The Invasion of ShuffleMaster

Shuffle Master shuffle machines seem to be popping up in a number of casino card games these days. Because they are having such a dramatic impact on the quality of the games they are being used in, it is time for us to take a closer look at these machines, how they work, and how they impact the games we play. We will end each examination with ideas on how to exploit these machines for fun and profit.

In this issue we will take a look at Shuffle Master[I] (for single deck games) and Shuffle Master II (for double deck). In the next issue we will look at the Shuffle Master "Box" (for 4, 6 & 8 deck games), as well as an exercise that you can do to duplicate the 1 & 2 deck Shuffle Master shuffle at home. You will be amazed at what you see.

Casino executives are always looking to increase the drop and therefore the table hold. Dealers are implored to deal faster, and newer (hopefully faster) shuffle procedures are introduced from time to time. In recent years, electronic peeking devices have been introduced to check for dealer naturals without exposing the holecard to the players. Turning over a dealer Blackjack without first playing out the rest of the hands improves the hands-per-hour ratio even further.

In 1992, Shuffle Master Corp. introduced the first practical shuffle machines to the casino industry. While they are hardly the first shuffling devices in use (you may remember the clunky shuffle boxes in use at the Golden Nugget in downtown Vegas, later moved to the Mirage before finally being discontinued) Shuffle Master machines are the first to gain widespread popularity. Today, we are finding Shuffle Master I & II (for single and double deck games respectively) throughout Nevada, as well as casino locations across the country; even around the world. Shuffle Master has even invaded Atlantic City.

In addition to Blackjack games, Shuffle Master machines are to be found at a majority of Caribbean Stud games and ALL Let It Ride games, being bundled with the game by the Shuffle Master Corp., the inventor and licensor of the game.

Although Shuffle Master Corp. promotes many benefits to using these machines, the most overtly obvious benefit (from the casino point-of-view) is a virtual elimination of shuffle down-time. Different color decks are used. For example, while the red deck is being shuffled, the blue deck is in play at the table. This offers a hands-per-hour increase of up to 40%. More hands per hour translates into greater profits for the casino. With a near 40% increase in throughput, most casinos recoup the $500/unit monthly rental fee in 1-3 days!

There is another aspect to these machines, which although denied by Shuffle Master corp. is of considerable interest to us. In their sales literature for the devices, Shuffle Master corp makes considerable claims that their machines produce randomly shuffled cards and therefore are a deterrent to card counters and shuffle trackers. Since the first day I saw one of these machines at Bally's Vegas, it has been clear to me that cards coming out of these machines are HARDLY random.

Despite claims to the contrary, BOTH Shuffle Master I & II produce non-random games; at least in comparison to what one would expect of a 1 or 2 deck game. Then again, this non-randomness is not so surprising when you understand how the machines work. In case you are interested, methods for determining the relative clumpingness of a game were detailed in Issue 2 of this Blackjack Journal.

As you may know, the Boris Casino Blackjack Simulator is the ONLY computer software to include the Shuffle Master machines in the available casino selections. To program these shuffle machines, I did a thorough study of the subject. The information I used to develop the shuffle algorithms came from observing dozens of different tables sporting a Shuffle Master machine, followed by a lengthy interview of several representatives of the Shuffle Master corp. over the phone and in person at the World Gaming Congress. Later, I verified many of their claims and descriptions after detailed studies of a half-dozen U.S. patent filings for the Shuffle Master machines themselves. At the congress, they gave me quite detailed demonstrations of the machines in action.

With the Shuffle Master II, we were able to shuffle red/blue decks together allowing me to examine the results. What I observed was quite astonishing. A close examination of both the Shuffle Master I & II machines reveal that although the cards are split in more or less equal-sized picks, the top 50% of the right pick is NEVER intertwined during each riffle. This creates, in effect, an unbalanced shuffle; even more devastating than the infamous Barbary Coast double deck shuffle of times past (they now use Shuffle Masters almost exclusively).

After each riffle, Shuffle Master cuts the top half of the cards to the middle (bottom). In effect, this results in 50% of the cards being riffled only 50% of the time. Shuffle Master I riffles the cards 4 to 6 times (although Shuffle Master people deny it is as little as 4). Shuffle Master II riffles the cards 6 to 8 times.

Reputable studies have demonstrated that a minimum of 7 riffles is required to randomize a single deck of cards. Another reputable study claims that 25 riffles are required to completely randomize a deck. As you can see, NEITHER Shuffle Master I or II pass the test for randomness because: A) the number of riffles per deck is insufficient; B) a thorough intertwine is not utilized.

Washing the Cards

To begin with, most casinos introduce new decks with the card order:

A-->K	    A-->K	K-->A	    K-->A

Usually, the deck is hand shuffled once or twice and then placed in the Shuffle Master. Let's watch the order of the cards as this is done. Because single-decks are easily riffled, we will assume this achieves a near-perfect 1-to-1 card intertwine, in which case we will end up with the following result:
A-K-2-Q-3-J-4-T-5-9-6-8-7-7-8-6-9-5-T-4-J-3-Q-2-K-A-A-K-2-Q-3-J-4-T-5-9-6-8-7-7-8-6-9-5-T-4-J-3-Q-2-K-A

