#1 Roulette Forum & Message Board | www.RouletteForum.cc

Roulette-focused => General Discussion => Topic started by: Roulettebeater on Apr 06, 08:44 AM 2020

Title: How many tests are needed before concluding that a system is a winner
Post by: Roulettebeater on Apr 06, 08:44 AM 2020
Do we have testing experts here ?

Comments are welcome.

Thx
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: Bebediktus3 on Apr 06, 09:44 AM 2020
Really matter not in a number of tests :). Must be simply calculations that system has edge.

Guys, when you finally will understand simply thing - the system is winner when it gave an advantage to the player, otherwise, it is looser! How you that cant understand?

Negative bets create a negative result. You cant be a winner if you with every bet gave some part of the money...

Title: Re: How many tests are needed before concluding that a system is a winner
Post by: gizmotron2 on Apr 06, 10:30 AM 2020
Quote from: Bebediktus3 on Apr 06, 09:44 AM 2020
Really matter not in a number of tests :). Must be simply calculations that system has edge.

Guys, when you finally will understand simply thing - the system is winner when it gave an advantage to the player, otherwise, it is looser! How you that cant understand?

Negative bets create a negative result. You cant be a winner if you with every bet gave some part of the money...

That's not actually true. It's just more axiomatic altruism. It's just trying to confirm a belief by saying enough times. Research is done like this. It takes a control group where half the participants in the control group are given a placebo. Well your conjecture that it must have an explainable edge is that placebo. Why must it be explained?  If enough people in the control group show as being successful at confirming that a gambling method works then the belief in the placebo side becomes validated as being without merit. At the same time evidence then exists where the method in experimentation has been shown to be both confirmed and validated. The need for demonstrating the so called requirement for an edge, the placebo, is also validated as confirming no merit.  That experiment is under way in Reading Randomness. There is just too little data at this time. Meanwhile people will still keep flogging the placebo.
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: maestro on Apr 06, 10:30 AM 2020
QuoteNegative bets create a negative result. You cant be a winner if you with every bet gave some part of the money...



well if you bet 6 numbers and manage to hit in 5 spins you will win even there is negative exp.
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: Bebediktus3 on Apr 06, 10:43 AM 2020
Quote from: maestro on Apr 06, 10:30 AM 2020well if you bet 6 numbers and manage to hit in 5 spins you will win even there is negative exp.
But in your sentence, the main word is IF!  And that is a bad word because we not need any IF ...we want to be sure in what will be :) And because of that you must learn to calculate chances...
If my hit rate when cover 6 numbers is 1/5 I have 20% advantage and win without problems. :)
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: Bigbroben on Apr 06, 10:44 AM 2020
Quote from: maestro on Apr 06, 10:30 AM 2020


... and manage to hit

Good management means positive edge.  Which can be calculated.  If it cannot be calculated, than there is no reason the expectation to be positive.
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: Roulettebeater on Apr 06, 10:47 AM 2020
 :thumbsup:
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: Roulettebeater on Apr 06, 10:48 AM 2020
Quote from: Bebediktus3 on Apr 06, 09:44 AM 2020
Really matter not in a number of tests :). Must be simply calculations that system has edge.

Guys, when you finally will understand simply thing - the system is winner when it gave an advantage to the player, otherwise, it is looser! How you that cant understand?

Negative bets create a negative result. You cant be a winner if you with every bet gave some part of the money...

You are missing the point - every strategy requires verification
advantage player sometimes face situation where they miss multiple spins, its normal but in the long term they win.

my question is, how many spins does a player (either system or advantage player) need to test before he concludes that his approach is a winner?
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: Roulettebeater on Apr 06, 10:51 AM 2020
Quote from: maestro on Apr 06, 10:30 AM 2020


well if you bet 6 numbers and manage to hit in 5 spins you will win even there is negative exp.

so in other words, hitrate should be at least 83.33%, right ?
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: gizmotron2 on Apr 06, 11:00 AM 2020
Quote from: Roulettebeater on Apr 06, 10:48 AM 2020my question is, how many spins does a player (either system or advantage player) need to test before he concludes that his approach is a winner?

