Standard Deviant: Comp dependant indices...I don't get it

[Standard Deviant]

As a newcomer, I'm slowly but surely working my way down this board. Just came to a January thread on composition-dependant indices for 16v10.

I don't understand why anyone would care. I thought there was only composition dependant basic strategy, i.e., if you don't card count, you can slightly adjust your basic strategy to reflect the cards you see in your hand and the dealer's up card.

But if you're counting, your count takes all this into account, so you only need one index per hand, no? Whether I hold a 10-6 or a 9-7 or a 2-2-2-2-2-3-3 vs. dealer 10 shouldn't matter, should it? No matter what the composition of my hand, I'm keeping a count, so the composition of my hand is already reflected in the running/true count.

I saw the long thread, and no one ever addressed the "why" element, so I'm wondering if I have some fundamental misunderstanding of the value of composition-dependant indices. To me, they seem superfluous. If you're not counting, you can't apply any index because you don't know what the count is, and if you ARE counting, it seems to me that the count already properly reflects the composition of your hand.

So confused...!!!

[Norm Wattenberger]

It takes into account low and high cards, but not specific cards. In 16v10, drawing a five is perfect. So 5,5,6 is a tougher hand than 6,6,4 even though they have the same count since there are fewer fives left in the deck.

> As a newcomer, I'm slowly but surely working
> my way down this board. Just came to a
> January thread on composition-dependant
> indices for 16v10.

> I don't understand why anyone would care. I
> thought there was only composition dependant
> basic strategy, i.e., if you don't card
> count, you can slightly adjust your basic
> strategy to reflect the cards you see in
> your hand and the dealer's up card.

> But if you're counting, your count takes all
> this into account, so you only need one
> index per hand, no? Whether I hold a 10-6 or
> a 9-7 or a 2-2-2-2-2-3-3 vs. dealer 10
> shouldn't matter, should it? No matter what
> the composition of my hand, I'm keeping a
> count, so the composition of my hand is
> already reflected in the running/true count.

> I saw the long thread, and no one ever
> addressed the "why" element, so
> I'm wondering if I have some fundamental
> misunderstanding of the value of
> composition-dependant indices. To me, they
> seem superfluous. If you're not counting,
> you can't apply any index because you don't
> know what the count is, and if you ARE
> counting, it seems to me that the count
> already properly reflects the composition of
> your hand.

> So confused...!!!

BJStats

[Don Schlesinger]

> I don't understand why anyone would care. I
> thought there was only composition dependent
> basic strategy, i.e., if you don't card
> count, you can slightly adjust your basic
> strategy to reflect the cards you see in
> your hand and the dealer's up card.

We're talking about indexes. They are used uniquely by card counters. We're talking about different indices for different compositions of the same total. We're not talking about basic strategy.

> But if you're counting, your count takes all
> this into account, so you only need one
> index per hand, no?

No! That's precisely what the notion of C-D indices addresses. The index changes according to the composition of the hand.

> Whether I hold a 10-6 or
> a 9-7 or a 2-2-2-2-2-3-3 vs. dealer 10
> shouldn't matter, should it?

Yes, it should. and does!

> No matter what
> the composition of my hand, I'm keeping a
> count, so the composition of my hand is
> already reflected in the running/true count.

No, not true.

> I saw the long thread, and no one ever
> addressed the "why" element, so
> I'm wondering if I have some fundamental
> misunderstanding of the value of
> composition-dependant indices. To me, they
> seem superfluous. If you're not counting,
> you can't apply any index because you don't
> know what the count is, and if you ARE
> counting, it seems to me that the count
> already properly reflects the composition of
> your hand.

See above. As the hand composition changes, the index changes. Just because you counted the cards in your hand doesn't mean that the count hasn't been influenced by other factors, such as the cards in everyone else's hands.

Don

[Hollywood]

> We're talking about indexes. They are used
> uniquely by card counters. We're talking
> about different indices for different
> compositions of the same total. We're not
> talking about basic strategy.

> No! That's precisely what the notion of C-D
> indices addresses. The index changes
> according to the composition of the hand.

