Suppose one could remember all of the cards played and those yet to be played, and could calculate the exact percentage that the next card won't bust. How could you use this to your advantage, how would you bet? I mean, obviously, the higher the percentage, the higher you bet, but it gets more complicated when you think about playing smart rather than just hit/stand. What could you do to maximize winnings. Thanks.
If you knew all the cards played...
- Thread starter somstuff
- Start date
They do in a sense, but even the most well known advanced count is a simplification. Wondering what is given up in the simplification? I'm guessing surprisingly less than one would think, but of course just a guess.
Just an interesting question on the theoretical upper bounds of a perfect super count.
Just an interesting question on the theoretical upper bounds of a perfect super count.
I think you all are thinking linearly because you use linear counts. What would be employed here is a straight 10 dimensional EOR counting system. I doubt anyone could actually do it even if they could remember all the cards but you would gain a lot if you could. Especially with few cards left to be played.
In order to actually make it within the limits of what anyone but a superhuman freak there would need to be compromises made. The compromises would give back some of the theoretical but I think you would still be amazed at the gain.
In order to actually make it within the limits of what anyone but a superhuman freak there would need to be compromises made. The compromises would give back some of the theoretical but I think you would still be amazed at the gain.
Any system derived from EORs, or even using them precisely, is still just a linear approximation. Combinatorial Analysis applied to each deck composition that you encounter is the only way to achieve 'perfect' betting and playing decsisions.
The difference between this perfection and the EOR-based approximation will be a function of the number of decks and the penetration.
The difference between this perfection and the EOR-based approximation will be a function of the number of decks and the penetration.
Ken Uston wrote about "George", a blackjack computer that does exactly what you're talking about. It's been a while since I picked up his book, so my memory may be a bit fuzzy, but I think he stated that over thousands of man-hours the team was winning at a rate of something like 3%. This is probably close to twice the EV of what a card counter can expect.
Gamblor said:
Ah yes, thanks Sucker, something like that is what I had in mind. Eh, wearable computer, 3%, probably enough cover from a bunch of oddball plays, not too shabby 
Today I think it could just about be done on a portable device (certainly for SD, which is where it would pay the greatest dividends), but see Sucker's warning!
London Colin said:
George will still have been EOR-based, as computer power back then could not come close to running a CA in real time.
Today I think it could just about be done on a portable device (certainly for SD, which is where it would pay the greatest dividends), but see Sucker's warning!
Today I think it could just about be done on a portable device (certainly for SD, which is where it would pay the greatest dividends), but see Sucker's warning!
That's not a practical proposition (except perhaps for an SD game with very shallow penetration). There would need to be a run of the CA for every possible subset of unplayed cards that you might come across. That's a very big number.
I'm not sure what you mean by indices in this context.
Pre-deal, the CA would simply give a precise figure for the advantage (notwithstanding the pair-splitting uncertainties discussed recently in other threads, which means that even a CA may not be entirely precise).
Post-deal, the CA tells you what actions to take on the initial two cards and after drawing each of any further cards.
As I said, a computer can more-or-less do it in real time now.
I'm not sure what you mean by indices in this context.
Pre-deal, the CA would simply give a precise figure for the advantage (notwithstanding the pair-splitting uncertainties discussed recently in other threads, which means that even a CA may not be entirely precise).
Post-deal, the CA tells you what actions to take on the initial two cards and after drawing each of any further cards.
As I said, a computer can more-or-less do it in real time now.
Yes, right LC, I assume we're dealing with such huge #'s that its not realisitic to pre-compute everything.
Lets say if this was pre-computed (tracking the 13 cards), in a 6 deck shoe, would have to account for approximately:
24 ^ 13 permutations ~ 900 quadrillion permutations
24 being the 0-24 remaining cards of each rank, 13 being each different rank.
I assume its a huge non-feasible amount of computing time for a CA of each permutation. And is this such a big number that it can't be stored on a computer)?
Maybe its possible a grouping of similar hand results and culling out non-realistic situations (for example less than 1/2 deck cut) might significantly reduce the #'s of calculations needed to be made and stored. Have no concrete idea how to do this, but somebody smarter might
Also, wondering how real time you can get with CA, as you mentioned "more or less", and I think your right. There would be situations where you can't really do it in real time in a real life situation, like lets say at a 8 deck game, you hit, and get a card, and have to make another decision, a computer would not be fast enough to do the CA of this new situation in real time?
Lets say if this was pre-computed (tracking the 13 cards), in a 6 deck shoe, would have to account for approximately:
24 ^ 13 permutations ~ 900 quadrillion permutations
24 being the 0-24 remaining cards of each rank, 13 being each different rank.
I assume its a huge non-feasible amount of computing time for a CA of each permutation. And is this such a big number that it can't be stored on a computer)?
Maybe its possible a grouping of similar hand results and culling out non-realistic situations (for example less than 1/2 deck cut) might significantly reduce the #'s of calculations needed to be made and stored. Have no concrete idea how to do this, but somebody smarter might
Also, wondering how real time you can get with CA, as you mentioned "more or less", and I think your right. There would be situations where you can't really do it in real time in a real life situation, like lets say at a 8 deck game, you hit, and get a card, and have to make another decision, a computer would not be fast enough to do the CA of this new situation in real time?
More pen = more value
Last year I wrote a .dll (.dll = Windows "dynamic link library") version of my composition dependent program that includes the ability to run sims either playing by using cd basic strategy or playing by computing what is best for the present shoe composition. If resplitting is allowed the decision of whether to resplit or not takes into consideration all of the cards played up to that point using my most optimal fixed strategy algorithm to compute split EVs (CDP1).
The .dll has no shuffle algorithm so a user of it needs to furnish the shuffle.
A sim that computes every playing decision does not generate a lot of rounds very fast because it spends a lot of time calculating. Basic strategy sims are faster. Also my basic cd program handles all up cards at once and I didn't change this in the .dll so even though one up card has been dealt in the sim, all of them are computed and the values for the particular up card dealt are used.
The value of considering all cards in making playing decisions will increase as penetration increases.
Last year I wrote a .dll (.dll = Windows "dynamic link library") version of my composition dependent program that includes the ability to run sims either playing by using cd basic strategy or playing by computing what is best for the present shoe composition. If resplitting is allowed the decision of whether to resplit or not takes into consideration all of the cards played up to that point using my most optimal fixed strategy algorithm to compute split EVs (CDP1).
The .dll has no shuffle algorithm so a user of it needs to furnish the shuffle.
A sim that computes every playing decision does not generate a lot of rounds very fast because it spends a lot of time calculating. Basic strategy sims are faster. Also my basic cd program handles all up cards at once and I didn't change this in the .dll so even though one up card has been dealt in the sim, all of them are computed and the values for the particular up card dealt are used.
The value of considering all cards in making playing decisions will increase as penetration increases.