Blackjack Strategy

[Q_and_A]

Mathematically, you can expect to see more blackjacks dealt from a single deck than would be dealt from multiple decks. This is a mathematical truism even before any cards are dealt.

[Q_and_A]

Oddly enough, the continuous shuffle program used by most online sites results in blackjack ratios closer to a single deck than a multiple deck blackjack ratio.

[Q_and_A]

As a player, you want the highest ratio of blackjacks. This is true even though the dealer's chances for blackjack are the same as the player's chances. The dealer can only win or tie. The player can either tie or be paid MORE than even money (unless the player accepts even money, just like buying insurance).

With the higher blackjack ratio for single deck, the typical casino only offers 6/5 on a singledeck blackjack rather than the normal 1.5/1 payoff.

[MicroBob]

the 6/5 payoff on single-deck BJ's has only become common in the last few years.

[Sandra Bullett]

[ QUOTE ]
Oddly enough, the continuous shuffle program used by most online sites results in blackjack ratios closer to a single deck than a multiple deck blackjack ratio.

[/ QUOTE ]

Blackjacks can be dealt in two ways: a ten then an ace, or vice versa.

Chance of a blackjack being dealt from a single deck is:

10+A: 16/52 * 4/51 = 0.02413
A+10: 4/52 * 16/51 = 0.02413 (the same, obviously)
Total: 0.04826 (or 1 in 20.72).

Chance of a blackjack being dealt from an 8-deck shoe is:

10+A: (16*8)/(52*8) * (4*8-1)/(52*8-1) = 16/52 * 31/415 = 0.02298
A+10: 0.02298
Total: 0.04597 (or 1 in 21.75)

Note 1: The above assumes that the cards are perfectly shuffled. This is an entirely reasonable assumption.

Note 2: The ONLY reason that the chance of a blackjack changes when the number of decks changes is because, after the first card has been dealt, its non-replacement alters the probabilities for the second card.

There is NOTHING about an on-line continuous shuffler program which affects any of the above. If that program uses 8-decks (which a large number of them do), then the chance of a BJ will be identical to that produced by an 8-deck physical shoe in a casino.

The poster's comment that I have quoted above is wrong, unless he can prove that such shuffling programs somehow put Aces in closer than random proximity to tens, and vice versa.

[Q_and_A]

See Wizard of Odds/Blackjack/Appendix 10.
"This appendix shall attempt to answer the question about the effect on the house edge of the continuous shuffling machine (CSM). The continuous shuffling machine is a device that randomly inserts discards back in the deck. With one in use it is like playing against a freshly shuffled shoe every hand. This machine is not to be confused with an automatic shuffler that shuffle and entire deck or shoe.

Although the CSM gets a lot of criticism from players the truth is that it actually lowers the house edge. To prove this I ran almost large simulations both with and without a cut card under typical blackjack rules. The following table shows how much the house advantage is reduced by the use of a continuous shuffler compared to a cut card game."[end quote]

I have only read second hand that online sites predominately use continuous shufflers. I have not personally surveyed any casino TOC or or site shuffle proceedure so my premise may be wrong.

[kyleb]

Sandra,

The CSM results in a lower HA for the casino. Please see Shackelford's work on them for proof.

[Sandra Bullett]

Casino CSM's are mechanical devices that have several and quite severe constraints on their ability to shuffle randomly (eg. mechanical reliability, wear-and-tear on the cards, card reappearance latency, etc).

CONTINUOUS Software shuffling programs have no mechanical constraints, and very few constraints of other types (I guess if they shuffled very, very fast (eg. a million times a second) they could end up using too much CPU power, but that is easily fixed simply by doing one random shuffle before each card is dealt. It'd no longer be "continuous", per se, but that's semantics only).

[FWIW, Non-Continuous software shuffling programs, as used in most on-line casinos, have even fewer constraints. They typically have (in software form) a cut-card very near the front of an 8-deck shoe (eg. 35 cards into the 416 card shoe). When the cut-card is reached, the hand is finished, and a completely random shuffle is executed.]

