ROULETTE BIAS?

Hi there, i'm new to this forum, and i have a question about what advantage can i have to a roulette that has a bias.

I started all of this when i saw the show of history chanel breaking vegas, the episode of Gonzalo Garcia-Pelayo and how he won at roulette by seeking for biased roulettes.

so i did the same and went to my local casino, and start taking stats, and this is what i have after 15069 spins:

20 - 526 times
34 - 520
14 - 465
31 - 454
3 - 452
13 - 444
4 - 438
10 - 437
21 - 437
26 - 424
1 - 421
16 - 421
18 - 421
6 - 420
24 - 419
7 - 417
29 - 413
17 - 412
30 - 401
12 - 397
28 - 396
36 - 394
35 - 389
5 - 387
27 - 379
15 - 378
32 - 377
2 - 375
23 - 375
33 - 372
8 - 369
0 - 366
22 - 360
11 - 359
25 - 359
19 - 353
9 - 342

(its european so no 00)

i've been gambling the last 3 months while im taking the numbers, and i already have won about 13000 dolars, when the first day i started with about a 100, so im pretty sure this works

my question is, do i have an edge over this roullette?? what edge do i have? and what is the best way to attack this roulette with this information (actually im betting to 18 numbers)


thanks !! (sorry if my english is not perfect, its not my first lenaguage)

I'm sure that a lot of folks here would say that 15,000 spins is too small of a sample to show any bias, but if your figures are correct, I know how I would bet.

Did you happen to notice that the four most frequent numbers fall in the center of the two regions of the orphan's bet? I would make that bet continually, or I would put bets on the 20, 14, and 31, which are adjacent on the wheel and happen to be 3 of the four most frequent numbers.

Actually, I might also bet the 1 and 9, as those are two numbers just outside of the 20, 14 and 31 on the wheel. On the other hand, the 9 is the least frequent number of all, so I might leave that one out.

Have fun!

Tilted wheel? I don't know.
If you are consistently winning with this information...... go for it!!!!!

Note: The duplicate thread with the same question has been removed.

Quote: tupp

I'm sure that a lot of folks here would say that 15,000 spins is too small of a sample to show any bias, but if your figures are correct, I know how I would bet.

Them folks would be wrong. It depends on the degree of the effect you are measuring.

With strongly biased numbers 15,000 spins would be more than enough to show up any bias.

You're wheel could be fixed before you know what has happened.

Nice done ...
Hope you get much out of that wheel ...

Join us at myrulet dot com and we might help you more about bias and advantage play ...

Cheers

Are you sure it is the same wheel? If it is are you sure there is no 'maintainance' done on it at some point?
Anyway, with that number of spins the average is 407.27 for each number. If you believe you have a biased wheel that will cntinue with the same bias, how about betting 3 units on the top 2 numbers, 2 units on the next 3 numbers, and 1 unit on the next 6 numbers, which is 18 units per spin, since you seemed to have been betting that much previously. Good luck!

Quote: ruletero

and i already have won about 13000 dolars, when the first day i started with about a 100, )

Are you up 13k exploiting what may be a bias or with some other play? I'll tell you if you have a biased wheel if you tell me where it at....

I watched that "breaking vegas" documentary only last week. What I found interesting was his claimed level of success in LV, yet when in Madrid he waiting for 30,000 spins before determining flaws of the wheel. Then again some of it didn't make sense to any clocker, such as betting two numbers either side of 19 but not 19 itself.

Bias can be due not only because of any tilt, but also track wear, milk it to the max and while you can.

Quote: ruletero

Hi there, i'm new to this forum, and i have a question about what advantage can i have to a roulette that has a bias.

I started all of this when i saw the show of history chanel breaking vegas, the episode of Gonzalo Garcia-Pelayo and how he won at roulette by seeking for biased roulettes.

so i did the same and went to my local casino, and start taking stats, and this is what i have after 15069 spins:

20 - 526 times
34 - 520
14 - 465
31 - 454
3 - 452
13 - 444
4 - 438
10 - 437
21 - 437
26 - 424
1 - 421
16 - 421
18 - 421
6 - 420
24 - 419
7 - 417
29 - 413
17 - 412
30 - 401
12 - 397
28 - 396
36 - 394
35 - 389
5 - 387
27 - 379
15 - 378
32 - 377
2 - 375
23 - 375
33 - 372
8 - 369
0 - 366
22 - 360
11 - 359
25 - 359
19 - 353
9 - 342

(its european so no 00)

i've been gambling the last 3 months while im taking the numbers, and i already have won about 13000 dolars, when the first day i started with about a 100, so im pretty sure this works

my question is, do i have an edge over this roullette?? what edge do i have? and what is the best way to attack this roulette with this information (actually im betting to 18 numbers)


thanks !! (sorry if my english is not perfect, its not my first lenaguage)

You said you play 18 numbers, what numbers?

Is this a real Wheel? t has some isolated strong numbers

I would play 4 to 6 numbers.

How did you possibly get that much data? At 1 spin per minute(an impossible pace), that's 250 hours of spins.

