KO Question/Confusion

[b]

I was at an 8deck game yesterday using the KO count system and noticed that as I got deep into several shoes that the count got between -5 and higher more often, promting me to up the bet at that point.
As I pondered that on the ride home I started thinking more about the system in general. At an 8 deck game the start count is -28. Assuming a normal distribution of cards, at the end of say 6 decks the count should be
-4 all by itself and prompting a hgiher bet when the high/low would be at 0.
It appears that as the further into the shoe one goes, it pushes you naturally to raise your bet even if the count is not in your favor.
Everything I read says that KO pretty much matches High/low but now my simple testing at the kitchen table may show this not to be the case. Can anyone help me understand why my logic is wrong and what I am looking at incorrectly before I switch to high/low? And why I should stop "thinking" so much!
Thanks
Bobp

[sabre]

Your logic isn't wrong. KO is most accurate in the middle of the shoe. It understates your edge early in the shoe, and overstates your edge near the end of the shoe. KO was designed so that the pivot (+4 RC) would represent an advantage of about 2% consistently, regardless of where you are in the shoe. Unfortunately, the key count (the point where you first have a player advantage) isn't so consistent.

That's the tradeoff for not having to do a true count conversion. Counting an extra card (the 7) simply isn't as accurate as dividing by the remaining number of decks.

Fortunately, this effect isn't terribly devastating, especially if you are using a betting ramp that starts at the key count. The cut card prevents you from making too many terrible betting errors late in the shoe. (Unless you're playing somewhere with 90%+ penetration) The biggest problem is that you're missing profitable betting opportunities early on in the shoe.

Overall, you have to compare simulation results of KO vs Hi-Lo for your typical game conditions, and decide if switching to the Hi-Lo count is worth it for you. Personally, I prefer using an unbalanced count and avoiding a TC conversion.

[21forme]

I agree with the above post. I did some calculations for 6 decks and made some adjustments in key counts based on the number of decks played in this thread:
http://www.blackjackinfo.com/bb/showthread.php?t=5213

If you play 8 decks, you can do something similar, and not miss some advantageous counts early in the shoe.

[b]

Originally Posted by 21forme

I agree with the above post. I did some calculations for 6 decks and made some adjustments in key counts based on the number of decks played in this thread:
http://www.blackjackinfo.com/bb/showthread.php?t=5213

If you play 8 decks, you can do something similar, and not miss some advantageous counts early in the shoe.

Has anyone done the math for an 8 deck game and various deck penetrations?
Also, would you not need a different betting ramp level at each new adjusted key count based on decks played to take a higher advantage at each level?

[21forme]

Yes, you would need a different betting ramp. Just follow my example and refer to Renzey's post and the calculations for 8 decks is fairly simple.

The pivot (+4) remains about the same throughout the shoe and is equivalent to a TC of +4. So just interpolate and spread your bets accordingly.

For 6 decks from my other post:
All the following comments apply to 6 deck shoes with an IRC = -20...

The KO book says wong in at -4. However, Renzey made some excellent point in his post in this thread: http://www.blackjackinfo.com/bb/show...41&postcount=3

I followed up with a few more calculations and came up with the following:
Let's presume you want to wong in at TC = 1.5
at 1 deck this equals a RC of -8
at 2 decks RC = -6
at 3 decks RC = -3
at 4 decks RC = -1

As an example, suppose you reach a RC of -8 at one deck. PP remains +4, so that's 12 points. So if you're doing a 1:12 spread, increase 1 unit per point of RC.

It makes intuitive sense if you think about the TC conversion. With 1 deck played, the denominator for TC conversion is 5. With 4 decks played, it's 2, so the TC (and the ramp) would increase 2.5x as fast.

[EasyRhino]

Actually, this kind of confuses me. Here's why:

In a balanced count, if the RC moves, say 5 points in the first deck, then your TC is only +2 or so (10 decks left). In other words, the "positive signals are less strong early in the deck. If we're deep in he shoe (1 deck), then our TC is a whopping +10.

In KO, if the RC has shifted 10 points in the first deck, then you should wong in, using the new numbers. But if you've had a shift in the RC 10 late in the shoe, then you should even be playing.

And it's all because of the stupid 7's.

[Automatic Monkey]

Originally Posted by EasyRhino

...And it's all because of the stupid 7's.

It looks like that doesn't it? But it really isn't just the 7's that cause this effect, it's all the low cards put together. So the unbalance in KO is caused just as much by the 2's and 5's as the 7's.

The problem of overbetting towards the end of a shoe with an unbalanced count (especially KO) is very real and it is the reason I moved away from unbalanced counts for shoe play. My last trip to Vegas I saw a team of counters there, and I knew they were using KO because they were missing opportunities to bet early in the shoe and overbetting (sometimes significantly) towards the end.

[21forme]

Originally Posted by EasyRhino

Actually, this kind of confuses me. Here's why:

In a balanced count, if the RC moves, say 5 points in the first deck, then your TC is only +2 or so (10 decks left). In other words, the "positive signals are less strong early in the deck. If we're deep in he shoe (1 deck), then our TC is a whopping +10.

In KO, if the RC has shifted 10 points in the first deck, then you should wong in, using the new numbers. But if you've had a shift in the RC 10 late in the shoe, then you should even be playing.

And it's all because of the stupid 7's.

Rhino - I don't understand your reasoning. If you're playing 6 decks and your balanced count is +5 after 1 deck, your TC = +1. What's 10 decks left mean?
If you're deep in the shoe (say 5 decks) and your balanced RC is +5, then your TC is +5.

In KO, you're RC will rise +4 per deck with a neutral count, unlike a balanced count where it remains 0.

[EasyRhino]

For my example I meant to imply a running count shift of +10 on both systems. Oops.

Overbetting at the end of the shoe doesn't worry me that much from a fiscal standpoint. Where it DOES worry me is from a camouflage perspective. Floating out giant bets at the end of the shoe is the sign of a counter, and seems to lead towards getting half the shoe cut off at some joints. Having the ability to gradually unwind a big bet would be much more useful. After all, if one ends a shoe with minimum bet, there's nothing strange at all about starting the next one at min bet.

And a team using KO? While I have previously mused how it might be well-suited for Big Player call-in games... I didn't expect to hear of any teams actually using it. Was it a good team? Or at least good enough?

[ScottH]

Originally Posted by Automatic Monkey

It looks like that doesn't it? But it really isn't just the 7's that cause this effect, it's all the low cards put together. So the unbalance in KO is caused just as much by the 2's and 5's as the 7's.

The problem of overbetting towards the end of a shoe with an unbalanced count (especially KO) is very real and it is the reason I moved away from unbalanced counts for shoe play. My last trip to Vegas I saw a team of counters there, and I knew they were using KO because they were missing opportunities to bet early in the shoe and overbetting (sometimes significantly) towards the end.

Overbetting your advantage actually makes you more money though, it just increases your risk a tad. It doesn't seem like too big of a problem to me.