Enhanced 6-deck KO for Indiana Casino trip
  • I found the KO system pretty easy to learn so I thought I would take it to the next level and learn the full matrix for 6 decks. I started out trying to make flash cards for the exact count at which to depart from basic strategy. I ended up with 26 cards, 18 for regular plays (probably corresponds to the illustrious 18) and 8 for surrender (counting 8s v 10).

    The first thing I noticed was that I had a lot of odd numbers that would probably not be practical for me to memorize. The thing about KO is most of the index plays come in at +4, or 24 above the initial running count (RC) of -20. I spend a lot more time playing at counts between they key count of -4 (trigger to increase bets and first index play of stay on 16 v 10) and +4 than I do at counts above +4 (the so-called "Pivot Point").

    I decided that instead of dealing with all of the odd numbers, I would work in increments of 4 to further refine the KO system. Here is what I came up with:

    -12 or less: Hit 13 v 2, don't surrender 16 v 9
    -8 or less: Hit 12 v 4, don't surrender 15 v 10, Hit on 16 v 10 (or always surrender if allowed)
    -4 or more: Stay on 12 v 3, surrender 8s v 10,Double 9 v 2
    Zero or more: Double 11 v A, Stay on 12 v 2, Double 8 v 6, Surrender 14 v 10, surrender 15 v 9 or A
    +4 or more: Double 9 v 7, Double 10 v 10 or A, Double 8 v 5, Stay 15 v 10 (if surrender not allowed as above)
    +8 or more: Double 8 v 4, stay on 16 v 9, surrender 16 v 8 and 14 v A
    +12 or more: Stay on 14 v 10
    +16 or more Stay on 16 v 8 or A

    I think these adjustments are worth the extra effort to learn. Some of the differences are noteworthy:

    The 13 v 2 and 12 v 4 plays are omitted in the KO book for the 6-deck game. I figure they would at least make good cover plays.

    The dreaded 16 v 10 play: In a game without surrender, regular KO has you hitting it more often than you should (-4 count vs -8 above).

    The dreaded 12 v 2 or 3: Regular KO requires that you wait until the +4 pivot point to stay on these hands. You should actually stay once the count reaches zero or -4, respectively. No wonder I have busting these so often with a 10!

    Doubling: most of the doubling index plays still come at the pivot point (+4). However, we see that there are several cases where we should be doubling sooner! 9 v 2 comes at the key count of -4. At a zero count, we should now be doubling 11 v A and 8 v 6.

    I realize this analysis is a bit hard to read with surrender mixed in. If you are playing without surrender it is easy enough to take it out. From what is left, I think you will see that there is some nice benefit from learning just a few deviations from regular KO to improve playing efficiency. My guess is a gain of at least 0.10% The KO book seems to imply about a 0.20% gain from learning the full system, but I have done a little rounding and I think they tend to be optimistic.

    So I have tried to compute my expectation for this Indiana trip, with 6 decks, S17, DAS, lsr, spread of 1-5 with KO kelly preferred wagering. KO preferred with a 1-10 spread claims a 0.73% advantage. They don't state an expectation for 1-5 spreads, but it is 0.36% without kelly wagering and you gain 0.07% with it on the 1-10 spread. So let's estimate 0.40% and adjust from there. I will be wonging out of bad shoes so that has to be worth at least 0.10% (if not 0.20%). Late surrender should worth at least 0.10% (and some have told me that might be closer to 0.20% for a counter). And the aforementioned 0.10% gain in playing expectation from using the full, slightly rounded index.

    That comes out to 0.70%, and like I said I think that is conservative. It may be closer to 1%.

    Now is the time for my wise fellow counters to poke holes in my analysis. I appreciate your insights. Please refrain from bashing my 1-5 spread. I am fully aware that I can increase my EV (and definitely my variance) by doing that. When I am ready for a 1-10 spread (or if I find a $5 table), I will implement it.
  • Knox said:
    I found the KO system pretty easy to learn so I thought I would take it to the next level and learn the full matrix for 6 decks. I started out trying to make flash cards for the exact count at which to depart from basic strategy. I ended up with 26 cards, 18 for regular plays (probably corresponds to the illustrious 18) and 8 for surrender (counting 8s v 10).

