Are roulette wheel flaws trackable?

By Mark Lasser

(Second of two parts)

Last issue we got into the difference between single- and double-zero roulette and basic betting etiquette. We discussed inside and outside bets and basic payoffs. I also explained the important differences between felts that list payouts as being “for” and “to.” This week we’ll look at betting systems and biased wheels.

In a perfect world, with a perfectly randomized roulette wheel, any given number on the wheel has the same likelihood of coming up. Of course, that doesn’t stop gamblers from believing that a number is “due.” That’s one reason the casinos often have an electronic display next to the wheel showing results. Players will see that maybe “black” numbers came up four times in a row and think that “red” is due and either bet it or increase the size of their initial bet. This is important

In the very long run, it is true that all the numbers and bets will occur at the expected rates, but that in no way has anything to do with what happens on an individual spin. The wheel has no memory. In statistics, we call roulette spins and dice rolls independent events. The previous results have no bearing on the next or any other future results. Blackjack and poker are different because when cards are removed from a deck, it affects the probability of future outcomes. We call this conditional probability.

So, you may be asking, do we live in a perfect world? No, we do not, but I’m not sure how imperfect the roulette world is. Marten Jensen, in his book “Secrets of Winning Roulette,” describes roulette wheels as vulnerable to flaws that can bias the outcome to favor certain results. Among the things that can bias a wheel are wear and tear of the bowl, causing the ball to drop in the same place every spin; dealer accuracy of spinning the ball; worn pocket pads and separators that can cause a ball to drop into certain pockets more than others; and even an unbalanced wheel. This can benefit a player who can identify which section of the wheel is more prone to winning numbers than others.

One way designers try to thwart this is by making the numbers on the wheel non-consecutive. In other words, the wheel does not have consecutive numbers around the wheel head. However, there is, despite the best efforts of randomization, one section of the wheel that is almost grouped together, and that can be found in the central of the three betting columns.

Identifying a biased wheel takes patience and a certain degree of brazenness. The idea is that you would record about 400 spins of a wheel without betting and look to identify certain trends. For example, you would expect any given number to come up 1 in 38 spins. So, in 380 spins you’d expect each number to come up 10 times. But let’s say you observe a number coming up 10 times in only 200 spins. Well, that might look like bias. If the numbers to either side of this number are also hitting above average, like seven or more times, then that could indicate a biased wheel.

Here’s my issue with this: I just can’t imagine sitting in a casino with a little chart, clocking spins for a few hours and not looking ridiculously suspicious. If the casino is quiet, you’d be incredibly obvious, and if it’s crowded, you might have trouble finding a place to stake out the wheel while recording results. If casino security came over to you, what could you possibly tell them that wouldn’t sound awkward and suspicious. Let me know if you decide to try this and succeed. I’d like to hear firsthand accounts.

The other ways that people can get an edge on roulette are team-playing the wheel, with one person learning how to visually track the ball, while the other places bets; and by using computer programs to analyze where the ball is likely to drop. Like wheel tracking, I think both of these would attract suspicion, and the latter may be illegal if you happen to be deploying a mechanical device.

If you do decide to play roulette, there are a number of betting systems that are better than the Martingale system that I’ve previously been opposed to. These include the D’Alembert system and the Reverse D’Alembert system, which are better but not good systems. There’s also a decent but not perfect betting system called the Labouchere system that can be managed better with more limited bankrolls. Look these up online or wait a few issues and we’ll cover them in more detail.

Mark B. Lasser is Denver writer and international poker player. He regularly plays in Colorado, Arizona, California, Missouri and Nevada. You can hear him talk about gambling and casinos every Friday at 5 PM on KEZW AM 1430. Readers can send questions and comments to him at ColoradoPokerMark@comcast.net.