Get the edge at roulette
Christopher Pawlicki

Another roulette related title I picked up out of curiosity. I fully expected this one to be more tosh published with the aim of relieving some of those that believe in the gamblers fallacy, or who are numerically challenged, of a few dollars in return for a short period of renewed hope. Having read through it in full, however, I was very pleasantly surprised to find this not to be the case, and between the covers there's a remarkable amount of "tell it as it is" objectivity around a punter's chances of findng an edge when playing against the wheel of doom, and how this might be achieved by your common or garden degenerate gambler.

Get the edge at roulette runs to 223 pages, around half of which are devoted to the history of roulette, a study of notorious betting systems and why they don't work and online roulette, which in 2001 was a lot less sophisticated than it is today - notwithstanding the fact that it hasn't fundimentally changed. For me, and probably most other seasoned roulette players and students of the game, this could all have been ditched and the book limited to the remaining subject matter, which are discussions around wheel bias, wheel tracking and picking up on "dealer signatures" in order to elicit an edge that's greater than the mathematical disadvantage built into the pay-table. If you also took a red pen to the 40 odd pages dedicated to "Techniques used by Professionals", much of which I suspect is unworkable in practice, what you're left with is some thought provoking stuff (taking up around a quarter of the book) discussing how roulette wheels are constructed, differences in component parts, what to look for in the way of consistent actions of individual dealers and how to test to see if recurring numbers and sectors may be due to any bias or not.

Does it stack up? Well, I think that with the precision builds, setups and maintenance of modern wheels of doom, spending time looking for a wheel where it can be proved statistically that something's amiss is a complete waste of time. In the past perhaps, but now - nah. However, I think the issue of dealers' signatures is a different matter, and these certainly do exist, although I've always been sceptical of claims by croups that they can hit sectors of the wheel on demand. A few days back, before I'd completed this book, I sat and watched the late night live roulette session on a cable channel, and the lady croup hit the same number (spinning from that number) on two successive clockwise spins, and two adjacent numbers on two successive anti-clockwise spins. Coincidence? Possibly. But when you consider all of the dynamics involved in spinning a number, all of these across her four spins appeared very uniform and consistent. However, the big "but" in all of this is all of those things that can upset that consistency. After this short run of results, the following ones were all over the place as a consequence of the the ball hitting a deflector en-route to the track and bouncing randonly across it before coming to rest. And therein lies the problem with all of this.

It only takes a very minor, possibly undetectable, change in the croup's initial launch speed of the ball, or an inch or so's difference in the launch point in the rim, to result in the ball catching a deflector on it's drop towards the rotor, after which it becomes nothing more than an agent of chance. There are eight deflectors on a wheel, four of which are set vertically, and hitting any of the vertical ones tends to redirect the ball due south towards the rotor which is a complete change from the regular degrading trajectory if it hadn't caught one. The horizontal deflectors have a similar, although diluted, effect. Let's face it, the deflectors have been built in to wheels for a very long time for a reason, and generally they do what they're supposed to.

From a maths point of view, the consideration around betting on sectors is fairly simple. If you're covering a nine number sector of the wheel with straight up bets, which is about a quarter of all of the options, you need to be hitting it (either through judgement or luck) once every four spins in order to breakeven; 4 x 9 chips bet = 36, a win on a straight up number is 35 + 1 (initial bet) = 36. If you're covering a six sector range, you need to hit it at least once every six spins. There does come a point, though, where a sector is so wide that you're not really betting with a (supposed) advantage at all; at the far end of the scale betting on a sector of 18 numbers, half the wheel, means you might as well be betting on red or black? I suspect the tipping point is around the 9 sector point, and if you're having to bet across a wider range than this, the exercise starts to be counter-productive; the more numbers you cover, the greater the amount that has to be risked each spin to cover the target sector and, therefore, the greater the cost of failure if things don't work out.

You can of course fall back on using the combination betting options to cover your target sector; so betting on the tight, four number, sector between nineteen red and two black inclusive (19,4,21,2) could be achieved by betting the one to six line bet and the nineteen to twenty-one street bet. This would require just two chips, have the potential to return 6 and 12 chips respectively and provides the potential for any of the other five numbers to pay out if the deflectors turn any particular spin into the lottery it's supposed to be. The trouble with diluting down the exposure in this way though, is that you're not really getting a return to reflect any advantage you think you may have hit upon - betting these four numbers straight up would mean that if you hit them more often than once every nine spins you'd be in front.

There are lots of other combinations of risk and return that could be discussed, but I'll leave all that up to you to explore if you have an inclination to look into this further. The book's still available from all major book-sellers, although the £19.95 that it's being offered at by Amazon.co.uk at the time of writing is well over it's orginal cover price of $13.95 - which works out at about nine quid at the current rate of exchange.

In conclusion, I'd say Get the edge at roulette is akin to a blueprint for machining up the components to create your own rifle. The explanations around the theory are sound, and the instructions on how to do it adequate to do the job. Put all of the components together, and what you end up with is a firearm that looks the part and will certainly get you nicked if you're caught with it without a firearms certificate. But the finished item will still be subject to all of the tolerance variations within the machinery used, may fail to zero to a suffiently precise degree to be of any use and when you load it with a round and pull the trigger it might just blow up in your face.

A final, final word? Identifying a consistent dealer signature may provide an advantage, albeit for a short time, but an advantage isn't a guarantee. By all means experiment with this, but do tread with care. I think it could be quite easy to throw a lot of money at a perceived bias where one doesn't actually exist and the results are nothing more than a case of the variance fairy having a bit of fun. Good luck.

August 2015.