General question about unbalanced counts

[muckz]

It is often said that ub'd counts underestimate advantage early and overestimate later. With the concept of "floating advantage" wouldn't this counteract the overbetting later in the shoe?

If so, then an ub'd count is weakest early. In shoe games the advantage you find early is rare even when using true count methods.

I guess my question about ub'd counts is; is the only major weakness they have due to not having index plays that correlate with a TC?

I understand that PE is not measurable for ub'd counts, what about a true counted ub'd count? How much does true counting increase PE? For KO, UBZ2?

Lastly, George C. said it was not worth True counting UBZ2 although I have never seen evidence of such, is there any sim evidence? or are the card tags a limiting factor?

[ZenMaster_Flash]

"Under-betting" … "leaves money on the table"

"Over-betting" … radically increases your variance,

(thus your Risk of Ruin.)

The combination is deadly, especially in H17 games,

8 deck games, and games with shallow penetration.

[Jabberwocky]

A perfect indictment to unbalanced counts.

[Norm]

It is often said that ub'd counts underestimate advantage early and overestimate later. With the concept of "floating advantage" wouldn't this counteract the overbetting later in the shoe?

There is very little advantage in early parts of the shoe anyhow. Later parts of the shoe rarely exist these days. Yes, unbalanced strategies have a sweet spot. Fortunately, that sweet spot happens to coincide with the sweet spot of most current games.

I am not a huge fan of unbalanced counts. But, that's a personal preference. Frankly, I think the criticism about under-betting early in the shoe isn't a strong criticism. What matters is overall results. Unbalanced strats perform very well.

[seriousplayer]

It is often said that ub'd counts underestimate advantage early and overestimate later. With the concept of "floating advantage" wouldn't this counteract the overbetting later in the shoe?

I think you are talking about the KO Count because that is the only count that I can think of that underestimate advantage early and overestimate later. This is because the pivot for KO is TC = 4. For other counts like KISS III, Red seven, and UBZII their pivot is TC = 2. Those counts does not overestimate later in the shoe but they do miss early betting opportunities at TC of 1 or lower.

In term of "floating advantage" Don Schlesinger in Blackjack Attack distinguish it as more of a theory than practical in nature.

I understand that PE is not measurable for ub'd counts, what about a true counted ub'd count? How much does true counting increase PE? For KO, UBZ2?

For unbalanced count one thing to remember when you use unbalanced count as running count system is the count has a betting correlation and playing efficiency at the pivot not throughout the range of the count. For example, the pivot for KO is TC= 4. So at the TC = 4 (pivot) it will have PE= 55%. If you true count unbalanced counts the BC and PE will be throughout the range of the counts. Which for KO the PE =.55 throughout the range of the counts.

Lastly, George C. said it was not worth True counting UBZ2 although I have never seen evidence of such, is there any sim evidence? or are the card tags a limiting factor?

To know for exact you will need to do your own simulation with true counting UBZII and compare it to UBZII-Full. From there you will find the answer. I haven't seen any tables or charts posted from George C.

Below are the steps:

1. If you don't have the true count indices for UBZII. Generate true count indices for UBZII using index generating software.
2. If you don't have the full indices for UBZII listed at the end of the "Unbalanced Zen 2" handbook. Use index generating software to create indices for UBZII.
3. Then use SCORE to compare both strategies.

[Mr. White]

Here's some SCORE results. Six deck play-all 1-15 Spread 3 players, H17 DAS RSA Surrender, the same 24 play variations were used in each.

The UBZ indices used are more or less based on those found here: http://www.blackjackinfo.com/bb/showthread.php?t=6996.

True-UBZ indices are from Modern Blackjack print edition. I happened to have TUBZ already in my CVCX, so here ya go.

4.5/6 Pen
UBZ: SCORE 30.97
TUBZ: SCORE 32.67

5/6 Pen
UBZ: SCORE 50.13
TUBZ: SCORE 53.18

About a 5 to 6 percent increase. For some, not worth it; for some, a big deal.

Using the true-count strategy also allows you to add more indices to squeeze out a little more performance, if that's something you're inclined to do. You really can't add too many more worthwhile indices to the running count strategy.

[seriousplayer]

You really can't add too many more worthwhile indices to the running count strategy

I got this idea from "Modern Blackjack". You can add more worthwhile indices to running count UBZII by generating separate running count indices for each individual whole deck. It is like using the real true count indices of TUBZII. It involves using six different sets of playing indices for a six deck game. Even closer you can generate the indices for each individual half decks but at that time it will involve remembering to many indices.

[Three]

About a 5 to 6 percent increase. For some, not worth it; for some, a big deal.

I never turn down a 6% raise.

[Norm]

I never turn down a 6% raise.

Assuming you don't have to kill somebody.

[vegaskid]

Here's some SCORE results. Six deck play-all 1-15 Spread 3 players, H17 DAS RSA Surrender, the same 24 play variations were used in each.

The UBZ indices used are more or less based on those found here: http://www.blackjackinfo.com/bb/showthread.php?t=6996.

True-UBZ indices are from Modern Blackjack print edition. I happened to have TUBZ already in my CVCX, so here ya go.

4.5/6 Pen
UBZ: SCORE 30.97
TUBZ: SCORE 32.67

5/6 Pen
UBZ: SCORE 50.13
TUBZ: SCORE 53.18

About a 5 to 6 percent increase. For some, not worth it; for some, a big deal.

Using the true-count strategy also allows you to add more indices to squeeze out a little more performance, if that's something you're inclined to do. You really can't add too many more worthwhile indices to the running count strategy.

I know I'm digging this up from the grave, but this very topic has been on my mind lately. I'm assuming part of the increase in SCORE is due to more accurate betting due to true counting, and the rest is due to using TC indices. What I want to know is, how much is due to each? The reason I ask is, I use UBZ in RC mode for pitch games, but I true count it for betting purposes (strategy decisions still made by running count with the exception of insurance, 16 v. 10, and maybe one or two more of the biggies) in shoes. I really like using it in RC mode in pitch games, and I don't want to have two sets of indices in my head, so I use the 25 composite indices form the UBZ OS thread you referenced. I've always assumed that I'm not giving up much by not using TC indices in the shoes since the indices aren't nearly as important to begin with in shoe games. So, what am I giving up?

[vegaskid]

Thoughts, anyone? How much is it costing me to use true counted UBZ for betting purposes but not for index play (I just use running count for playing decisions)?