Don/other experts, How to find covariance of a game ?

[Gramazeka]

I always thought it would be better if those votes were not anonymous.

Let them put a minus. The main thing is that they do not avoid discussions))

[Gramazeka]

I wrote to Snyder sms, maybe he will comment personally, I will share with you. Although I'm sure he meant specific conditions, where specifically you can't channel an ace into a specific hand. Better to split it with the dealer 50/50 and not necessarily as the first card. But that doesn't hold up for sequencing as a universal advice about playing with only one hand. But it is completely mathematically suitable for slug tracking.

[aceside]

That's top of the deck, ignores counts, ignores indices, ignores cut card, ignores changing number of hands.

I have been thinking about the calculation of the covariance analytically. Actually we can include all these factors you mentioned, all by changing the card composition of the original deck. For example, we use a man made shoe with extra ten valued cards to do the same calculation of covariance.

[aceside]

And while variance obviously increases with optimal bets on multiple simultaneous hands, so does e.v., by the same percentage, leaving risk of ruin the same.

Don

I agree that variance increases a lot with multiple simultaneous bets, but disagree that the ev does so. That is faulted math. The only situation for playing three or more multiple hands is when the cut card is about to come out but the count is still skyrocket high, but this means the player has been losing.

[DSchles]

I agree that variance increases a lot with multiple simultaneous bets, but disagree that the ev does so. That is faulted math. The only situation for playing three or more multiple hands is when the cut card is about to come out but the count is still skyrocket high, but this means the player has been losing.

One of these days, you will read BJA3, especially, in this case, pages 24-26, and then I won't have to read, for probably the first time in the past 30 years, about my "faulted math."

I will say that, for someone who posts something incorrect in virtually every thread that you participate in, you aren't shy. But, that isn't necessarily a good thing.

Don

[aceside]

One of these days, you will read BJA3, especially, in this case, pages 24-26, and then I won't have to read, for probably the first time in the past 30 years, about my "faulted math."

I will say that, for someone who posts something incorrect in virtually every thread that you participate in, you aren't shy. But, that isn't necessarily a good thing.

Don

I will definitely read your BJA3, but I must get the money out of blackjack. That is why I am not shy at all.

[G Man]

Indeed.
What Im guessing is he was trying to advice against playing multiple hands at max bets following the intuitive logic that more hands = more chances of getting the ace.
I do find it very surprising that he didn't write anything about the possibility of doing that with minimum bets for steering purposes or to cover for the max bet.
Since this is the strategy used by almost all sequencing teams (at least the ones I know) maybe he didn't write about it on purpose.

Hope you are doing well my friend.

What Snyder meant was: When tracking a specific Ace, using key cards for location and alone with the dealer. If you're not 100% (and you rarely are) sure of the position the Ace will land, keep it simple and play only one hand. By doing this, you will "split the Aces" with the dealer, giving you an overall edge of approximately 9% on each of these trials.

52% when you get it.
-34% when the dealer gets it.
This gives 18% on two trials for a net of 9% per trial.

[aceside]

What Snyder meant was: When tracking a specific Ace, using key cards for location and alone with the dealer. If you're not 100% (and you rarely are) sure of the position the Ace will land, keep it simple and play only one hand. By doing this, you will "split the Aces" with the dealer, giving you an overall edge of approximately 9% on each of these trials.

52% when you get it.
-34% when the dealer gets it.
This gives 18% on two trials for a net of 9% per trial.

I generalized this process to the situation of known Ace/Ten ratios. For now, I am in the process of experimenting my new card counting strategy. Thank you for your hard work.

[21forme]

I will say that, for someone who posts something incorrect in virtually every thread that you participate in, you aren't shy. But, that isn't necessarily a good thing.

+ infinity

[bjarg]

What Snyder meant was: When tracking a specific Ace, using key cards for location and alone with the dealer. If you're not 100% (and you rarely are) sure of the position the Ace will land, keep it simple and play only one hand. By doing this, you will "split the Aces" with the dealer, giving you an overall edge of approximately 9% on each of these trials.

52% when you get it.
-34% when the dealer gets it.
This gives 18% on two trials for a net of 9% per trial.

I know, but when he says this he is assuming that the alternative to playing one hand is playing multiple hands also with max bets out, and this is not the only alternative, since you can place your max bet in the box where you have the ace tracked and play minimum bets in all other boxes, either for steering purposes, or to cover that max bet by reducing the chance of the ace going to the dealer.
This is something that every single sequencing team I've seen play does.