    The first thing I noticed was that I had a lot of odd numbers that would probably not be practical for me to memorize. The thing about KO is most of the index plays come in at +4, or 24 above the initial running count (RC) of -20. I spend a lot more time playing at counts between they key count of -4 (trigger to increase bets and first index play of stay on 16 v 10) and +4 than I do at counts above +4 (the so-called "Pivot Point").

    I decided that instead of dealing with all of the odd numbers, I would work in increments of 4 to further refine the KO system. Here is what I came up with:

    -12 or less: Hit 13 v 2, don't surrender 16 v 9
    -8 or less: Hit 12 v 4, don't surrender 15 v 10, Hit on 16 v 10 (or always surrender if allowed)
    -4 or more: Stay on 12 v 3, surrender 8s v 10,Double 9 v 2
    Zero or more: Double 11 v A, Stay on 12 v 2, Double 8 v 6, Surrender 14 v 10, surrender 15 v 9 or A
    +4 or more: Double 9 v 7, Double 10 v 10 or A, Double 8 v 5, Stay 15 v 10 (if surrender not allowed as above)
    +8 or more: Double 8 v 4, stay on 16 v 9, surrender 16 v 8 and 14 v A
    +12 or more: Stay on 14 v 10
    +16 or more Stay on 16 v 8 or A

    I think these adjustments are worth the extra effort to learn. Some of the differences are noteworthy:

    The 13 v 2 and 12 v 4 plays are omitted in the KO book for the 6-deck game. I figure they would at least make good cover plays.

    The dreaded 16 v 10 play: In a game without surrender, regular KO has you hitting it more often than you should (-4 count vs -8 above).

    The dreaded 12 v 2 or 3: Regular KO requires that you wait until the +4 pivot point to stay on these hands. You should actually stay once the count reaches zero or -4, respectively. No wonder I have busting these so often with a 10!

    Doubling: most of the doubling index plays still come at the pivot point (+4). However, we see that there are several cases where we should be doubling sooner! 9 v 2 comes at the key count of -4. At a zero count, we should now be doubling 11 v A and 8 v 6.

    I realize this analysis is a bit hard to read with surrender mixed in. If you are playing without surrender it is easy enough to take it out. From what is left, I think you will see that there is some nice benefit from learning just a few deviations from regular KO to improve playing efficiency. My guess is a gain of at least 0.10% The KO book seems to imply about a 0.20% gain from learning the full system, but I have done a little rounding and I think they tend to be optimistic.

    So I have tried to compute my expectation for this Indiana trip, with 6 decks, S17, DAS, lsr, spread of 1-5 with KO kelly preferred wagering. KO preferred with a 1-10 spread claims a 0.73% advantage. They don't state an expectation for 1-5 spreads, but it is 0.36% without kelly wagering and you gain 0.07% with it on the 1-10 spread. So let's estimate 0.40% and adjust from there. I will be wonging out of bad shoes so that has to be worth at least 0.10% (if not 0.20%). Late surrender should worth at least 0.10% (and some have told me that might be closer to 0.20% for a counter). And the aforementioned 0.10% gain in playing expectation from using the full, slightly rounded index.

    That comes out to 0.70%, and like I said I think that is conservative. It may be closer to 1%.

    Now is the time for my wise fellow counters to poke holes in my analysis. I appreciate your insights. Please refrain from bashing my 1-5 spread. I am fully aware that I can increase my EV (and definitely my variance) by doing that. When I am ready for a 1-10 spread (or if I find a $5 table), I will implement it.



    I'm not a "KO-er" so I won't comment on those. But the idea of "quantization" (multiples of 4 here) has been done with other systems. I don't remember the who's or when's about this, but several have run detailed sims using this kind of approach (you have a group of +0 indices, then +4 indices, then +8, etc.) So that you remember the "groups" rather than a bunch of different plays, each with a different index (split 10's vs 6 at +4, 10's vs 5 at +5, 10's vs 4 at +6 for example. You could split em all at +4 or +5 and not lose a lot, and make the memory issue less significant.