Quote: surrender88s

How did you possibly get that much data? At 1 spin per minute(an impossible pace), that's 250 hours of spins.

There are ways to get the data without just sitting there clocking.

The data on the 20 pocket suggests p=1/28.46 an edge of 26%. Assuming the data is accurate that is a definitively biased number. So is the 34 (p=1.28.97, edge of 24%).

Given the bias on these numbers some of the other numbers are probably biased too, but progressively harder to be certain without Bayesian refinements.
Of course, there's a big chance when you find a wheel bias that you just screwed up somehow collecting the data.

This is an extremely biased wheel. A chi-squared test has a p value of 1 in 23,254,325,925,947,900. That means that is the probability a fair wheel would produce results this skewed or more.


Image source: eofdreams.com/data_images/dreams/roulette/roulette-02.jpg

However, the wheel doesn't seem to be obviously biased to one side. Note the 4th most common number (31) is right next to the 37th (9).

Here is the data in a table, for the benefit of others.

Number	Observations
20	526
34	520
14	465
31	454
3	452
13	444
4	438
10	437
21	437
26	424
1	421
16	421
18	421
6	420
24	419
7	417
29	413
17	412
30	401
12	397
28	396
36	394
35	389
5	387
27	379
15	378
32	377
2	375
23	375
33	372
8	369
0	366
22	360
11	359
25	359
19	353
9	342

Wizard, I have doubt of this wheel, I guess it isn´t real.
We wait for Ruletero to clarify how he won.

ok, so here is the thing, this roulette is an electronic roulette, not an actual roulette with croupier (not a digital one, just an automatic roulette where you bet in a touch screen). so it actually has a pace of one ball por minute.

the thing is, that the casino move it because they bought another one so they have to make some space, so i think that they actually changed the bias or whatever, but i started loosing so i stoped.

these are the numbers i had when i stopped:

34 679
20 675
14 611
21 597
31 584
4 581
13 579
3 576
6 561
10 560
17 560
7 555
26 554
1 550
16 549
24 538
15 536
18 536
29 530
35 525
28 518
0 514
2 509
30 509
36 509
27 506
23 505
12 504
19 501
25 496
5 492
22 488
32 479
8 478
33 476
9 464
11 458

but a month ago i started to take numbers from scratch, this is what i have at the moment

20 212
27 193
34 188
6 178
17 176
4 175
12 169
26 169
18 168
22 163
29 163
7 161
10 158
3 154
14 153
23 153
1 152
5 152
21 151
0 150
24 149
28 149
13 144
16 144
32 142
19 141
30 139
9 138
35 136
15 135
2 133
31 133
11 130
8 129
33 123
25 121
36 121


it shows that must the good numbers are still good numbers, but something happened that numbers like 22, 12 and 27 start appearing more, and numbers like 31, 21 or 14 decreased.

i want to know how do people would attack this roullette with this information

cheers :)

I would first make a chi-squared test to ascertain that there IS a bias.
If there is:
Then I'd play around with math to create probability distributions that could be +EV for a correct play. I would individually (unilateral) test the observed data against probabilities rising progressively towards that distribution. Keep the figures that are just significant and compute if you have an edge.

This is NOT standard statistics, either Bayesian or frequentist. Just dirty heuristic. Don't blame me if the results suck...

IMPORTANT ! Don't use your data while buiding winning distributions. If you do that, your test will be overconfident.

Quote: ruletero

ok, so here is the thing, this roulette is an electronic roulette, not an actual roulette with croupier (not a digital one, just an automatic roulette where you bet in a touch screen). so it actually has a pace of one ball por minute.

the thing is, that the casino move it because they bought another one so they have to make some space, so i think that they actually changed the bias or whatever, but i started loosing so i stoped.

these are the numbers i had when i stopped:

34 679
20 675
14 611
21 597
31 584
4 581
13 579
3 576
6 561
10 560
17 560
7 555
26 554
1 550
16 549
24 538
15 536
18 536
29 530
35 525
28 518
0 514
2 509
30 509
36 509
27 506
23 505
12 504
19 501
25 496
5 492
22 488
32 479
8 478
33 476
9 464
11 458

but a month ago i started to take numbers from scratch, this is what i have at the moment

20 212
27 193
34 188
6 178
17 176
4 175
12 169
26 169
18 168
22 163
29 163
7 161
10 158
3 154
14 153
23 153
1 152
5 152
21 151
0 150
24 149
28 149
13 144
16 144
32 142
19 141
30 139
9 138
35 136
15 135
2 133
31 133
11 130
8 129
33 123
25 121
36 121


it shows that must the good numbers are still good numbers, but something happened that numbers like 22, 12 and 27 start appearing more, and numbers like 31, 21 or 14 decreased.

i want to know how do people would attack this roullette with this information

cheers :)

At the first actuals, 19834 spins. No doubt of this wheel of bring biased.Just play 17 34 6 and 1 20 14 31

5645 spins of the 2nd data, 17 34 6 27 and 1 20 14

A confirmation.
Ruletero, as I said before I have serious doubts of this wheel.
In case it iis real you will have no problems to win a lot