The Simulation Fallacy

[Scarface]

Why don't you test your system for only one spin? 

All test will fail if you run enough simulations.  Even your chi square can fail.  Strategy is more important.  What is your strategy if your tests show a bias wheel that is not bias?  Do you play until your bankroll is wiped out like the simulation shows?  Or do you stop, and take your loss?

[Tinsoldiers]

Strategy is more important. 

Let's talk real Scarface. You think you will be able to give one strategy. I will try to do it in a disciplined manner and see how that strategy holds. Like everyone else, I am looking for something that wins and unfortunately dont have one. Please dont do riddles and puzzles.

[Joe]

Even your chi square can fail.  Strategy is more important.  What is your strategy if your tests show a bias wheel that is not bias?

Don't a lot of casinos these days run Chi square tests on their wheels? If the casino does it then any advantage player looking for biased wheels can't just be using what the casino uses, unless they're only looking in old backstreet casinos where they're unlikely to monitor the wheels closely.  But these kind of casinos are more likely to be ripping you off anyway.

[Scarface]

Tinsoldiers, I think you are asking about a system to test?  Not so much a strategy.  When I talk about strategy, I'm referring to my own personal rules which is almost impossible to simulate because they may change depending on how the game is going.

My strategy is to bring no more than 20-25% of my bankroll to casino.  This way I can never lose my entire bankroll in a single session if variance takes an ugly turn.  My strategy is sometimes start with a parachute system, playing many numbers with few chips to grind out small gains.  Normally, I'll always switch to a repeater system playing few numbers.

My goal may be 500 units, and leave.  But at times I hit this early and raise my goal to 1000 units using a up as you win positive progression.  If it goes back down to 500, I'll stop.  If it reaches 1000, I may even change my goal to 1500, with a stopping point at +1000. 

Some days wins don't come so quickly.  I'll play for up to 6 to 8 hours alot.  Usually when I get to that 6 hour mark, that's when I'm looking for a good point to exit the game.  Main reason for this is really because I'm tired of playing at that point.

My strategy will never involve steep negative progressions.  I will never go from betting 1 unit a number to 50 units.  If I do a negative progression, it will only be after a hit and only if there was a long gap between hits.  Normally, I will never go more than 1 unit to 5 units max.

Positive progressions are a different story.  If variance is on my side, I will keep adding to my wagers as long as the wins keep coming.

But looks like you are looking for a system to test.  I like playing recent repeaters, no more than 4 numbers.  Check out the post for Repeaters and Gaps.   

[falkor2k15]

Tinsoldiers, I think you are asking about a system to test?  Not so much a strategy.  When I talk about strategy, I'm referring to my own personal rules which is almost impossible to simulate because they may change depending on how the game is going.

There's nothing you can do with RNG Roulette that a computer cannot be programmed to do in your place.

Winkel said the same thing: he follows the trot and claimed a computer cannot have "gambler's intelligence". 

This is very telling - says more about Scarface and winkel than it does about simulations.

[Scarface]

There's nothing you can do with RNG Roulette that a computer cannot be programmed to do in your place.

Winkel said the same thing: he follows the trot and claimed a computer cannot have "gambler's intelligence". 

This is very telling - says more about Scarface and winkel than it does about simulations.

Falknor, do you play in a b&m much?  I know you said you spend alot of time testing systems.  What do you hope to find?  A system that never fails?  This is pointless.  If you run enough simulations for long enough, all systems will fail.  Run enough simulations then you'll see the same number repeat 36 times.  All things all possible if you test enough.  So, what are you left with?

I agree with Winkel.  I like to call it gamblers intuition.  I'm not talking about remote viewing what numbers that will come up next.  I'm talking about what tells me to stop, and take that small loss.  Or, maybe to change my system.  Or when to call it quits.

Don't be so quick to dismiss intuition.  There is some science to back this up.  There have been experiments that suggest that our bodies react to certain stimuli before our conscious minds are aware.  I can point you to some of the science, if you're interested in this.

