Value extraction from live roulette

[alani123]

As you know live roulette doesn't follow an algorithm to generate randomness. The dealer spins the ball with his imperfect human hands and the entropy of the butterfly effect leads to hard to predict outcomes every round. There is physics involved but I'd venture to guess that even a very powerful computer couldn't predict the outcomes perfectly in the given time between each spin.

Now, the question of if positive EV can be achieved on live roulette is on if somehow a pattern can be detected. Some have argued that the physics is enough to predict the outcome in terms of which section the ball would land on, at least making better predictions than random ones. But then again there's the house edge, so even if you make predictions better than predicting randomly, the predictions need to be notably and reliably better than just 1%.

The other limitation is that the window to use the physics visible and make a prediction is approximately 3 seconds while the dealer spins the wheel, then the ball and then the game gives this time window before the bets are closed.

A live casino setting allows for a physics calculation to be done via video feed without having to carry equipment in a physical casino and get the police called on you. But the question remains if it's possible to beat entropy to an extent that the house edge also gets beaten and lets the player win long term. Because even with bonuses, rake back/loss back and french rules the house edge might still be too high.

Has this been attempted? Do you think it's possible?

[eisen33]

As you know live roulette doesn't follow an algorithm to generate randomness. The dealer spins the ball with his imperfect human hands and the entropy of the butterfly effect leads to hard to predict outcomes every round. There is physics involved but I'd venture to guess that even a very powerful computer couldn't predict the outcomes perfectly in the given time between each spin.

Now, the question of if positive EV can be achieved on live roulette is on if somehow a pattern can be detected. Some have argued that the physics is enough to predict the outcome in terms of which section the ball would land on, at least making better predictions than random ones. But then again there's the house edge, so even if you make predictions better than predicting randomly, the predictions need to be notably and reliably better than just 1%.

The other limitation is that the window to use the physics visible and make a prediction is approximately 3 seconds while the dealer spins the wheel, then the ball and then the game gives this time window before the bets are closed.

A live casino setting allows for a physics calculation to be done via video feed without having to carry equipment in a physical casino and get the police called on you. But the question remains if it's possible to beat entropy to an extent that the house edge also gets beaten and lets the player win long term. Because even with bonuses, rake back/loss back and french rules the house edge might still be too high.

Has this been attempted? Do you think it's possible?

I think that over time, a dealer can notice certain patterns, how hard they spin the wheel, which ball they use, and how the roulette wheel behaves, I mean how accurately it is positioned on the table, whether there are any slight inclines or other factors that could make the ball land in certain sectors more often. Basically all of these things can potentially influence the outcome. But honestly, I think that with a reputable casino, these kinds of issues should be ruled out as well.

[danherbias07]

Win long-term in live roulette? If that ever happens, I think that gambler will probably not boast about it to preserve his profit-making machine.

When it comes to live roulette, there are so many superstitions about it, and some would even say there's a pattern. There's none. But I do believe that the weight of the ball could make a difference, and physics could be applied. I guess that is where the pattern will make sense because the weight and size of the ball may differ, and then some balls are sometimes chipped, and it could also change the results of the game.
But I honestly doubt one could do it long term because they change it and so do the dealer.

[Dr.Bitcoin_Strange]

This topic reminded me of an article I read a long time ago when I was making some inquiry about roulette game as it is also one of my favorite casino game. The article made reference to 2 students that use science to beat the odd of roulette, you can read it Isaac Newton Vs. Las Vegas: How Physicists Used Science To Beat The Odds At Roulette.

Those two friends won a lot of money from roulette not only because they applied physics but they saw some defects and weakness that the game equipment had and they used it to their advantage. This was a long time, right now, many upgrades has been done and would be difficult to even see such loophole to exploit using physic or maths.

The game is fast, one can not put together some calculations to get any exact prediction before the spin stops and some casino would not also allow players to bet after the ball has already started rolling.  It's actually delusional to think that one can beat the house now using any knowledge of science, that's why we should risk what we can afford to lose and have fun.

[xenomorfo]

Win long-term in live roulette? If that ever happens, I think that gambler will probably not boast about it to preserve his profit-making machine.

When it comes to live roulette, there are so many superstitions about it, and some would even say there's a pattern. There's none. But I do believe that the weight of the ball could make a difference, and physics could be applied. I guess that is where the pattern will make sense because the weight and size of the ball may differ, and then some balls are sometimes chipped, and it could also change the results of the game.
But I honestly doubt one could do it long term because they change it and so do the dealer.