> Yes, it should. and does!

> No, not true.

> See above. As the hand composition changes,
> the index changes. Just because you counted
> the cards in your hand doesn't mean that the
> count hasn't been influenced by other
> factors, such as the cards in everyone
> else's hands.

> Don

I'm a bit confused by what is being said here, so please allow me to ask something.

In the counting system I use, in a 6 deck game doa, das split 3 times max. When the count gets to -4 my first indice kicks in. For the first time I go from 1 to 2 unit bet. And for the first time I do not hit my 16 against a dealer 10.

Now, when it's my turn to go, weather I have a 10 and 6 or I have 4 cards that add up to 16 as long as that count is at -4 or less, then I don't take a card.

If on the other hand I had a 4 and a 2 and I was at -4 and the dealer gave me a 10 which now is -5 and I have 16 I would hit that 16.

This is the way I always understood it to be.

If I am missing something, please tell me?

Thanks,

Hollywood

[Don Schlesinger]

> If I am missing something, please tell me?

You're using a single index for all hands totaling the same amount. That's fine. But, it's possible to learn composition-dependent indices, which vary according to what cards you have in your hand to make that total.

Don

[Sun Runner]

> You're using a single index for all hands
> totaling the same amount. That's fine. But,
> it's possible to learn composition-dependent
> indices, which vary according to what cards
> you have in your hand to make that total.

RE: Composition-dependent indices

Is ther an Ill 18(22) generated for these?
If so, what would be the % payoff in playing these as opposed to the standard Ill 18?

Thanks again.

[Zenfighter]

Is there an Ill 18(22) generated for these?

Yes, at the request of Mr. Lewis, I did the work for SD. The indexes were derived by algebraic approximation. For full tables see Wong's PBJ.

If so, what would be the % payoff in playing these as opposed to the standard Ill 18?

A very difficult question to answer with an ultimate precision. As with many BJ related questions an extensive simulation may be necessary to achieve a satisfactory answer.
According to Griffin, to compute the gain from learning cd indexes as opposed to the generic ones, you will need to compute the correlation for each two card hand, so as to obtain your average mean for the particular play and compare and see how much you have increased your correlation, in relation to the other one, the non composition dependence hand.

An example: Sd, s17, ndas, rsp = 4

 

Hand			Hi-lo corr.	Prob. of occurrence 

16 vs T		        .558358 

T,6 vs T		.589865		 0.014479638 

9,7 vs T		.522364          0.0038612368 

 

Here your weighted correlation for T, 6 and 9, 7 is:

Wc = (Sum Ci*Pi)/2 = .575654

So your total gain in efficiency using both indexes has increased by:

.575654/.558358 ~3, 1%

From a practical point of view, I?ll bet that the complexity and memory efforts added are not fair compensated in a substantial amount of EV increase. I could simple be plain wrong, a possibility, which should not be disregarded.

Sincerely

Z

[Don Schlesinger]

>From a practical point of view, I?ll bet that the >complexity and memory efforts added are not fair >compensated in a substantial amount of EV increase.

> I could simply be
> plain wrong, a possibility, which should not
> be disregarded.

No, surely, you're right.

Don

[John Lewis]

"I'm a bit confused by what is being said here, so please allow me to ask something.
In the counting system I use, in a 6 deck game . . . "

Are you confused about the fact that these CD indices were developed only for SD, not any other game?

The CD effect is pronounced at SD, but very diluted and much less significant (ie, not worth the trouble) @ eg 6 deck.

[John Lewis]

"From a practical point of view, I?ll bet that the complexity and memory efforts added are not fair compensated in a substantial amount of EV increase." -- Zenfighter

"I agree." -- Schlesinger

This sounds a little absurd coming from a novice responding to two experts, but I do not agree with either one of you on this subject.

For those CD indices that identify a very modest difference with the mean index, you two are, no doubt, correct.

But many of Zen's CD indices identify quite a range of indices within a single play.

For example, the mean index for 13 v 2 is 0. However the CD index for 10,3 and 9, 4 is +2. The index for 8,5 and 7/4 is -2. That'a a considerable difference in indices in an extremely common play. CD indices for 13 v 3 differ by 3 points.