    The other thing is that you can pull off a 1-10 easily by just betting 2x after a shuffle and going up or down according to the count after that point, which will have the pit noticing your $10 start and $50 max and nobody gets too concerned. This is a good plan for SD/DD games to hide that 1-4 spread also for SD or 1-8 for DD, which are the spreads I use... It isn't as effective in a shoe where the count will likely hover around 0 for the first deck or so no matter what...
  • Hey Knox

    Great effort on all the work. I have been using KO preffered for approx. 6 months now. I play strictly 6ders as the pitch games in my area are very weak. For starters, I believe their are only 13 matrix plays, as the last 5 are ommitted for shoe games. I know you could easily add more, as you have seen fit. But, how much do you actually add to your EV, especially for nickel players like ourselves, by remembering more indices, and what do you give up by making possible mistakes with added indices ? Where is the trade off. We know that Insurance & 16v10 comprise something like 46% of all plays. Check out Arnold Snyders site on thissubject. Im no math wiz by any stretch, but unless you are a high roller, is it worth it ? Also, I dont find KO indices that difficult to remember as per book. I have alsoadjusted IRC to -14, so when I hit the key count @ +2, I know '2 units' and so on. Here is how I try to remember indices:

    Insurance +9
    16v10 +2

    rest of indices are at +10:
    16v9
    15v10
    12v3
    12v2
    11vA
    10VA
    10V10
    9V7
    9V2
    8V6
    9V5
    As you can see, I go from highest hand to lowest for ease of remembering.Also, their is only 3 counts at which to remember making plays;
    +2, +9, and +10. It would be interesting to see if in fact your sims hold true, as I have read before that Fulks & Vancuras numbers are a little liberal. But what you would actually give up at the table as a nickel player, is minimal at best. Well, I dont know if any of this even helps, and as a fellow KOer, it is very interesting to see the work you have done. Keep it up and maybe we can help each other out. In closing, I will say I have had very promising results, and wonging really enhances this.

    good luck
  • Richard Munchkin did an interview with Johnny Chang and his wife (of M.I.T. TEAM FAME) and they basically said they didn't use indices-just basic strategy and hi lo count. No need to put yourself through changes memorizing long lists of indices-just get the money out!
    something to think about.
    rogue1
  • Thanks for the comments. I have had very good results using KO 6-Deck with S17. I have not yet played anywhere that offers late surrender and S17, but I will be tomorrow. There are three IN casinos in the Cinci area, and I intend to hit them all on my way up for business tomorrow which is 100% a travel day. 1 or 2 of the 3 offer lsr. I should also get to play Wed evening and Thursday afternoon.

    The tip about not starting off the shoe with a min. bet is a good one. I will probably start off with a green and let it ride until I lose. Then back to table minimum, with a pained look on my face!

    I think there is enough to be gained to make these additional plays worth knowing. I also customize the KO count by starting at zero (add 20) on a six-deck game. Then I hardly ever have to deal with negative numbers, I just count up. I exit if it stay at or near zero!

    I doubt there is much benefit to the index plays for +12 and above (unadjusted KO). Those would be +32 for me. However, I think there is certainly some benefit to fine-tuning, mainly those plays that occur between the key count and pivot point. Those are common situations that the KO authors have rounded a bit too much for my taste.

    So for me the most important plays will be, starting with the initial running count at zero:

    +16 (key count): Stay on 12 v 3, double 9 v 2
    +20: double 11 v A, stay on 12 v 2, double 8 v 6
    +23: insurance
    +28: double 8 v 4, stay on 16 v 9

    Anthing that affects doubling is a critical play to me. Notice there is only 6 plays that differ from the regular KO and one added.

    The other positive count plays (IRC = 0) are consistent with the regular KO I have already memorized. Obviously the surrender plays are important also. I think I will break those out separately and definitely will be memorizing those before tomorrow!

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!