Here's what I know.  I can quit and take may profit/ loss at anytime I choose.  I can change the system I'm playing at anytime I choose.  But, you tie your own hands with your own rules.  You will play according to your system.  You will envitiablly lose your whole bankroll with to your system, by sticking to your rules. 

[Tinsoldiers]

You are now deviating from what you started this whole thing or at least it sounded like.  Intuition and Biased wheel conditions cannot be simulated and no one is saying simulate that to death. People only say simulation when someone claims there is a mechanical method and it will never lose.  I am sure anyone will agree this is completely different from intuition. There is no strategy here. Strategy can be defined, it can be repeatable executed and it can be simulated. That’s different from when I play what my mind says, that’s not strategy - as you rightly put it is intuition and no one is saying here simulate it. 

[The General]

I know you said you spend alot of time testing systems.  What do you hope to find?  A system that never fails?  This is pointless.  If you run enough simulations for long enough, all systems will fail.  Run enough simulations then you'll see the same number repeat 36 times.  All things all possible if you test enough.  So, what are you left with?

If you're system will fail in the long run then it can fail in the short run too.  99.9% of the time there's no reason to test a system because the results can easily be calculated mathematically.  The goal is to develop a method that overcomes the house edge and that will win in the long run.

I agree with Winkel.  I like to call it gamblers intuition.  I'm not talking about remote viewing what numbers that will come up next.  I'm talking about what tells me to stop, and take that small loss.  Or, maybe to change my system.  Or when to call it quits.

I call it guessing.   And if you test you're "gambler's intuition" then you too will find that you're really just guessing. 

Here's what I know.  I can quit and take may profit/ loss at anytime I choose.  I can change the system I'm playing at anytime I choose.  But, you tie your own hands with your own rules.  You will play according to your system.  You will envitiablly lose your whole bankroll with to your system, by sticking to your rules.

No you can't!  You can try and try, but you can't step outside of probability.  This means in the long run you can't win or lose at a rate that exceeds the house edge. A good example of this is Maestro on the MPR game.  The last time I'd checked he'd played so many spins that he'd locked into losing at exactly the house edge and had no hope of ever breaking even again.

Biased wheel conditions cannot be simulated.

If you have a statistically relevant number of spins from a biased wheel, which I do, then you most certainly can simulate and test.  As a matter of fact I have millions of spins from such wheels.  On some of the wheels I have over 300k spins.

[Scarface]

General, you can't sidestep probability right?  Someone can claim that 20 repeaters in a row can never happen.  But you can claim that if you simulate a statistically revelent number spins, at some point it will.  So, with randomness and enough test anything is possible.  So, that means even your chi square method of spotting bias wheels at some point will fail in the long run.  Random can spit out numbers way outside standard deviation on a wheel that isn't biased at all.  You can't get around it.  Your bankroll will eventually all be lost to the house edge 

[The General]

Scarface,

What do you consider to be a relevant chi square demonstrating a playable biased wheel?

[Scarface]

Scarface,

What do you consider to be a relevant chi square demonstrating a playable biased wheel?

You tell me, General.  You're the expert on it.  I'm just saying that random can through you a curveball.  If you can't admit this is possible, then you are saying there are limits to randomness.  Can't have it both ways. 

[The General]

I don't play on a biased wheel unless I'm absolutely sure that it's biased.  However, contrary to popular believe, the wheel is found before the numbers are written.

[Herby]

the wheel is found before the numbers are written.

Since the invention of the wheel not so hard to find one ...

[Kattila]

Since the invention of the wheel not so hard to find one ...

I found the general's wheel, lost by his soldiers......

[Joe]

I don't play on a biased wheel unless I'm absolutely sure that it's biased.  However, contrary to popular believe, the wheel is found before the numbers are written.

That's what I meant when I said in my earlier post that you must be doing more than just recording numbers and using chi-square, though I'm not sure what.