It's not by the spins for me. I can't control or quantify situational awareness or optimum coincidence. So It comes down to sessions won and the eventual win to loss ratio. To get a realistic number this takes more than a thousand sessions.
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: Bebediktus3 on Apr 06, 11:01 AM 2020
Quote from: Roulettebeater on Apr 06, 10:48 AM 2020You are missing the point - every strategy requires verification
No not every, and at all mostly are simply calculated, not verified...
I will give an example: say I Grandmaster in chess play against you 10 games and winner of each game gets 1000$. Now if you are not Grandmaster your chances to be winner after 10 games near to 0 but my are near to 100% and not need any verification...
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: Bebediktus3 on Apr 06, 11:03 AM 2020
Quote from: Bigbroben on Apr 06, 10:44 AM 2020Good management means positive edge.
That is wrong - managment and edge are different things....
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: Bebediktus3 on Apr 06, 11:07 AM 2020
Quote from: gizmotron2 on Apr 06, 10:30 AM 2020Research is done like this. It takes a control group where half the participants in the control group are given a placebo.
I said not need any research in most situations - that is simply clear before all researches as in the example with Chess Grandmaster...
In roulette, if you do not have some weapons to know which number is more likely to fall your bet will have -2.7% negative expectation no matter what you will do !
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: Bebediktus3 on Apr 06, 11:12 AM 2020
The main moment what is strange why you all so avoid all what gives advantage and spent so much for arguings about abilities negative bets transfer to positive ....?
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: gizmotron2 on Apr 06, 11:18 AM 2020
Quote from: Bebediktus3 on Apr 06, 11:07 AM 2020
I said not need any research in most situations - that is simply clear before all researches as in the example with Chess Grandmaster...
In roulette, if you do not have some weapons to know which number is more likely to fall your bet will have -2.7% negative expectation no matter what you will do !

That's more of the same thing. It does not make it true.  Everyone would agree that if I bet 1000 bets in a row on red at $5 per bet that I will lose at or around the rate or percentage you say it must. But If I know that in those 1000 bets I decide to sit out or bet on black when black spins are grouped and in a condition of domination then those expectations will then be numerically different. Now tell me how you know that I can't know when blacks are in domination.
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: Bebediktus3 on Apr 06, 11:29 AM 2020
Quote from: gizmotron2 on Apr 06, 11:18 AM 2020Now tell me how you know that I can't know when blacks are in domination.
Do you understand that if I am Grandmaster in chess I simply know that will win accordingly 99.9% random peoples? From where I know? Simply because I know how to play and others 99.9% or not know or know much worse than me.
So in roulette, you simply must create the same situation how with that Grandmaster...that is all.

And opposite to that Grandmaster - you do not know when black will fall and because of that your expectation to win is negative... all is super simple.
All this talk is about expectations to win -  you simply not understand. what you must be able to do, that expectation to win will be positive.
In chess that is simply -,, you want to win must learn how to play that game - better than others.

Why do you think that in roulette is other?

Why do you think, that in this game you can be winner by knowing some simple rule ?
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: gizmotron2 on Apr 06, 11:42 AM 2020
Quote from: Bebediktus3 on Apr 06, 11:29 AM 2020Why do you think, that in this game you can be winner by knowing some simple rule ?
I have spent a few years teaching people how to see domination and to exploit it. After that someone suggested that I could teach everyone for free and most of them would not see it or do the work it takes to see it. That has pretty much been validated. So it sits there in plain sight. Eventually it will be validated. Until then the world will still believed to be flat. And that ends me trying to suggest what is required to validate a real experiment.
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: Joe on Apr 06, 12:41 PM 2020
Quote from: Roulettebeater on Apr 06, 10:48 AM 2020my question is, how many spins does a player (either system or advantage player) need to test before he concludes that his approach is a winner?

It depends on the amount of numbers bet and how sure you want to be. More bets = more certainty that the system has an edge if it continues to win. You can calculate the empirical edge using the formula: Edge = total profit / total amount bet.
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: maestro on Apr 06, 01:34 PM 2020
QuoteIf my hit rate when cover 6 numbers is 1/5 I have 20% advantage and win without problems.


big brain..... :xd: :xd:
Title: Re: How many tests are needed before concluding that a system is a winner
Post by: Bebediktus3 on Apr 06, 03:45 PM 2020
Quote from: maestro on Apr 06, 01:34 PM 2020big brain.....
Say that to autor  :twisted:
Quote from: maestro on Apr 06, 10:30 AM 2020well if you bet 6 numbers and manage to hit in 5 spins you will win even there is negative exp.