The post of yours that I said was wrong was,

[ QUOTE ]
Oddly enough, the continuous shuffle program used by most online sites results in blackjack ratios closer to a single deck than a multiple deck blackjack ratio.

[/ QUOTE ]

Your statemetn dealt exclusively with "continuous shuffle program used by most online sites". You then use the wizardofodds site's analysis of physical CSMs as proof. I doubt very much whether the writers of online CSM's have attempted to reproduce the way physical CSMs are built.

I know how many types of physical CSMs work. Writing a program to reproduce their logic would be far more difficult than writing a clean, simply, true continuous shuffler. Physical CSMs have all sorts of engineering problems that are solved/ameliorated in ways which are simply unnecessary, and are overly complicated, in a program.

Most Physical CSM's do not, in fact, CONTINUOUSLY shuffle. What they do is when the previous hand's discards are put into the input hopper, randomly place each of these cards. If there are no new cards to 'digest' (in the hopper), and no cards being dealt (dead table) CSMs are usually totally inactive. They are not "continuously shuffling". There are several real-world engineering reasons for this, none of which apply to a computer program which can, very easily, be continuously shuffling.

Unless you can show that on-line sites have gone to the totally excessive, unnecessary and sub-optimal extent of exactly duplicating how the mechanics of a physical CSM works (thereby making the wizardofodds reference valid), then I will repeat my statement that your assertion is wrong.

[Banks2334]

[ QUOTE ]


Lets say you bet $1 and lose.
If you bet $2 and win you are now even.



[/ QUOTE ]
Um...You would be up $1. Good luck with your Martingale/Progressive/Busto system genius.

[DpR]

[ QUOTE ]
Casino CSM's are mechanical devices that have several and quite severe constraints on their ability to shuffle randomly (eg. mechanical reliability, wear-and-tear on the cards, card reappearance latency, etc).

CONTINUOUS Software shuffling programs have no mechanical constraints, and very few constraints of other types (I guess if they shuffled very, very fast (eg. a million times a second) they could end up using too much CPU power, but that is easily fixed simply by doing one random shuffle before each card is dealt. It'd no longer be "continuous", per se, but that's semantics only).

[FWIW, Non-Continuous software shuffling programs, as used in most on-line casinos, have even fewer constraints. They typically have (in software form) a cut-card very near the front of an 8-deck shoe (eg. 35 cards into the 416 card shoe). When the cut-card is reached, the hand is finished, and a completely random shuffle is executed.]

The post of yours that I said was wrong was,

[ QUOTE ]
Oddly enough, the continuous shuffle program used by most online sites results in blackjack ratios closer to a single deck than a multiple deck blackjack ratio.

[/ QUOTE ]

Your statemetn dealt exclusively with "continuous shuffle program used by most online sites". You then use the wizardofodds site's analysis of physical CSMs as proof. I doubt very much whether the writers of online CSM's have attempted to reproduce the way physical CSMs are built.

I know how many types of physical CSMs work. Writing a program to reproduce their logic would be far more difficult than writing a clean, simply, true continuous shuffler. Physical CSMs have all sorts of engineering problems that are solved/ameliorated in ways which are simply unnecessary, and are overly complicated, in a program.

Most Physical CSM's do not, in fact, CONTINUOUSLY shuffle. What they do is when the previous hand's discards are put into the input hopper, randomly place each of these cards. If there are no new cards to 'digest' (in the hopper), and no cards being dealt (dead table) CSMs are usually totally inactive. They are not "continuously shuffling". There are several real-world engineering reasons for this, none of which apply to a computer program which can, very easily, be continuously shuffling.

Unless you can show that on-line sites have gone to the totally excessive, unnecessary and sub-optimal extent of exactly duplicating how the mechanics of a physical CSM works (thereby making the wizardofodds reference valid), then I will repeat my statement that your assertion is wrong.

[/ QUOTE ]

I did not read the link refered to, but it is possible the edge is linked the the "cut card effect," which would have nothing to do with the mechanics of a CSM or computer shuffle.