Another betting ramp question.

Question:
(Assume 6 deck shoe game with 1.5 decks cuttoffa and playing Hi-Lo).

When there is a high RC (ie +12) but there are only two decks in the shoe, and one and half of those are out of play, does anyone scale back their betting ramp? On balance of probability, the majority of the 12 additional 10s/Aces will be out of play - with an assumed even distribution, there will be only 3 additional cards in play. I appreciate that the TC average will still be +6, but at this stage is it fair to assume that 75% of the count advantage is negated?

Although I'm sure that things will average out over millions of hands run from a sim, as a way of minimising risk does it make sense to scale back the betting ramp to, say, 50% (4 units instead of 8 when using 1-8, 8 units instead of 16 when using 1-16 etc).

A similar instance happened to me recently where there was a RC+12, but only enough cards in the shoe for a final round for the seven players at the table. Never had the bottle to put out the 8 units the count demanded.

Comments ?

Nope. When the TC is high you have +EV, and you need to put the $ out. Developing a stomach for this is part of learning the game. With a high TC, you have a higher likelihood of 10s appearing, regardless of how few decks left.

(It is a bit annoying when the TC gets high and it's time for a shuffle...)

There must be an optimum point up to which it is prudent to ramp up bets, and after which increasing the number of units is detrimental. Perhaps why in some counts, adjustments are made to index plays in the second half of the shoe? In the example given, where (assuming an equal distribution) of 12 additional high cards only 3 will be in play, as these would be dealt out amongst 7 players the advantage would be all but diluted away - in contrast to playing the last 7 hands yourself heads up.

In the early stages of a shoe, it's not possible to know where any additional high cards will be (unless you've shuffle tracked), but as more cards hit the discard and the count stays high it becomes easier, and eventually with nearly all cards played out the conclusion can be reached that that may never have been in play in the first place, and the advantage anticipated isn't actually there. The question is really, whether one adapts play to take account of this or just sticks to the betting plan knowing that on balance of probability these additional high cards are not going to put an appearance in?

If you're at the point where the liklihood is that additional high cards are behind the cut card, why increase the bet if those cards will never be played?

Not really, at least not that has anything to do with the cards. Your maximum bets are limited only by your bankroll, and what you can get away with at the casino. Read up on Kelly betting.


Nope; not in a shoe game (perhaps in SD/DD, it's a bit more complex). It's all about ratios and odds of having high cards come out, and without any other information, you always have to assume an even distribution. If there are lots of high cards left, they're more likely to come out, period.

cut card effect

this link might be interesting as far as this stuff:
http://www.blackjackincolor.com/blackjackeffects1.htm

Well, without meaning to sound argumentative, that's the point. If a RC+12 is assumed to be equally distributed across 25% of cards in play, and 75% of cards out of play, then the odds are that most of the additional high cards won't come out? If you accept that, then there follows a case for not raising a bet in the same way as you would if you thought the probability existed that they would? There must come a point when it becomes apparent that the odds are that a significant number of additional high cards (75% in the example given) won't be dealt.

Looking at another, but related, matter, if it was purely about averages and making no allowances for what is known they'd be no difference in playing heads up against the dealer and on a full table with six other players. But of course there is - if playing heads up against the dealer, any advantage is all yours, if playing on a full table any advantage is diluted away amongst all of the players at the table. Hence APs avoid crowded tables, as dilution of the slim advantage eats into the +EV.

Thanks for the link but I'm not sure it's the same thing. My question is whether you should use what you know about the composition of the remaining shoe to adjust the betting schedule.

Taking my query to the silly extreme, if the RC of a remaining shoe is +12, but the cut card has just come out, do you still have an advantage? If you had the option to bet at that stage, would you put out a big fat one based on the TC? No, because there can be no advantage assumed at that stage as 100% of cards are out of play and will be reshuffled and a new shoe started. So why continue to assume there is an advantage of TC+6 when 75% of the remaining cards will not be played? Can't think anyone here would play a shoe game with 25% penetration from the outset, so why continue to keep betting to a ramp when similar conditions occur later in the game?

me, i dunno, i think it has to do with the limitation of what we can know.
like ok let's say we are heads up with the dealer. say there are six cards left before the cut card. say the true count is 4 and there are 52 cards behind the cut card. well for one thing you don't have to necessarily get all high cards to have an advantage. your first two cards might be 5 & 6. so in this case you have an advantage because of a good double down scenario. hey you might even double down against the dealer in that case even if he has an ace (after you take insurance) .
but let's even ignore the advantage in this scenario that the possibility of double downs affords us. let's just consider the possibility of us getting a snapper. well it's true we don't know where the heck those aces & faces are lurking. maybe they are all behind the cut card, but here is the thing, we don't know and can't know, this is one of those rare opportunities and cut card or no cut card when you have 58 unseen cards and the true count is 4 then in our ignorance beyond that information all we can say is this just might be a good chance for us to receive a snapper. so you just pretty much gotta go for it or at least should want to go for it.

You're absolutely right. We can't know and don't know where the additional high cards are. So assume equal distribution. Where does that leave us? On the risk/return consideration, does it warrant a max bet?

:=)

but you can't assume equal distribution in the case of a positive true count.

In this situation, if it's possible, as this requires you to be heads up or with one other player and no restrictions on how many boxes you can play. I estimate how many hands there will be played before the cut card and try to manipulate the shoe by the amount of boxes I play, so that the cut card will hopefully appear a card or two in to the last round, I then spread to as many boxes as possible. This basically gives you better pen in a positive count but with the drawback of higher varience.

This is where you're confused. It doesn't matter that some of the high cards are expected to be cut off; if the remaining cards are rich in high cards, they're still more likely to come out than they were before. This still holds even if some aren't going to be dealt. It's still more likely - the odds hold.

Calculate the odds if you need to convince yourself.

"Sort of". It's true that the advantage is shared, but if you're playing all hands so is the disadvantage. There's a far larger effect from a much lower hands/hr; that's the real reason for playing alone (also wonging effects). But aside from this the number of players doesn't matter much.

I've read your response several times, and I'm not sure we're talking about the same thing.

Some thoughts on the example discussed:

If a shoe has a running count of +12, but the cut card was pulled during the last hand, there is a zero chance that the additional high cards will come into play in that shoe. So there's no advantage within that count at all.

If a shoe has a running count of +12 and there's no cut card (all remaining cards are in play) then there's a 100% chance that the additional high cards will come into play. Bet to full extent of one's ramp, quarter Kelly or whatever system you use.

With these two extremes, I can't see how it can be assumed that all of the additional high cards will come into play when 75% of the cards are cut out of play? - assuming the twelve additional high cards in the shoe are equally distributed. There's always the possibility that all twelve additional high cards, together with the normal distribution of 10s/As per deck, are in the half deck in play but on balance of probability they all won't be. In this case, would continuing to bet to the full extent of one's ramp etc not be overbetting, bearing in mind that the table is full with 7 players.


I'll mull it over some more. . . . . .

Thanks for your input.

I think I understand your confusion. The improvement in EV is not based on having *all* (or most) of the high cards appear before the end of the shoe; it's whether your *next* hand is likely to have high cards come out.

If there are two decks left, and 90% of them are high cards, it seems pretty likely that a high card is coming out soon, doesn't it? It doesn't matter if 1.5 decks are cut off.

Still mulling . . . .