Will I regret learning KO instead of High-Low?

[ScottH]

I don't get it. Here are a couple of reasonable situations.

Let's say 4 decks are gone and we're at the KO key count. With KO preferred, you raise your bet, but the high-low true count is approximately only zero. With the above system, you would not raise your bet.

Let's say 1 deck is gone, and we're just below the KO key count. With KO preferred, you still bet minimum, but the high-low true count is approximately +2, meaning you should have tripled your min bet in shoe, which you would do with the above system.

Obviously, I'm not attributing my upswing to this improvement, but...

KO is about even with High-low even with these flaws, right? You're saying these situations would only increase expectation from about 0.62 to 0.626? You gotta be kidding me...

Hi-Lo also has flaws that KO does not. It doesn't matter how you "reason it out", just look at the simulations, they are both equal so what's the choice is not which one is better, it's which one is easier for what you are doing.

 

[supercoolmancool]

The ease of KO comes with the price of higer variance. Exactly how much more I don't know.

 

[Automatic Monkey]

The ease of KO comes with the price of higer variance. Exactly how much more I don't know.

It depends on how far away from the center of the shoe you are. If you are playing a deeply-dealt 8D shoe, you will almost always be overbetting towards the end and underbetting at the beginning. In SD or DD it won't be so bad.

One time I encountered a group of skilled counters in a casino and I could tell they were using KO because of this problem.

 

[21forme]

It depends on how far away from the center of the shoe you are. If you are playing a deeply-dealt 8D shoe, you will almost always be overbetting towards the end and underbetting at the beginning. In SD or DD it won't be so bad.

That's easy enough to overcome by adjusting the betting ramp based on number of decks played (without actually calcualating a TC). I've been playing with deck-adjusted KO numbers for several months, with excellent success. Yesterday, KO-Rob updated a small spreadsheet with everything you need to know and posted it on the board.

 

[jee_pack]

it can't be that bad right? I mean, if the sims come dead even, who care wha the variance is? Unless it affects your risk of ruin? What exactly do you guys mean by "variance"? Are you talking about the swings (ups and downs) in your bankroll?

 

[supercoolmancool]

it can't be that bad right? I mean, if the sims come dead even, who care wha the variance is? Unless it affects your risk of ruin? What exactly do you guys mean by "variance"? Are you talking about the swings (ups and downs) in your bankroll?

Someone using Hi Lo can use a smaller bankroll than someone who is using KO. I don't know whether the diffence is significant or not.

 

[Brock Windsor]

Back of Basic Strategy Card

Brock - If you don't mind, please scan the index side of your card and post it as an attachment. I'd be most interested to see it. Thanks.

For 8 Decks:
I attached the word doc I used to create it. Decks remaining are down the left side. The two rows of numbers along the bottom are using the BRH bet spread for full wong depending on whether playing 1 spot (top) or two spots (bottom).
The indicies are from TC 1-6, I wong out negative shoes. The idicies are not identical to Hi-Lo, but rather Hi-Opt 1. Though Hi-Lo index #s would likely be more accurate I only had Hi-Opt 1 at the time I made the card, and I prefer doubling 10v10 at TC 5 instead of sooner as I find it a bit more risk adverse.

BW

 

[Brock Windsor]

Brh

It seems to me that you would still need IRC to be based on #starting decks -and- the indices to be recalibrated for the KO tags imbalance.

Lets see the BRH wong ramp. zg

Yeah I've adjusted for an IRC of -4* decks and True Counted the tags. The BRH 8 deck wong ramp is attached. Keep in mind with the BRH spread -2 is about TC of 1 and 0 would be TC of 2, 2 a TC of 3, etc.

 

[schismist]

Sim results in: damn.

So, I've been silent in this thread for some time now. The reason is that I've been preparing a sim of this system I proposed. I havent shelled out the dough for a simulator yet, so I wrote a program in R (the open source version of S-Plus statistical computing software). Plus, I had to do this to test this custom system. It turns out you were right. In fact, it does worse than KO.

Game 6D H17 DAS RSA LS, 80% pen

10 k shoe sim house advantage: -.57%

I know that many hands isn't much but my program is slow and this was really just a test.

KO Preferred, IRC=0, wong in at 12,14,16,18,20 wong out at -7,-2,3,8
spread 1-10 from RC 16 to 24

50 k shoe sim: +1.07%

My KO adjusted system, wong in at TC>=1.5, wong out at TC<-1
spread 1-10 from TC=2 to TC=5, same strat deviations with hi-lo indices

50 k shoe sim: +.90%

These were run with the same starting random number generator seed, so although not likely to be the true expecation with this many hands, should be comparable.

It seems to me that you would still need IRC to be based on #starting decks -and- the indices to be recalibrated for the KO tags imbalance.

As zengrifter suggested, I would guess the problem would be the ramp and the fact that the index plays aren't adjusted for the different tags. Probably, it this were corrected, it would outperform, but eh...

I guess I'll just go back to KO with decks-remaining adjusted ramping, and work on my act!

Thanks