalanced by 2, so its pivot is
> equivalent to Hi-lo TC +2. Some feel that
> this is a more useful point than +4, given
> the mediocre pen so common in today's games,
> and sims seem to support this. Red 7 will
> perform as well or better than KO in many
> games with average pen, but in
> well-penetrated games KO usually edges out
> Red 7.

> This is actually a simplified explanation of
> how unbalanced counts work in running count
> mode. If we were to use Hi-lo in running
> count mode, we would have accurate
> information at RC 0. However, we have no
> edge at this point. By the time the RC gets
> to, say, +4, our information is greatly
> affected by where we are in the deck, that
> is, RC +4 with 5 decks left is very
> different from RC +4 with one deck left. By
> unbalancing the count, we move the pivot
> point (where we have accurate information in
> running count mode) to somewhere more
> useful.

> When we true-count KO, we adjust the pivot
> point to 0, but we must keep in mind that
> this is still equivalent to Hi-lo TC +4, so
> we will start ramping up our bets around TC
> -3, and our indices will be adjusted
> accordingly.

> (Parts of this post are excerpted from my
> upcoming book.)

[Sun Runner]

> How much better does TKO perform when
> compared to KO and Hi-Lo? Lets just assume a
> 6D game with AC rules with a 1-20 spread. Is
> there a significant difference in
> performance to justify true counting KO?

For the following conditions ..

1:20 spread / 6D, S17, DOA, DAS, SPA1, SPL3, NS, 5/6, C22, 1 player

Cacarulo posted the following c-SCOREs ..

TKO/A .. 48.96
TKO .. 46.59
Hi-Lo/A .. 46.55
KO .. 44.55
Hi-Lo .. 44.24

You'll have to decide how 'signifigant' the differences are to you.

Good luck.

[MJ]

Thanks for the info! Seems they are all pretty close. I think I'll just stick with KO or Hi-Lo.

-MJ

> For the following conditions ..

> 1:20 spread / 6D, S17, DOA, DAS, SPA1, SPL3,
> NS, 5/6, C22, 1 player

> Cacarulo posted the following c-SCOREs ..

> TKO/A .. 48.96
> TKO .. 46.59
> Hi-Lo/A .. 46.55
> KO .. 44.55
> Hi-Lo .. 44.24

> You'll have to decide how 'signifigant' the
> differences are to you.

> Good luck.