I don't think we can win in the long term, i don't remember who explained the mathematical reason to me.
But I'm bad at math.
I can only tell you, it's better that you play to just pass the time, it's no coincidence that it's called gambling.
Money is only made by working, honestly.

[Free Market Capitalist]

This topic reminded me of an article I read a long time ago when I was making some inquiry about roulette game as it is also one of my favorite casino game. The article made reference to 2 students that use science to beat the odd of roulette, you can read it Isaac Newton Vs. Las Vegas: How Physicists Used Science To Beat The Odds At Roulette.

Those two friends won a lot of money from roulette not only because they applied physics but they saw some defects and weakness that the game equipment had and they used it to their advantage.

They weren't the only ones, some matematicians who were also poker players exploited the imperfections in several casinos.

The TRUE STORY of the Family Who BEAT THE CASINO | Los Pelayos

You're late, alani123. After those incidents, the casinos made sure that the roulette wheels don't have any flaws, and they check them on a regular basis.

[Smartprofit]

Ideally, both cryptocurrency trading and roulette are games of chaos. The outcomes in this game are completely random, so the player has no mathematical advantage. As a result, they inevitably lose in the long run. This is true for both traders and players. 🙋

In my opinion, a player can gain a competitive advantage only by discovering a pattern in the chaos of the game. Such a pattern could theoretically arise either in the roulette system itself or in the human (dealer) system. This means that the dealer could simply be tired or have tendon problems in their hand, causing them to throw the ball in a very specific way. Such a tired or ill dealer is a chance for the player to gain a competitive advantage over the casino.

In fact, casinos aren't completely stupid either. As you may have noticed, dealers are changed periodically. And casinos also change the balls periodically. Moreover, the balls are made of different weights and diameters. This is done precisely to eliminate any pattern in the throw.  However, technology is advancing, and I don't rule out the possibility that in the future, players will be able to conduct more effective attacks on casinos. For example, one could imagine a powerful computer program that could create a complete simulation of a roulette wheel spin and the dealer's ball roll. One could also imagine a large number of players pooling their resources to systematically search for patterns in real roulette. However, all of this, of course, is more science fiction than reality. And even if this were to happen, casino employees would very quickly discover that someone was trying to steal their money this way. 💰

[cryptomaniac_xxx]

They weren't the only ones, some matematicians who were also poker players exploited the imperfections in several casinos.

The TRUE STORY of the Family Who BEAT THE CASINO | Los Pelayos

You're late, alani123. After those incidents, the casinos made sure that the roulette wheels don't have any flaws, and they check them on a regular basis.

Yes, that's what they call biased roulette tables, some imperfections that it's hard to spot for a ordinary player. But if you are those that are really observant, in the past they could have found some of that and exploited the system until casinos where able to catch up with them.

And so they make a lot of move as well, like rotating the roulette machine itself so that those who are recording might not follow the pattern. Or change the ball or maybe let their dealer knows about it and so they can change how they are going to throw the ball.

[viljy]

There is already a fairly old study about roulette: Predicting the outcome of roulette
Here is a quote from the abstract of the article:

The first system relies only on a mechanical count of rotation of the ball and the wheel to measure the relevant parameters. By applying this techniques to a standard casino-grade European roulette wheel we demonstrate an expected return of at least 18%, well above the -2.7% expected of a random bet. With a more sophisticated, albeit more intrusive, system (mounting a digital camera above the wheel) we demonstrate a range of systematic and statistically significant biases which can be exploited to provide an improved guess of the outcome. Finally, our analysis demonstrates that even a very slight slant in the roulette table leads to a very pronounced bias which could be further exploited to substantially enhance returns.

[Cointxz]

A live casino setting allows for a physics calculation to be done via video feed without having to carry equipment in a physical casino and get the police called on you. But the question remains if it's possible to beat entropy to an extent that the house edge also gets beaten and lets the player win long term. Because even with bonuses, rake back/loss back and french rules the house edge might still be too high.

Has this been attempted? Do you think it's possible?

I created before a fun experiment thread dedicated for roulette using chi square method just to determine if there’s a biased on roulette. Finding biased result is the only why to have some boost on the EV for this game.

Sadly, it’s nearly impossible to find one since modern roulette usually have frequent maintenance and they intentionally change ball and dealer to avoid having bias result.

In the past, I believe some gambler attempted to crack roulette just like other casino games.

I saw this study https://arxiv.org/abs/1204.6412 and many more on the earlier years.

