The Invasion of ShuffleMaster

ShuffleMaster shuffle machines are becoming more of a reality every day. In the first article on this subject from Boris' Advanced Blackjack Journal #3 (re-published on this website), we took a detailed look at the ShuffleMaster I and II machines (for single and double-deck games, respectively). We also considered a few things you can do to take advantage of the non-random cards produced by these machines. In this issue, we are going to take a similar look at the ShuffleMaster "BOX" for today's shoe games. Similarly, we will examine things that you can do to take advantage of the cards produced by this machine.

If you have paid much attention to the multi-deck shoe games in today's casinos, you have probably noticed dozens (if not hundreds) of table games which sport an ominous black box, sitting on a shelf mounted behind the table; usually on the left side. This particular unit, manufactured by ShuffleMaster Corp., to my knowledge has no name (previous units were known as: ShuffleMaster I and II). For discussion purposes, I have given it a name: "ShuffleMaster BOX ".

ShuffleMaster BOX is an ingenious device; especially when you take a closer look at how it was designed. My background with this machine began as a one-on-one interview with a ShuffleMaster V.P. of Sales (who also seemed to know a great deal about card shuffling), followed by a detailed study of the Unit's patent diagrams (obtained from the U.S. Patent Office). Admittedly, the ShuffleMaster people gave me information about the inner-workings of the unit not specified in the patent disclosure.

Because it is easy to get bogged down with insignificant-significances (an invented phrase I borrowed from Introspection Therapy), the purpose of this Article is to take you on a fascinating tour of the BOX's main points. Later I will offer some Insight into exploiting some of the BOX's anomalies. If you have not yet read the previous article on ShuffleMaster I and II, you may well find it to be useful supplementary information.

To establish a point of registration, let's review some of the casino benefits that ShuffleMaster "brings to the table". You will remember from the previous article that ShuffleMaster corp. is on record as claiming that amongst other advantages, ShuffleMaster machines allegedly produce random cards (ostensibly to thwart card counters) and offer a hands-per-hour increase of up to 40%. I went on to point out that, in actual fact, the cards coming out of the ShuffleMaster are FAR from random (per the Clump-tracker's definition of the term), although the 40% hands-per-hour increase is indeed accurate.

Additionally, cards coming out of ShuffleMaster I & II fail the mathematical test for randomness which requires that each deck be riffled AT LEAST seven (7) times (2 decks s/b riffled 14 times). These machines riffle 4-6 times (single-deck) and 6-8 times (double-deck). Additionally, the "riffles" created by these machines are unbalanced, further decreasing the randomness of the cards, producing even more like-card clumping. As it turns out, with ShuffleMaster Box, there is a similar lack of randomness, in part due to a "forced" (i.e. intentional) un-shuffled clump of cards during each riffle-step

Because the BOX-unit "riffles" the cards only 5 or 6 times, ShuffleMaster's claim of card randomness can be easily questioned; especially when you consider (from the above-mentioned shuffle-randomness figures) that in order to be considered a "random shuffle", the machine should riffle the cards at least 7 times.

Unlike ShuffleMaster I & II which use mechanical "fingers" to reasonably simulate a riffle, the Shuffle Master Box uses a series of "elevators" instead. When it comes time to shuffle, the dealer loads the stack of cards into the center elevator and pushes the red button atop the box. This closes the glass window and the shuffle procedure begins.

To begin the shuffle procedure, the BOX randomly determines whether this shuffle will utilize a "tighter" intertwine (1 to 5 cards per "drop") or a "looser" intertwine (5 to 10 cards per "drop"). There is no way to externally know which kind of intertwine was selected, unless possibly, you are capable of [clump] "reading" the cards; and even then possibly not. ShuffleMaster corp. claims that the intertwine selection is COMPLETELY random and is NOT selectable by casino personnel.
(As an aside: a drop of 1 to 5 cards is roughly equivalent to the dealer taking 3/4 deck picks during the shuffle process, while a drop of 5 to 10 cards is roughly equivalent to taking 1-deck picks.)

Once the intertwine is selected, the SM Box lifts the center elevator as it "fires" the first half the cards down the right hand elevator shaft, followed by the other half of the cards down the left elevator shaft; effectively splitting the cards into two stacks.

Next, a random number (from 5 to 60) is generated and this number of cards is fired from the right elevator into the center [stack]. This is the equivalent of the un-riffled 25% of the right-hand pick produced by ShuffleMaster I & II. Remember that ShuffleMaster corp. claims that these un-riffled clumps of cards (their term, not mine) are necessary in order to insure that the cards are random. Amazing! Non-random cards are necessary to insure that the cards are random. Huh?
(As a side note, even the patent disclosure use the word "Clump" to label this step in the shuffle procedure.)

