Let's say you put a black chip on the table. By the time the cards come out and the dealer comes to you, the count is so sour you wish you hadn't. Luckily, you have a double down hand. Do you double for less? Do you just take a hit? Do authors of counting books consider this when telling you how to bet?
This situation is another reason to play with as few other players as possible. (Others are: maximizing your hands/hour & maximizing the number of +EV rounds before the shuffle.)
Part One: You must be counting right! Many times you will have your big bet out when the count sours. For the count to dump, 10's and Aces are falling all around. You put the money out and the big cards fell, they just missed your hand. The bad side is that the deal also gets the big cards and you are stuck with a soft/double-down hand. Now the best thing to do is follow your Basic Strategy or index for the count you now have.
doubling index for the hand by the time you play, then don't double.
An interesting question may arise if you don't know the doubling
index for that hand, but you do know that the current count would
warrant a much lower optimal bet. I wonder if some sort of optimal
bet reasoning may come into play. Probably not, because it would
still depend on the specific hand. If it is 11 v 6 absolutely double.
If it is 11 v 10 that is different. It would be nice to see a real math
wonk answer this.
not where it was. If he count indicates a double, then double. Don't get hung up on where the count was or else you may end up not playing optimally, and that is going to cost ya.
A a general rule, Never Double for Less. When you double for anything, you give up your right to take a fourth card. That lowers your win probability. If it isn�t worth tie to double,. Then just hit.
There are a a few esoteric situation where optimal Risk-Averse strategy does call for a partial double. But these are very usual, and they really have more of an academic interest. If you have seen 1000 doubles for less, there is a very high probability that 999 of them were poor plays.
In fact, for most doubling plays, we would prefer to double for more if they let us. Unfortunately, I have not had much success in talking a casino into allowing doubling for more.
It is possible that the count has fallen so that you are below the doubling by the time you get to your hand. It is also possible that you are below the Risk-Averse index. Most players only learn one risk-averse index, but theoretically, you could have a different index for each bet size. More realistically, you might have an index for �big bet� and one for �small bet�. It is possible that the count is now below the �big bet� risk-averse index.
However, unless you are sure about that, just play your standard game. If your standard index calls for a double, then double for the full amount.
I agree that you should almost never double for less.
However even if you don't have a Risk Averse index for the play, it seems quite reasonable to at least require the True Count be one higher than your index. Given the approximate nature of indexes and that they attempt to identify the point at which an option just barely becomes better, you are hardly giving up any EV to increase the index by one, maybe even two.
likely accurate and valuable regarding how to play with indexes.
But, I think part of the original point is still missed.
Recent state of the art recommends NOT memorizing indexes,
except for the "Illustrous 18". So, this thread arises:
one does not know the doubling index, and yet the count has
dropped dramatically before the play decision. Perhaps one
answer is to consider play with several other players. In that
case maybe it IS warranted to memorize more indexes, so that
this dilemma can be avoided.
If you do not know an index for a play, and your basic strategy is to Double, then you should always double. Don�t make �Hunch Plays�.
I personally play with more than 18 indices. Total is about 30; I actually don�t keep an exact count.
My suggestion is that if you don't know the correct index number for a play, you should stick to basic strategy. Most of the index numbers that call for hitting rather than doubling in what would otherwise be a basic strategy double are fairly low. The frequency of occurance of these plays is rare enough that using basic is going to have very little effect on your overall EV. Guessing will probably cost you more in the long run than sticking to basic strategy.
Recent state of the art recommends NOT memorizing indexes, except for the "Illustrous 18".
I think I'm pretty well read on blackjack, but I sure never heard that one before! Original research?
ETF
Don Schlesinger, 2nd edition, 2000.
Also "Get the Edge at Black Jack" by John May.
Also "Burning the Tables in Las Vegas" by Ian Andersen.
ETfan means that the Ill18 have been around a while. A bit uncalled for, but that the rough-and-tumble of blackjack discussion, I guess.
I'm not being sarcastic. They don't say what you said: "Recent state of the art recommends NOT memorizing indexes, except for the "Illustrous 18". It's too strong a statement(especially with the capital NOT ;-) ). About the strongest statement Schlesinger made was you "might as well" throw 90% of your numbers away. But he said more study was needed, and since then, he's come up with the "fab four" plus two single deck exceptions, plus some other goodies from Cacarulo ...
I just thought you made it sound like a positive prohibition against learning more than 18 indices. All the recent, state of art stuff I know says every added index adds to your "score" and adds a certain positive EV.
ETF
sarcastic, then what? sardonic?
And I did understand Schlessinger a little differently than you.
To some readers,
like me, it does come off as a recommendation to only memorize
the Ill 18 (possibly also the fab 4). If not, then there seems to
be no "state of the art" recommendation at all as to how many indices
should be memorized. And in that case, that would contradict the
triumphalistic tone of these books themselves.
Myself, I use many more indexes, by habbit and because I feel much
more comfortable knowing "what the heck is going on".
The I18, sweet 16, fab 4 etc make a lot of sense if you only play shoes, but even then once you have them down solid you have a choice. Is the effort of learning and remembering one more index worth an additional .001% expectation for all your play from now on?
You will almost certainly play single and double deck at some time, and more indexes will add a lot more to your return in that case than shown by the analysis done for 4 decks. Once you know an index of course you'll use it whenever it applies. Starting with a small set of indexes and adding gradually does seem best, and the reassurance that you've got most of the gain against shoes from the first few helps.
It would be useful to select from a list of indexes based on your normal playing conditions, count, and bet spread. With them ranked by their effect on your return you could pick your own cutoff point in the list. Such a list can be generated without even needing any programming. You can use a spreadsheet with the "Virtually Complete Strategy Tables" from Theory of Blackjack, your count's indexes, sim results for TC frequencies, your tag values, and your betting schedule.
"sarcastic, then what? sardonic?"
How about THE TRUTH? I said "I sure never heard that one before!" and I had, in fact, never heard a recommendation NOT to learn indexes stated so strongly before.
"If not, then there seems to be no "state of the art" recommendation at all as to how many indices should be memorized."
Does there need to be? OK. Here ya go: Learn a minimum of 4 indices, up to a maximum of however many you feel comfortable learning.
"And in that case, that would contradict the triumphalistic tone of these books themselves."
Books about winning gambling systems tend to be triumphalistic. They sell better that way. Good word. I like that.
"Myself, I use many more indexes, by habbit and because I feel much more comfortable knowing "what the heck is going on".
So you're violating the most recent, state of the art recommendations???
I just didn't want newbie readers to get the impression there is some kind of new prohibition against learning indexes.
ETF
quibble: You say, "Does there need to be [a state of the art]?"
The fact is, in any endeavor with an intersubjective component, there
IS a state of the art. It can be valid to dialogue/argue about what
that state of the art is, but it does exist.
