Roulette dealers get into a rhythm. They pick up the ball from the pocket where it last hit the same way each time. They then spin it the same way each time, with the same force. Thus, the ball will spin the same number of spins and should land the same number of pockets away from the last spin. This unconscious ability or trait to place the ball, not so much in a given section, but a given distance from the last hit is called dealer signing or a dealer signature.

Experts have lined up on either side of the dealer signature issue -- some yea, some nay. Essentially, it boils down to a matter of belief. Some experts believe that some dealers unconsciously do this; some experts believe there is no consistency to the distance that the ball travels from one spin to the next and, with the changing rotor spin over time, it would be impossible to achieve consistency anyway.

However, the dealer signature is an interesting concept and could have some credence because -- one: it makes sense, and two, like Mulder on The X-Files, we want to believe it?s true. Of course, if one could follow a thousand dealers and record their spins over thousands of decisions, it might be possible to determine once and for all whether the signature idea is a myth or a reality. To this date, we know of no such extensive study as this. The actual accomplishing of such a study would be much more difficult than just the recording of thousands of dealers and thousands of their spins.

Since we are looking for an unconscious ability that these dealers might have, we wouldn?t want them to know that we are studying them because that would affect their spins. Once dealers realized that we were looking at them specifically, they would become highly conscious of what they were doing. If any signature actually existed with such a dealer, their awareness of their spinning technique would very quickly erase whatever signature they had. So the study of possible signatures would have to be done surreptitiously, without the dealer noticing, and there?s the rub. How could a researcher stand by a dealer?s table, follow the dealer from table to table, hour after hour, day after day, recording and analyzing his spins without the dealer becoming aware of such a person? At first the dealer might think that the wheel was being clocked, but it wouldn?t take long for him to suspect that it was his spins that were being clocked. Even if the dealer thought it was the machine being clocked, this still might cause him to alter his spin.

Now, wouldn?t a biased-wheel watcher also affect a dealer?s spin? (A biased wheel is one with a physical defect of some type that causes certain numbers to appear out of proportion to their probabilities.) Yes. Certainly a dealer would eventually become aware of someone standing near his table recording thousands of spins to ascertain whether the wheel was biased. Yes, it might make the dealer self-conscious. Yes, the dealer might change how he spins the ball. But, in truth, none of this would matter. On a biased wheel any number of dealers will spin the ball with any amount of force, causing the ball to make any number of revolutions around the wheel, and all this wouldn?t affect the bias one whit because that bias was in the wheel itself and had nothing to do with the spinning of the ball or the dealer. But anything that is dealer-dependent and dependent on the dealer being unaware is immediately changed when the dealer becomes aware. In this case, as in Quantum Mechanics in physics, the observer interferes with the observed by the mere fact that he is observing! And good-bye dealer signature. So I doubt if an actual practical test of dealer signatures is workable in the real world of casino play.

The other way to look at it is that dealer signatures must necessarily vary as the wheel speed slows down over time. The dealer might spin the ball the same way time and again but with each ball-spin he is playing into a fractionally different wheel-spin. Thus, dealer signatures would move slowly around the wheel as the wheel itself slowed. When the wheel was respun and thus accelerated, you would see the same patterns as last time. You would have to then analyze the pattern of the signature -- an even more complicated task.

Still, if dealer signatures exist, certainly they would be exploitable in long and short-term play, especially short-term play as the gradual slowing of the wheel?s spin would not affect the signature quite so drastically. If the dealer signature exists, then it will help us to win. If it doesn?t exist, it can?t hurt us -- since it isn?t increasing the house edge against us. We merely face the same house edge we would have faced had we played any layout strategy. Still analyzing signatures is a technique of play that might be able to change the edge in our favor. It?s worth a try.

In any series of decisions, it is easy to find the ?average? distance of a dealer?s spins from decision to decision. Just add up the distance of each spin and divide by the total number of spins. Unfortunately, this is not the same as finding a signature. What we want to know is if the dealer?s average is within the confines of, say, one-third of the wheel. That is to say, will the dealer spin the ball in such a way that it tends to land within a nine to 12 pocket grouping consistently?

Looking at the double-zero roulette wheel, let us say that on spin number one the dealer picks the ball up from the 00 pocket and spins it so that it winds up in the 12 pocket, five pockets away. On his next spin, he picks the ball up from the 12 pocket and spins it so that it lands in the 6 pocket, five pockets away from the 12. Finally on this third spin, he lands it in the 23 pocket -- another five pockets away. The dealer?s signature here would be five. Of course, three spins of the wheel is not enough to determine without a doubt that a dealer has a signature, but for purposes of a glaring example it is sufficient. Since the last number that hit was 23, we would now bet on 9 -- five pockets away. In the best of all possible worlds, the dealer would again spin the ball in such a way that it would land in our 9 -- five pockets away.

In the real world, as opposed to the world of our roulette dreams, we would never see a perfect signature. What we would want to find is a dealer who places the ball more than a third of the time within a 12 pocket grouping an average distance away. Let us say that the dealer is able to hit a 12 number grouping, one half of the time. In this way, we would bet those 12 numbers (12 units), lose half the time (-12 units), win half the time because one of our 12 numbers hit (which means we lose 11 units on the numbers that didn?t hit) but win 35 units on the number that did hit for a net win of 24 units (35-11 = 24). Therefore, in two spins we are ahead 12 units or six units per spin. We would soon own the casino. Naturally, we would take any kind of win, even one unit per spin. Thus, if the dealer were able to hit our 12 pockets three times every eight spins we would average 1.5 units per spin as a win (lose 12, lose 12, lose 12, lose 12, lose 12, win 35-11, win 35-11, win 35-11 = 12 units = +1.5 units per spin).

Some of Frank's most popular books can be purchased by clicking one of the image
links below: