Done beating Blackjack - now, onto Roulette!

Dear WoV,
It's been a long time since my last post. WoV inspired me to start focusing my gambling inclinations into advantage play. My first approach to AP was with BJ in the fall of 2013 and an initial bankroll of $5k. It grew to be much bigger than that. I know the pain of the $15 tables; I know the rollercoaster of the $300 tables; I've shuffle-tracked with mild success; I've max-betted with much failure. I got comped many nights, some concert tickets, much food, a windows surface tablet... got trespassed at two casinos and backed off at a handful of others... saw the major wins as well as the very painful, very awful, month-long 5-digit losses of death .... and now I am on for the next challenge -- especially since I am now done with PhD school and have a real job with real work hours.

So on to Roulette. Yes, for a truly random machine, the roulette is the worst near-even-money game in the house with an EV of -$0.0526 for every dollar bet on the American wheel (vs BJ, which is just fractions of one penny for perfect BS). But when you're live, it is also the ONLY game in the house where you can bet on the outcome after it's been de facto determined by physics. Enter visual ballistics.

There are just shy of 100 posts on VB on this forum, vs. many tens of thousands for BJ. I'm hoping to get some discussion going.

Step one is, just like memorizing BS for BJ, memorizing the wheel and the layout; getting comfortable placing bets on neighboring numbers. Somewhat arbitrarily**, I decided to chop up the roulette wheel into 10 semi-overlapping sectors, each 9 consecutive pockets in length. For the most part, that means one of the numbers between 1 and 10, and their four neighbors on either side. (interestingly, it makes much more sense to do this on the American wheel than on the European one... also interestingly, I find the American wheel much easier to memorize... more later). I built myself an Excel macro that simulates placing bets on a table, with a timer, a checking function, some additional options -- pretty helpful to understand the way the wheel is designed and how to move your hand across the layout to minimize the time it takes to place the bets you want. Some robo-roulette wheels have touchscreens that come pretty close to what my interface looks like.

So I have my 10 sector bets, I know which of the 10 sections each of the 38 numbers belongs to, and for the most part, I know each number's neighbors as well. And now I need to find a wheel that has very little deceleration and a dominant drop-point. Also helpful is a croupier who doesn't alternate the rotor direction every spin as they typically do in Europe... that's the worst. Lastly, most helpful is a croupier who's dozing off and calls NMB real late or is launching the ball on autopilot with similar force each time... ("dealer signature"). But these things I can't control -- they are field variables similar to how penetration and rule variations are for BJ. On to the controllable variables!

I searched the interwebs for tips on techniques for clocking the ball, the rotor, and the ball relative to the rotor. More than half of everything I came across is a scam. Also, I don't want to use vibrator devices hidden in my shoe - just plain old skillful play by identifying the likeliest sector the ball will drop on. I understand that the skillful part is in identifying the ball's sweet speed when it just leaves the rim -- maybe a few rotations before that -- because that always happens at the same speed for a given ball and wheel. What gets me is how to deal with a rotor that might change speeds from spin to spin. You can't just "guess" that the ball's total travel relative to the wheel will be +120 degrees or -90 degrees versus the previous spin or whatever -- there needs to be a quantitative way to judge that. What do the veterans on this forum think?

@Keyser - I read a good number of your posts --- would welcome any pointers. I'm aligned with many of the "outrageous" claims you make that others seem to bash a bunch! :)

** The math comes out nicely this way. Winning once for every three losses gives you exactly a net-zero game. Objective: attain a greater than a 1:3 ratio of wins:losses.

Too many words for me. Are you saying you are going to ball track?
Maybe you should have dinner with EB and learn his skill.

Go here:

http://www.rouletteforum.cc/index.php?PHPSESSID=bfe3a2031adad92fc7180e6421e93c60&board=91.0

I cant bring myself to believe in ball tracking, even if you could ball track enough to swing the odds 1% into your favorite, the house edge is 5%, so you would have a 4% house edge

Ask Kentrys mom how to track balls.

If you have a job now, time and motivation are going to be a problem.

Tldr cliffs?

Quote: arcticfun

Dear WoV,
. . .
So on to Roulette. Yes, for a truly random machine, the roulette is the worst near-even-money game in the house with an EV of -$0.0526 for every dollar bet on the American wheel (vs BJ, which is just fractions of one penny for perfect BS). But when you're live, it is also the ONLY game in the house where you can bet on the outcome after it's been de facto determined by physics. Enter visual ballistics.

I searched the interwebs for tips on techniques for clocking the ball, the rotor, and the ball relative to the rotor. More than half of everything I came across is a scam.

** The math comes out nicely this way. Winning once for every three losses gives you exactly a net-zero game. Objective: attain a greater than a 1:3 ratio of wins:losses.

Seriously wishing you good luck.
I can believe that Visual Ballistics can help, but as already pointed out, the pre-existing edge is a big hurdle. Why run up against 5.26% when it's not much more effort to analyse 2.6%
Nice to hear of someone other than me that uses Excel creatively.
Personal recommendation: Try to video record a lot of spins with a tiny covert camera and analyse those spins to death. ( yeah, I know it's naughty, but I never said to do it while wagering)