Roulette Measuring Success

Hi precogmiles, hope things are good with you too.
I would say in general now I am effective more often than not -previous barren runs could last weeks (so yes maybe moon related?) whereas now I can usually snap out of them by resetting focus. I have changed my technique a couple of weeks ago which seems to be helping. Thanks for the tips as always, will look into those in detail.
The issue for me it’s MPR is the 30 second time cutoff. I am much less effective (often no edge at all) when under pressure and this is a general problem when I’m not in control of conditions (timing of bets, lighting, noise and other environmental conditions). I think this a barrier to most if not all precog players winning in a casino. There are so many distractions and trying to block everything out is so difficult. I’m trying to work on this but it is not proving successful at all.
I like the idea of a spreadsheet. I find good sleep and diet and general overall well-being is much more conducive to successful play but I’ve not recorded this so will do that.
Cheers

 

For once you make the argument that math is irrelevant to success. I can wait for the win streak in that 100 spins and beat the coincidences that are never controlled by mathematical expectations. All it takes is waiting for the right combination of confirmations to strike and walking away with a small enough unit win goal.

 

@ Klausy,

How many numbers are you betting?

@ LF, Klausy wants a simple formula, not a lecture in statistics. The video you posted is for stats students, and I'm guessing most members here aren't. I doubt if any got through the first five minutes. And Dr Sir is not WRONG; it's not as black & white as that. Realistically, nobody is going to get results so overwhelming that no statistical analysis is necessary (ie, 37 wins in 37 spins betting a single number), and in a short sequence of spins it's not that hard to get a result which statistically significant (just in terms of the p-value) and yet it would be ridiculous to come to the conclusion that you have a winning system just based on those 37 spins, or even 100 spins. There are other considerations like the effect sizes and statistical power (from which you can find the minimum sample size), as well as the probability of winning a bet. That's why I have asked Klausy how many numbers he's betting.

So you know the answer has something to do with hypothesis testing, but the fact that you just posted a video doesn't inspire much confidence that you know how to choose and actually do an appropriate test. I call BS. You couldn't even solve the simple probability problem Benas asked on the other forum, and now you're making unsubstantiated accusations that he's a fake. Whatever.

 

Typically 4 numbers per spin. Thanks

 

I don't find your argument to have any teeth to it. The Reading Randomness thread is an experiment in the process of hypothesis testing. The reader, if they choose to engage in the learning skill task, is expected to keep track of their own win to loss accumulations in unit value over the entire process of learning the skills from the very start to a month past the time that they are skilled in the expertise and process of using it.

I could not be more happy that another mathBoyz has passed up the chance to participate. It could be arrogance? But the challenge, gantlet has been thrown down. If you set out to lose you probably will succeed. But others have done the process and succeeded. This can't be a case of too small a sample and a few succeeding. This is a case of almost everyone that has trained succeeding.

 

Up till now you still don't get the clear bolded pointer I have given you. You are supposed through your self taught refresher course after more than a decade post graduation that hypothesis testing, statistical significance, p-value will not help you find the statistical anomaly for positive edge systems betting.

There are hundreds of thousands of students who have to write answers for questions whose scenario that are more complex than roulette outcomes. Go figure. You don't need exact calculations.

What you need is deeper understanding from detailed exploration of the permutations of roulette spins to find the anomaly.

This level of statistical analysis requires an understanding that goes beyond the basic probability of 1/37. That's the area(you used the word "sophisticated math") you are lacking. Admit that you are weak that's why you seek further education. You're going to find it tough.

You got it completely wrong because of the erroneous assumption that 1/37 basic probability is the ONE and ONLY PROBABILITY applicable for roulette outcomes. And you repeat this mistake over and over again.

I write this post factually so that readers can benefit from stop being misled by your posts.

No more one and only 1/37 basic probability again. End of conversation.

Btw I'm writing from the casino floor collecting as usual. Lol

 

And stop replying to my post.

I play in b&m casino.
No more time for forum. Lol

 

Awaiting the return of A F Wendell and his hidden math .

 

There is a formula but it's easier to use an online calculator. Here's one.

There are 4 input fields you need to fill in (actually, only 4 because you can leave the 4th as it is):

1. Sample proportion. This is the proportion of wins. So say you were betting 4 numbers over 100 spins (betting every spin) and you got 15 wins (the expected number of wins is about 11. So your sample proportion is 0.15 (15/100). Put that in the first box.
2. Sample Size. This is the number of BETS you actually made (NOT the number of spins). Since I assumed you're betting every spin, the number of bets in this case is the same as the number of spins. So put 100 in this 2nd box.
3. Population Proportion Estimate. This is the expected proportion of wins when betting 4 numbers, which is the standard probability of a win betting 4 numbers (4/37). You'll have to put this in as a decimal, so put 0.1018 in the 3rd box (4 significant figures is enough).
4. You can leave the last box as it is (it should contain 0.05).