Once the initial clump is created, ShuffleMaster Box alternates from left elevator to right, firing cards onto the center stack based on the intertwine selection. When both outer elevators have been emptied into the center, the entire process is repeated, using the same intertwine previously determined. The shuffle process is performed 5 or 6 times (randomly selected, independent of the loose/tight intertwine selection). Finally, the cards are reverse moved to the left elevator shaft completing the shuffle procedure.

For the each subsequent shuffle, cards to be shuffled are stacked in the center elevator. To begin the shuffle procedure, the previously shuffled cards are removed from the left elevator and the [RED] start button is pressed.. This closes the glass door and the above-described procedure is invoked.

Essentially, that's all there is to the ShuffleMaster Box procedure. Even BEFORE you study the cards produced by this machine, the above-mentioned description should make it rather obvious that the Shuffle Master BOX does not REALLY produce random cards.

I mentioned in Part I of this article-series that card-counters HATE these shuffle machines. This is even more-so with the Shuffle Master Box. In contrast, much of the time I LOVE the games produced by these machines; although admittedly, when I first encountered one of these boxes (at the Golden Nugget in Laughlin), I didn't do well at all. That's what practice is for. Fortunately, the Boris-for-Blackjack Simulation Software provides me with an accurate ShuffleMaster BOX Simulation to practice against.

Because the cards produced by ShuffleMaster are USUALLY reasonably non-random, these machines should produce clump-track favorable games, and in fact, they often do. ShuffleMaster BOX games are not as non-random as one might find in, say, Atlantic City hand-shuffled 8-deck games, however they compare favorably (from the clump-tracker's viewpoint) with most of the non-random games found virtually anywhere in Las Vegas. Because ShuffleMaster BOX games are often so readable, in the "olden days", at casinos which offered BOTH types of games (Ex: Aladdin and Bally's Vegas in the late 90's), I would usually PREFER the ShuffleMaster games; altho there were exceptions.

Double Deck and the ShuffleMaster BOX
On another note, in early 1998 I was wandering through the New York, New York casino and noticed an interesting oddity: a ShuffleMaster BOX was being used to shuffle their double-deck games. I also noticed that during this time-period the Soft-17 rule was essentially the inverse of what we normally encounter: (2-Deck) - Dealer Stands on Soft-17; (6-Deck) - Dealer draws to Soft-17.

When I asked the Floorman about the rather unique use of the BOX, his response was that it saved them money from not having to lease an additional ShuffleMaster II for their double-deck games - an expected response. However, another reason, came to mind; ShuffleMaster BOX might (from the casino point-of-view) produce less-exploitable double-deck games than ShuffleMaster II. Setting up a special casino environment simulation with the Boris-for-Blackjack simulation software, bears this out. I will have more to say about this in a future article

Double-Shuffling with ShuffleMaster BOX
On a play trip to Reno in November 98, I encountered an interesting situation at two different casinos: Dealers were instructed to run the cards through ShuffleMaster BOX twice. The downside of this double-shuffle is that it introduced approximately 2 minutes of downtime; the very thing ShuffleMaster was designed to eliminate. When I asked the dealers why this was being done, I received the same programmed response: "Some guy was caught [sic] cheating last week". Even if players WERE cheating, a double-shuffle would in no way prevent this. By "cheating", the dealers probably meant card-counting or card-tracking. (While Nevada courts have ruled that card-counting does not constitute cheating, many casinos treat counters as if they ARE cheating.)

My professional instinct tells me that there is a little more to this double-shuffle situation than might at first be obvious. The next article in this series details the results of simulations run with the Boris-for-Blackjack simulation software. The results turned out to be quite surprising.

Bright Lights and Masking Tape
Over the years, I have taken notice that a sort of paranoia abounds with regard to the BOX Shuffler. My first play against the ShuffleMaster BOX was in May-1998 at the Golden Nugget - Laughlin. Normally, the Golden Nugget takes a rather loose approach to their Blackjack games. However with ShuffleMaster BOX, things were a little different.

To begin with, casino execs in a clever move physically placed the machines in such a way as to reflect an overhead floodlight; precisely aimed at the glass front of each machine (there were two "BOX" units positioned essentially back-to-back on opposite sides of the pit, each matched with an "unusually placed" floodlight).

Months later they extended this "disruption" to include a piece of masking tape across the top of the BOX's glass front. Both approaches tend to obscure visibility of what is going on inside the box. I have concluded that was the intent.