The CD index of 10,2 v 2 is +6; 9,3 and 8,4 are +3; 7,5 is +2. 12 v 3 CD's show a comparable range. Again, a significant difference in indices in very common plays.

The CD indices for 16 v 9, 16 v 10 (both 5 points), and 11 v 10 (3 points) show a similar range of differences.

These are all very significant hands, all (with the exception of 11 v 10) in the Ill 18.

I would venture to guess that use of any of these groups of CD indices would be of greater consequence than use of virtually any non Ill 18 indices , and, as such, a useful refinement of the SD game.

Hopefully some day we'll be treated to an analysis of the question. Then I can apologize.

[Zenfighter]

Hopefully some day we'll be treated to an analysis of the question.

As been said above, computer sims are a sine qua non to solve these type of doubts. Btw, we have here computer programmers like Norm and Cacarulo that probably have the capabilities to address a solution via Monte Carlo simulation. An algebraic inference could be done also; maybe I?ll bite on this.

Nor Don, nor yours truly are saying that CD indexes when being used for SD are worthless.
Worthless is one thing and a fair compensation is another one. You have the same disjunctive when confronted with the option of employing multiparameter counts for SD versus simple and/or with and extra ace side count added. An individual who use Hiopt I with aces and sevens incorporated will raise his playing efficiency at the fantastic .736 level. Applause for him!

That is, certain individuals have the abilities to memorize and use complexities while feeling comfortable with them at the same time. All this stuff seems reserved for these ones. And for SD exclusively, of course! For the rest of the mere mortals, a simple count when used accurately will bring the Benjamins home. That?s tested!

Then I can apologize

No necessary at all. Being proud of yourself is much better!

Sincerely

Z

[Don Schlesinger]

> Hopefully some day we'll be treated to an
> analysis of the question. Then I can
> apologize.

You are almost certainly incorrect in your analysis. The use of all the indices you mention would be almost entirely inconsequential. But someone else will have to run a sim to prove it.

Don

[John Lewis]

Zen

Thank you for your reply.

"An individual who use Hiopt I with aces and sevens incorporated will raise his playing efficiency at the fantastic .736 level. Applause for him!"

Side counts are tough. Combine these additional mental efforts with those involved with counting, remembering the count, TC conversion, recalling and applying indices, and ordering a beer, and one is faced with a truely daunting task. I tried just a side count of aces for a time and decided it wasn't for me, at least not at this time. I have no personal experience with multilevel counts, but surely what's said about side counts largely applies there, too.

Comparing the mental work involved with learning and utilizing a complex count with the work of simply memorizing and utilizing a few extra indices is DISJUNCTIVE, however. Yes, I learned that word just now from reading your post, and it's a good one. You're teaching me English, and it's not even your primary language

Adding mental tasks to every play makes every play more difficult to process correctly. Index recall is quite simple, however. If the indices are well memorized, use of 100 is really no more difficult than use of 10, or 18. The mental work (memorization) is done prior to play. I use hi-lo with the full set of indices. My experience has been that recall of well memorized indices is essentially effortless, regardless of their number.

The number of balls you are juggling (balls=mental tasks) determines the difficulty of the effort. Memory recall remains one ball, regardless of information content (ie, number of indices), providing that information is adequately assimilated.

My assertion is that the CD indices cited in my post contribute more to one's game than do many of the other non-Illustrious 18 indices, and that some may be more significant even than most non Ill 18's. This may be incorrect, of course, despite Don's support of my opinion on the issue (kidding.)

But, certainly, the effect of any non-Illustrious 18 index on win rate is quite miniscule. I'm sure we all agree on that, as the numbers are indisputable. However, incorporation of these indices into memory and effective use during play is a relatively simple task. A task that, once performed, does not exert a constant tax on play effort.

I continue to feel these indices are worthwhile in my particular approach to the game, and I will certainly continue to use them. I may be the only one to do so on the planet, however. I'm starting to suspect that they may not make BJA 4.

Personally, thanks again for the indices.

JL