You can also leave the checked radio button on "Two-Tailed". Basically this will take into account the possibility that you get a statistically significant result which is LESS than expected. It's a more conservative option, which means you're less likely to overestimate your edge (if any).

Now click Submit, and you'll see the results on the next page. In the Results box you'll see that it says "Not Significant". That's all you really need to know. If you're interested you can read up and find out what the other values mean. Do make sure that it says "test ok" in the final line though.

Do the test again but this time put 0.17 in the sample proportion box (meaning you got 17 wins instead of 15). This time the result will be statistically significant.

If you're betting more or less than 4 numbers then calculate the average and use that in the sample proportion, for better accuracy. You should add the results from session to session so that you get a larger and large sample, that way you can see whether the results are stable, improving, or getting worse.

Good luck!

 

@ LF, if my posts irritate you so much I suggest you put me on ignore. And don't bother replying because I've put YOU on ignore. There, I've saved you some time, wasn't that nice of me? lol.

 

1. Define your bet.
2. How large do you think your edge is?
2. Flat bet. (No progression)
3. The bet selection should be able to exceed three standard deviation in two out of sample tests or five standard deviation in one test.

If you're betting on even chance bets, here's a grossly, simplified example of why 100 spins is a bit silly. It's not going to apply to individual number bets. Again, it's an example for even chances because this is the easiest way for people to understand what to expect.

If you DO NOT have the edge, your fluctuations will be the square root of the spin sample size of 100 (10) x 3 = 30 ...meaning because of good or bad luck (variance) you'll likely be somewhere between losing -35.26 to +24.74 units
Next subtract the house edge -5.26 units from the win side, and add it to the losing side. So you're expectation will likely be losing about 35.26 units or winning possibly as much as 24.74 units.

Now let's flip it around and see what happens if you have the edge of 5.26% instead of the house. You're likely fall somewhere between losing 24.74 units to possibly winning as much as 35.26 units.

Now why didn't you win every session, even though you had the edge??? It's because testing for only 100 spins was toooo small of a sample for your edge to dominate luck (variance). Meaning, the test sample size wasn't really statistically relevant!

Now let's see what happens if we increase the sample size to 1000 spins. (Square root of 1000) x (3) = somewhere between winning or losing 94.86 units because of the variance. Next let's add your edge 52.6 units. Now you're likely to fall somewhere between losing 42.26 units or possibly winning as much as 147.46 units. A 1000 spin test sample is still too small, but it's getting closer.

Let's try a 5000 spin sample. (Square root of 5000) x (3) = somewhere between winning or losing 212.13 units because of the variance. Next let's add your edge of 263 units. Now you're very likely to win somewhere between 50.87 units to as much as 475.13 units. As you can see, you've now tested a relevant number of spins...IF YOU HAVE THE EDGE INSTEAD OF THE CASINO! The size of the edge determines how long you need to test.


If you have a small edge, then you have to test for a very large number of spins. However, if you have a very very large edge, then you don't have to test for as many spins. Again, in the sample I grossly simplified things to make it easy to understand. I know most of your aren't betting the ECs. Testing individual numbers is going to require larger sample sizes than you think. Testing small samples is silly. Before testing, you need to have some idea as to what you think you're edge might be so that you'll know just how long you may need to test.

Best of luck,

Dr. Sir Anyone Anyone

 

Thank god you won't reply to my post.

It means I don't have to deal with this erroneous ONE and ONLY 1/37 basic probability that spells the death knell for systems betting.

Use your brains if you still have one, if 1/37 basic probability is all there is to it(ONE and ONLY) then why all those higher statistics studies?

Even those who don't have math background can logically figure it out.

Statistical significance math(nothing sophisticated) cannot help you find the anomaly for positive edge. Fact.

 

Fantastic, many thanks for this!

 

Thanks for this. Yes I can appreciate that things will be a lot clearer the more spins you do. However, the nature of my method is there is an assumption that there should be a large edge over the casino and it is tiring and time consuming to even put together 50 spins in one sitting hence the interest in small sample sizes.

 

One last matter and I'm done.

Notice that SirAnyone, MJ and Benas write only about statistical significance math. Nothing else. Fact.

Check their post history.

With this evidence you guys still don't get their agenda then too bad you deserve to be fooled.

 

This isn’t a strong area for me so I appreciate all the help I can get!
Thanks again

 

I wrote that you asked a good question. Check my first post.
Proper education is ALWAYS good for gamblers.
This is a good thread. Thumbs up.

However, misleading, false and agenda laced presumptuous info is damaging to forum.

 

I understand.

Slightly crestfallen that my most recent effort is “Not significant” but that’s fine.

Cheers

 

I wish this forum would have the ' Idiot ' button.

 

Although I cannot pinpoint exactly on the solution, that's how much I've already figure out.