I have seen the masking tape trick in other casinos sporting the BOX shufflers. Only the Golden Nugget in Laughlin has (to my knowledge) used the lighting trick. That such "modifications" are being made suggests to me that casinos doing this have something to hide; specifically, that alert players can observe the machine shuffling one shoe while they play the current one; one of the reasons I often like to sit in a seat at the Blackjack table offering me a relatively un-obstructed view of what is going on inside the machine. Sitting in a seat NEXT to the BOX allows me to count the "riffles" by sound, rather than visually.

While the use of masking tape and special lighting tend to interfere with such observations, luckily, Clump-trackers have other avenues available to win against these machines; some of which are documented in Part III of this article series.

Taking advantage of ShuffleMaster BOX
Depending on who you talk to, the ShuffleMaster BOX is either the best thing that ever happened to Blackjack, or the worst. While Blackjack players sport varying opinions on their ability to exploit these machines, it is generally agreed upon by most attentive BAC players that the BOX produces what they call "choppy" games.

To understand how Blackjack players can take advantage of these machines, we should consider the factors which contribute to the [non-]randomness of just-shuffled cards.

Randomness of shuffled cards is influenced by a number of factors:

1. The order of the cards in the discard tray.
2. Whether or not the un-played cards are "plugged" into the discard stack before the entire stack is transferred to the shuffle machine.
3. Size of the un-shuffled clump at the beginning of each "riffle" step.
4. The chosen intertwine for the entire procedure (Loose or Tight).
5. The number of riffle steps used for a given shuffle (5 or 6).

Depending upon the contents of the un-shuffled cards, plugging may or may not contribute to the eventual [non-]randomness of the cards. Shuffle-trackers and some card-counters may be able to deduce this, based on the count at the shoe's end.

The size of the un-shuffled clump during each riffle will of course effect card-randomness. In general, the smaller the un-shuffled clump, the more random we can expect the cards.

Shuffle intertwine has a pronounced effect on card-randomness. In general, the tighter the intertwine, the more random we can expect the cards. That loose intertwines contribute to card-clumping we already know from studying the effect of different pick-sizes during hand shuffling.

Finally, the number of riffles of course affects card-randomness. More riffling produces more random cards. Fewer riffles (especially with a loose intertwine) tend to produce more pronounced card-clumping .

Unlike with Shuffle-tracking, Clump-trackers and conventional card-counters cannot know in advance what card-sequences to expect during play. This does not however mean that counting and clumping are not effective methods of play against ShuffleMaster BOX . It simply means that such players will have to determine the degree of clumping as each shoe progresses.

Dual Shoe Signatures with ShuffleMaster BOX
I believe it was E.C. Davis by way of NBJ who first documented the concept of a shoe "signature". While criticized heavily as being unscientific, Davis based the concept of shoe signatures on the "local" Tens-ratio. You may recall that keeping a Tens-ratio was one of the first (if not THE first) approaches to card-counting. In essence, Shuffle-trackers also track a sort-of shoe signature.

A difficulty with tracking signatures in ShuffleMaster BOX games is due to the fact that there are always two sets of cards to deal with; one set being shuffled while the other is being played. Fortunately, in most cases, the cards are different colors; Ex: Red & Blue. This means that should a [un-]favorable signature be noted, it will not recur until the NEXT shoe after the one being played (while the "favorable" cards are being shuffled).

For example, if a favorable signature is noticed while playing against the red-backed cards, we will have to wait for the re-emergence of those red-backed cards in order to have another shot at it. The blue-backed cards will most-likely have a different signature altogether.

Some players "solve" this problem by taking a break during the play of the blue-backed cards, returning in time for the shuffle to play the red-cards. Playing against one set of cards and not the other may be tantamount to the heat-inducing method of playing against one dealer (who is say, exposing the holecard, or mis-paying favorably) and then taking a break during the dealer's 20-minute off-period (during which time a relief dealer is brought in). I will discuss further implications of this problem in another article.

Some final thoughts on ShuffleMaster BOX
Like it or not, shuffle machines are here to stay - we may as well get used to them. As we have seen, ShuffleMaster BOX is unique amongst the many shuffle machines on the market. If you develop playing methods to take advantage of the unit's anomalies you too will appreciate the up-to 40% increase in hands-per-hour.

The secret to survival against any shuffle machine is to keep your eyes open for nuances that may exist or develop over time. We have been doing that with hand-shuffled games for nearly 10 years. There is no reason we can't do that against machine-shuffled cards. I trust that this series of articles serves as a step in that direction.