[Findingnemo]

It is just too much complicated and we really can't find a pattern there even if there seems to be, because human hand is the random factor there, it jus lands on different angle everytime with different speed and should be enough to bring in the randomness here.

[alani123]

A live casino setting allows for a physics calculation to be done via video feed without having to carry equipment in a physical casino and get the police called on you. But the question remains if it's possible to beat entropy to an extent that the house edge also gets beaten and lets the player win long term. Because even with bonuses, rake back/loss back and french rules the house edge might still be too high.

Has this been attempted? Do you think it's possible?

I created before a fun experiment thread dedicated for roulette using chi square method just to determine if there’s a biased on roulette. Finding biased result is the only why to have some boost on the EV for this game.

Sadly, it’s nearly impossible to find one since modern roulette usually have frequent maintenance and they intentionally change ball and dealer to avoid having bias result.

In the past, I believe some gambler attempted to crack roulette just like other casino games.

I saw this study https://arxiv.org/abs/1204.6412 and many more on the earlier years.

Is it only about bias then?

Wouldn't the result be predictable based on the time between the spin and betting time still being open for a few seconds?

If for instance a spin starts from the number 0, with the speed of the wheel and the ball calculated one could theoretically make a prediction about the section it's most likely to fall on. If the predictive ability is just 2% better than pure luck then the house edge can be beaten.

This is of course very significant as it would make roulette be considered a solved game. If achieved even without utilising any occasional imperfections then it could be abused to make big and reliable profits.

Most probably even if this technology exists it's not very widespread. Otherwise I don't see them making any more live roulette games. But somehow I keep seeing new live roulette tables being opened left and right. So surely casinos must have a way to combat this.

[Cointxz]

A live casino setting allows for a physics calculation to be done via video feed without having to carry equipment in a physical casino and get the police called on you. But the question remains if it's possible to beat entropy to an extent that the house edge also gets beaten and lets the player win long term. Because even with bonuses, rake back/loss back and french rules the house edge might still be too high.

Has this been attempted? Do you think it's possible?

I created before a fun experiment thread dedicated for roulette using chi square method just to determine if there’s a biased on roulette. Finding biased result is the only why to have some boost on the EV for this game.

Sadly, it’s nearly impossible to find one since modern roulette usually have frequent maintenance and they intentionally change ball and dealer to avoid having bias result.

In the past, I believe some gambler attempted to crack roulette just like other casino games.

I saw this study https://arxiv.org/abs/1204.6412 and many more on the earlier years.

Is it only about bias then?

Wouldn't the result be predictable based on the time between the spin and betting time still being open for a few seconds?

If for instance a spin starts from the number 0, with the speed of the wheel and the ball calculated one could theoretically make a prediction about the section it's most likely to fall on. If the predictive ability is just 2% better than pure luck then the house edge can be beaten.

This is of course very significant as it would make roulette be considered a solved game. If achieved even without utilising any occasional imperfections then it could be abused to make big and reliable profits.

Most probably even if this technology exists it's not very widespread. Otherwise I don't see them making any more live roulette games. But somehow I keep seeing new live roulette tables being opened left and right. So surely casinos must have a way to combat this.

Based on my research when I created the thread, only biased is the potential criteria that we can use to determine potential advantage on roulette result. There’s roulette speed, ball weight, and other factor that varies per round but assuming there will be a constant measurement for all those factor then in theory it’s possible.

Considering that casino operators are vigilant then I believe it’s hard to measure those factor accurately in actual games even biased.

Arbitrage betting is the only method that I know that gives +EV excluding poker.

[Free Market Capitalist]

So surely casinos must have a way to combat this.

Casinos have it much easier today than they did two or three decades ago, when some people managed to exploit imperfections. In addition to ensuring that the roulette wheels have no imperfections, the balls are replaced periodically, as are the croupiers, and today all the results are stored in databases that they can analyze—even using AI—to see if anything seems off.

[kennycryptoitalia]

This topic reminded me of an article I read a long time ago when I was making some inquiry about roulette game as it is also one of my favorite casino game. The article made reference to 2 students that use science to beat the odd of roulette, you can read it Isaac Newton Vs. Las Vegas: How Physicists Used Science To Beat The Odds At Roulette.

Those two friends won a lot of money from roulette not only because they applied physics but they saw some defects and weakness that the game equipment had and they used it to their advantage. This was a long time, right now, many upgrades has been done and would be difficult to even see such loophole to exploit using physic or maths.