Notice the ready-made clumps of neutral cards and the area of tens-concentration (in the middle of the deck). Although not a perfect clump of ten-value cards, it is enough to destroy the 1 ten for every 3.25 card distribution that we would find in a "random" deck. If the dealer shuffles one more time we end up with the following:
A-A-K-K-2-2-Q-Q-3-3-J-J-4-4-T-T-5-5-9-9-6-6-8-8-7-7-7-7-8-8-6-6-9-9-5-5-T-T-4-4-J-J-3-3-Q-Q-2-2-K-K-A-A

Shuffled a third time, we get: A-7-A-7-K-8-K-8-2-6-2-6-Q-9-Q-9-3-5-3-5-J-T-J-T-4-4-4-4-T-J-T-J-5-3-5-3-9-Q-9-Q-6-2-6-2-8-K-8-K-7-A-7-A

Granted, stripping the cards will break up the clumps considerably. However in the real world we aren't seeing it much with Shuffle master. The tendency is to let the shuffle machine do the work. Then the play of the cards will keep clumping alive; even in these single and double deck games.

What good is more hands per hour if the house isn't winning most of those hands? Remember that single and double deck games are dangerous for the casino if they are exploitable by card counters. While Shuffle master boasts that card counting is combated by the alleged random shuffle of their machines, it is well known that in randomly shuffled single-deck games Basic Strategy players play essentially even with the house (altho admittedly with great statistical variation). Lately, casino classes are essentially teaching Basic Strategy; the Hit/Stand aspects, at any rate.. If the Shuffle Master machines REALLY produced random cards, very little money would be made at these tables.

Luckily (from the casino point of view) this is not a problem. Not only has the table action increased by 35%, the table hold in Shuffle master games is considerably greater than that found in hand shuffled games. It doesn't take a rocket scientist to realize that these shuffle machines are quite favorable to the casino; more so than hand the shuffled games. As I mentioned earlier, the Barbary Coast has retired their highly lucrative unbalanced shuffle in their double deck games, switching exclusively to Shuffle master II.

The Boris software simulator has demonstrated conclusively how lucrative an unbalanced shuffle is. Casino executives would not sanction replacing a money-making procedure unless they felt its replacement would produce an even greater improvement of their bottom-line. Shuffle master II is such a replacement.

Taking advantage of Shuffle master

Thus far you may think I have painted a bleak picture regarding shuffle machines. It is the function of this journal to present the facts as they are, hopefully followed by some insight allowing you to take advantage of this factual information.

Recently I was making conversation with the proprietors of the Gambler's Book shop in Las Vegas. I purposely asked them what they thought of the new shuffle machines. John informed me that card counters hate the machines; specifically the 6-Deck box (described next issue). This didn't surprise me in the slightest. I let John know that I absolutely love those shuffle machines. Like every niche in the game of Blackjack, shuffle machines have their little tricks. Let me share some of them with you.

I first encountered Shuffle master [I] at Bally's in the spring of 1992; at a $2 table no less. I was card counting in those days and researching the Boris Clump-Count procedure. I played 3 sessions (2 evenings and one day), winning all of them. The first session found me playing 1st-base (the only seat open). Incredibly, I won 18 hands in a row. Had I been paying attention I would have been using a positive betting progression (such as a 1-1-2-2-3-3 stepladder), capitalizing on the streak.. Instead, because only two rounds were dealt per shuffle, I chose a 2 to 1 spread. Either way, there is nothing better than starting off a playing session with a sizeable win.

Originally Bally's located the shuffle machines to the dealer's left (now they are on the right). That gave me a close up look at the mechanism, although at the time, I was oblivious to the possibility of clumping a single-deck game. An hour into the game, the machine jammed spitting the cards straight up in the air. At first I thought it was a joke. The floorman brought in a new deck and the game played on. Later, this got me to thinking. If a table goes bad, possibly we can "encourage" the pit to bring in new cards to change our "luck".

When I play single deck, I usually hold my cards so everyone can see them. This encourages everyone else to do the same. Remember, we are playing against the dealer, not against each other. One way I hold my cards is to stand them on their side, facing me. It occurred to me that furtively pushing down slightly on the cards can give them a slight bow. I am careful to do this in such a way that I don't favor any particular value of the cards, lest I be accused of "marking" the cards. Enough bowed cards will eventually cause the machine to jam, requiring a new deck. I only do this if the deck is developing a considerable dealer-favorable bias.

Clump Control

To "control" the clumping, the question to ask is whether you want to encourage random cards, break-card clumps (described in the last issue), or cleaner clumps. For the most part, there isn't much you can do with pat hands. With the exception of hard or soft 17, 2-card pat hands are made up exclusively of high cards. With the exception of hands involving combinations of 6's, 7's and 8's, stiff hands are made up of a high card and a low card.

Originally for ease in adding up my hand, I used to swap the 2 cards placing the high card behind the lower card. Later it occurred to me that swapping the card order can encourage or discourage card-clumping, although is an untested theory. When you finish your hand, the cards are tucked under your bet. When the dealer later turns them up they will end up ON TOP of any hit cards. In most cases the cards are always scooped up in the same way. The rest I think you can infer for yourself.

E.C. Davis has suggested that if break-hands are left on the table (and therefore scooped up with the rest of the hands) that clumping would be all but eliminated. I intend to research this for a future issue of the journal. Assuming for a moment this is true, you may be able to get away with not turning in your hand with a value of 22 (and possibly 23), feigning a mis-total if the dealer makes a comment about it. However, don't overplay this maneuver. You don't want to call attention to yourself.

In issue #4 of this Journal we will take a close look at the Shuffle Master "Box". Stay Tuned.