Blackjack Card Counting Question

I once heard from a very gifted card counter that, using hi-lo in double deck, once the count becomes "very, very positive" that the edge to the player is lost. First, is that true? If so, what would be considered "very, very" positive? What would this mean if playing on a 6-deck shoe?

The card counting indices are based on linear perturbations of the normal deck composition. Small changes in the deck result in small changes in player edge, and are linear based on the size of the deviation. Thus the removal of two 5's results in twice as much gain as one five. This is true over a wide range of possibilities and for all practical purposes it applies to all situations encountered in real play.

However, the linear theory does not hold up for large deviations from the standard proportions of cards. For instance, if there are nothing left but 8's left in the deck, how much should be player bet? The answer, obviously, is bet everything and stand. Split 8's to the allowed maximum number of split hands. The dealer will receive a hard total of 16. He must hit and bust. Therefore the player's edge is 100%, The Hi-Low, HiOptI and many other counting indices would not even hint at this opportunity.

So I would think it possible that at extremely high values, the Hi-Low might become non-linear and actually indicate a negative expectation.

Thorpe has a good discussion of this phenomenon in 'Beat the Dealer'.

Interesting. I think I know what you are talking about but counting cards is beyond me. I can barely count my fingers and come with ten each and every time. So my question to you is: unless you are able to count down a six or eight deck shoe, what good is the information going to be to you. Poker players who get lots of practice remembering what was dealt and to whom can probably count cards but it doesn't seem to make them winners at Blackjack. How many places deal a single deck game anymore and if they do, how many of them let the dealer get anywhere near the end of that deck without shuffling?

deleted

Quote: majmike

I once heard from a very gifted card counter that, using hi-lo in double deck, once the count becomes "very, very positive" that the edge to the player is lost. First, is that true? If so, what would be considered "very, very" positive? What would this mean if playing on a 6-deck shoe?

I think this page provides an extremely in depth answer to your question:

http://www.qfit.com/cardcountingindexes.htm

Quote: Ibeatyouraces

Counting cards is extremely easy. All it takes is a little practice. If you learned basic strategy, and remember how long it took to do that, you'd understand its not that hard. It just takes some concentration. As for counting 6-8 deck, it is exactly the same as counting 1 deck except the procedure is different since in a 1D game cards are dealt face down. Some people will disagree or have their own preference, but I find that a 6 deck game is a little bit better than both 1 and 2 deck games ONLY because there is less heat or scrutiny from the floor and that the rules tend to be better for a shoe game. And when you have that huge count in a shoe game, it lasts longer than it does in a hand held game. Nothing like seeing a +20 running count evaporate in 2 or 3 hands in a 2 deck game or the dreaded shuffle up happen, not that it couldn't happen in a shoe game but I have yet to see it.

Counting cards is NOT easy, you need to have basic strategy down FLAWLESSLY (there's "counters" who make BS mistakes, those guys obviously don't make any money), have to be able to count down a shoe flawlessly, and you need to know the index plays.

And huge count deviations happen less often in a 6 deck shoe then a 2 deck game.

f2d - that page exactly answers the question. It verifies what I was told. Who woulda' thought that the advantage tapers of at the extremes. Tks again.

We should be careful what conclusions we draw: those charts show that the advantage may decrease for each additional increase in the true count (TC) at very high counts (TC >20), but there is still a huge player advantage. Also, it would be extremely unlikely to be in a TC > 20 situation in 6 decks.

An interesting factoid that the marginal increase is not linear, but you should still be betting your bankroll-adjusted max at a TC of plus 20

You are right, of course. I would still bet a bankroll-adjusted max and seeing a TC of >20 is certainly unusual. I just had never seen the above referenced chart which seemed to collaborate what I had been told.