The game is fast, one can not put together some calculations to get any exact prediction before the spin stops and some casino would not also allow players to bet after the ball has already started rolling.  It's actually delusional to think that one can beat the house now using any knowledge of science, that's why we should risk what we can afford to lose and have fun.

well, after reading your message a thought arose in my mind, now we can do incredible things, definitely very advanced compared to just 2 or 3 years ago, now with AI we can reach truly very high levels, this is something that will go further and further, and those who want to risk more will do so in an even more extreme way.

[Junii]

Has this been attempted? Do you think it's possible?

Bro roulette is physically predictable but to some extent because the ball and wheel follow physics but it is not as simple like first of all the main problem is getting accurate  information in those few seconds and honestly it needs a lot of activeness and also a next level brain decision making and if someone do it then the prediction still needs to be accurate as well in the long run because there is also house edge available which beats the person in numbers . If i have to rate it in 100 there were 10% chances of possibility or may be less then that.

[free-bit.co.in]

The other limitation is that the window to use the physics visible and make a prediction is approximately 3 seconds while the dealer spins the wheel, then the ball and then the game gives this time window before the bets are closed.

You are talking about 3 second time window but live stream is not real time!

Live video we see on the screen from Evolution, Playtech or Pragmatic has video latency of at least 1.5 to 3 second. Casino server fires the No More Bets signal exactly 1-2 seconds after ball is released. By the time your AI Vision software detect the ball's initial speed and trajectory in video frame, the casino's software has stopped accepting your bets. That is, there is no physical time window to place the bet even though calculation is processed.

[Donk1]

Well the thing is even if there is a way to figure out how fast the ball spins with a computer, it's still not a certainty though. Because anything can happen, like what if the live video suffers from just a tiny minor network glitch/ delay or something just kinda interrupts it Basically, missing even a second can disrupt the timing that is required to get a precise prediction of where the ball will fall, so by the time the computer is done with the analysis, you might just still end up losing. In most cases stuff like this isn't impossible, but yeah, you would have to be willing to take the risk,because we often know that anyone can have the edge in the game, but yeah, the casino can be the one with more advantage though. So that's all.......

[tsaroz]

As you know live roulette doesn't follow an algorithm to generate randomness. The dealer spins the ball with his imperfect human hands and the entropy of the butterfly effect leads to hard to predict outcomes every round. There is physics involved but I'd venture to guess that even a very powerful computer couldn't predict the outcomes perfectly in the given time between each spin.

Now, the question of if positive EV can be achieved on live roulette is on if somehow a pattern can be detected. Some have argued that the physics is enough to predict the outcome in terms of which section the ball would land on, at least making better predictions than random ones. But then again there's the house edge, so even if you make predictions better than predicting randomly, the predictions need to be notably and reliably better than just 1%.

The other limitation is that the window to use the physics visible and make a prediction is approximately 3 seconds while the dealer spins the wheel, then the ball and then the game gives this time window before the bets are closed.

A live casino setting allows for a physics calculation to be done via video feed without having to carry equipment in a physical casino and get the police called on you. But the question remains if it's possible to beat entropy to an extent that the house edge also gets beaten and lets the player win long term. Because even with bonuses, rake back/loss back and french rules the house edge might still be too high.

Has this been attempted? Do you think it's possible?

There are always people looking for patterns to beat the house edge and its a cat and mouse game between them and the casino yet not everyone might have explored every of the casinos.
I've heard that physical roulette wheel however polished and calibrated has a very small degree of imperfection that favors some results than others. As they are not replaced quite often and if we have enough data, we might make better guess and compare that with the style and pattern of the operator we might beat the house edge.
But that's theory and we need a dedicated person to measure if its true.

[EluguHcman]

[~SNIP

I don't think we can win in the long term, i don't remember who explained the mathematical reason to me.
But I'm bad at math.
I can only tell you, it's better that you play to just pass the time, it's no coincidence that it's called gambling.
Money is only made by working, honestly.

Funny but the truth.
Some players there will spend a whole time trying to use every available resources just to make wins of their bets and with times being without numbered, no strategies is reliable on gambling especially that of roulette that defines it out randomly.

So those applying the mathematical terms with so much efforts like the bets they are playing is a serious activity which they relies on making income.
In the random outcome games, the dealers will never loose an inch to loose in the long time while players wins. Instead, they would allow you win for a while just to make the game interesting to you which is why first timers are usually lucky to win in their first stage.

If the dealer has also understand your strategy and you are winning streak, they still have alternative to manipulate the RNG just to scatter your algorithm.