Can Physics Be Used To Predict A Roulette Outcome?

[Chingy711]

Hello Everyone,
I'm new to your forum and was hoping to get some answers concerning
physics being applied to a roulette wheel. I'm a member of a roulette forum
where there is presently a member that states he can predict 99% of the time
where the ball will land within a +/- 18 number sector on the wheel. He states
he clocks the time it takes for the ball to make 1 revolution and the rotor speed
of the wheel and can determine where the ball will strike the wheel. He states
based on these visual measurements he can apply the scatter affect of the ball
to accurately 99% of the time predict which half of the wheel the ball will land.

Is this possible using physics and to the degree he states? I'm aware that if the
same amount of force was used to spin the ball everytime it would be possible
to some degree to know when the ball will start to fall, but is it possible to figure
how far the ball will scatter based on the amount of force used to spin the ball?

I would appreciate any thoughts on this matter, I just can't believe this is possible
using visual observations.

 

[Simon Bridge]

I'm a member of a roulette forum where there is presently a member that states he can predict 99% of the time where the ball will land within a +/- 18 number sector on the wheel. He states he clocks the time it takes for the ball to make 1 revolution and the rotor speed of the wheel and can determine where the ball will strike the wheel. He states based on these visual measurements he can apply the scatter affect of the ball to accurately 99% of the time predict which half of the wheel the ball will land.

Well you want to see the results of some in-field testing before paying any more attention. You certaily want to know how that 99% was calculated.

Is this possible using physics and to the degree he states?

Yes and no.
In principle it's a totally deterministic system with a very large number of variables.

In practice, as you rightly suspect, you may get a long run of successful guesses but you won't be able to keep it up. There are too many variables.

Notice that which side of the wheel does not really help much ... the red and black numbers alternate around the rim - which is the usual halving (leaves having to spread bets around, what, 20 numbers? Unless the claim is he can predict red vs black?

Can you even place a bet after the ball goes into play?

The description looks like something a croupier could use, but not a punter.

 

[Chingy711]

Well you want to see the results of some in-field testing before paying any more attention. You certaily want to know how that 99% was calculated.Yes and no.
In principle it's a totally deterministic system with a very large number of variables.

In practice, as you rightly suspect, you may get a long run of successful guesses but you won't be able to keep it up. There are too many variables.

Notice that which side of the wheel does not really help much ... the red and black numbers alternate around the rim - which is the usual halving (leaves having to spread bets around, what, 20 numbers? Unless the claim is he can predict red vs black?

Can you even place a bet after the ball goes into play?

The description looks like something a croupier could use, but not a punter.

Thank you Simon. I really appreciate the reply. Yes, a bet can be made once the ball
is in motion. Generally it's 5 to 8 seconds before the dealer calls, "No more bets."
It is in those few seconds that this guy clocks the revolution speed of the ball from the
area of the wheel of which the ball is released from.. Then he comes to a calulation where
the ball should stop and bets 9 numbers clockwise and 9 numbers counter clockwise. Based
on my knowledge of probability a bettor could make a fortunte if he could just be able
to exclude even 4 numbers from coming out per spin. From what this guys states he
can exclude an 18 number sector. What most people don't realize is the ball spins
in one direction while the wheel itself rotates in the opposite direction. Speed is naturally
a factor, especially when the ball hits the board or one of the reflectors, the scatter is
the only thing that should make this almost impossible to any degree of certainty.
Well thanks again. Louie

 

[Simon Bridge]

That's right - the scatter is the key.
One could, in principle, work out the results - but it would be very sensitive to the initial conditions.
Small wear on the components, for eg., can throw the whole thing off.

Computer simulated roulette may be predictable like this though - depending on the physics engine (and if it uses one).

hmmm... is it reasonable to be able to tell, quickly, which part of the wheel the ball is launched from?
How accurately? That may be an easy thing to check.

The other check, if you have access to a wheel, an experiment could be performed to find how deterministic tit is to release the ball with the same initial speed and angle, in the same general part of the spin. Looks like one for Mythbusters.

Don't roulette wheels get tested for randomness?

 

[A.T.]

I'm new to your forum and was hoping to get some answers concerning
physics being applied to a roulette wheel. I'm a member of a roulette forum
where there is presently a member that states he can predict 99% of the time
where the ball will land within a +/- 18 number sector on the wheel. He states
he clocks the time it takes for the ball to make 1 revolution and the rotor speed
of the wheel and can determine where the ball will strike the wheel. He states
based on these visual measurements he can apply the scatter affect of the ball
to accurately 99% of the time predict which half of the wheel the ball will land.

99% sounds a bit optimistic. But you need only a small bias, to make a net gain. Even 60% is enough, if you don't get greedy, and just systematically bet the same amount.

Roulette beater spills physics behind victory
http://www.newscientist.com/article...d-victory.html?DCMP=OTC-rss&nsref=online-news

Predicting the outcome of roulette
http://arxiv.org/abs/1204.6412

I also read a story about a european guy, who was first using computers, but then through training became so good, that he could clock the ball and do the math in his head. He is banned from many casinos for life.

 

[Danger]

I don't actually know how fast a roulette wheel spins, but from what I've seen on TV and in Vegas casinos, it's faster than the human eye can physically track. After all, TV and movies work, as do many "magic" tricks, because of "persistence of vision". Anything over a couple of hundred rpm's becomes a blur.

 

[schip666!]

For some social background, and a little technical detail, on the article ref'd above (Roulette beater spills physics behind victory) see the book The Eudaemonic Pie:

https://www.amazon.com/dp/0595142362/?tag=pfamazon01-20

 

[Danger]

see the book The Eudaemonic Pie

I would be mildly interested in reading it, but not nearly enough so to pay for it.

Anyhow, I saw a "documentary" about that caper on TV already. It's not the same thing that Chingy was asking about.