As most of you know, google and youtube are clever now.  Whenever I load up youtube on the right are links to what youtube thinks I’m interested in because of my history.

Usually on the right hand side, there are poker related videos and the other stuff that I’m interested in.  So sometimes there are gambling related videos it selects for me.

On one occasion there was a link to Online Roulette, with the head line “Make $600-$700 a day playing online roulette”.

Now I know this not possible but my curiosity got the better of me, (“Maybe just maybe it’s true” yea right!!!!)

So I loaded the video and sure enough there was a guy doing a 10min video showing his strategy and showing his bankroll increase as he was winning. He obviously showed a 10 min period where he was winning which can easily happen playing roulette, it’s only just over a 2%  house edge (with 1 zero or over 5% with two zeros, but for arguments sake I will only talk about 1 house zero) so you can win quite easily in short periods but lose it all in the long run.

And he had comments and testimonies below saying how it’s working for them and how much money they are making and blah blah blah.

Well I simply took out a pen a paper and went through his betting strategy and simulated the wheel spinning all the numbers  0 – 36 one after the other, wrote down the outlay/profit of the numbers and colours he was betting on when they actually came up and the outlay/loss when the numbers he was not betting on came up and of course it was negative. It always is, you can’t get around the mathematical house edge. It’s impossible!!! Really!!! Maybe not!!! Read on.

For eg you will hear roulette players say they bet the 2 column strategy which to them is a winning strategy. I’ve debated with so many of these people and they just don’t see it. They believe it’s a winning system and no matter what I show them, as far as they are concerned they have the advantage because they are betting on 2/3 of the numbers so they must win.

Again, it’s quite easy to see the flaw in their strategy. They pick 2 of the 3 columns and bet 1 unit on each. It doesn’t matter which columns just pick 2 and put 1 unit on each column.

24/37 times you will win. 24 times you outlay 2 units, and receive back 3 units, with a net profit of 24 units, but 13/37 times you lose 2 units, 2 x 13 = 26.   26 – 24 is -2 units. It’s that simple, you lose.

So what do these non fail roulette video guys get out of it?? They have a link to their website and on their website there are links to online casinos where you can register and play their strategy. They obviously get some sort of kick back if you register from the link on their webpage.

And then there are the strategies of doubling your bet every time you lose and eventually you will win and of course that is true if you had an unlimited bankroll and unlimited table limit, but casinos have table limits so this strategy can’t work. You only have to lose 5-8 in a row and you are wiped out because you can’t double your bet anymore. It doesn’t take long. Lose $1 bet $2, lose bet $4, lose bet $8, lose bet $16, lose bet $32, lose bet $64 and that’s it, you can’t bet anymore, usually a $1 dollar min has a $100 max on black or red and 1 – 20 on a single number.

But anyways, while checking out these so called non fail systems, I pulled up an online roulette table which I have available through Titan poker and saw they have a min bet of 10c to a max of $200 on black or red.  1 to 2000 spread.

Hmmm, interesting I thought. That is a pretty big spread, maybe I can work out a strategy to take advantage of such a spread. 10c is not much but with the introduction of online roulette you can actually play it in turbo mode, that is you stop the animation of the ball spinning around the wheel and just click and get the next number and they have a display of the last 15 numbers or so, you can keep track of an improbable run of colours and work out a way to take advantage of it.

In a live casino, you would be lucky to get 20 – 30 spins an hour, even less if the table is busy, but online you can click pretty quickly and I was getting 40 to 60 spins a minute, but with betting it dropped to 25 to 30 spins a minute. That’s an average of about 1500 spins an hour.  Hmm, this is getting interesting, maybe, just maybe.

The only casino game which you can beat is Blackjack if you count the cards, the other reason is the frequency of when the advantage swings to the player. It is a fast game. Without that, making a decent hourly rate would be impossible. Apparently, don’t hold this to me, but it makes sense as card removed from a Baccarat shoe would influence the remaining cards. But apparently there is a counting system to get a player edge in Baccarat but because the frequency is so low it is pointless, you wouldn’t waste your time to win $1 and hour.  But now 1500 spins an hour with a spread of 1 to 2000, maybe something can be worked out.

I searched some websites for roulette probabilities and didn’t find much but found one that gave some decent information.

…..bility.php

 It had a table that showed 15 red or black in a row was 1 in 50,000. Wow that’s huge. The next thing I did was work out how many times I can bet starting from 10c up to $200. Doubling my bet each time I lose, and I will win 10c anytime one of the following bets hits.

10c – 20c – 40c – 80c – $1.60 – $3.20 – $6.40 – $12.80 – $25.60 – $51.20 – $102.40 = A total of $204.70

Ok, 11 times. I have to hit on my 11^th bet or I lose $204.70. I looked at the table and 10 in a row was 1 in 1,350, so the 11^th would be roughly double that 1 in 2,700. Hmm, that’s not good if I’m getting 1500 spins an hour, it’s going to happen once every  2 hours. That’s gonna lose money.

Then I thought, why not wait for 4 colours in a row before betting. That will give me the 15 chances I need.  That will only happen 1 in 50,000, in  You can click a spin without betting. So spin the ball until I get 4 colours of black or red and then start my betting sequence after that. If 4 blacks in a row came up I started to bet on red and visa versa. It surprised me how often 4 in a row came up, the frequency was quite good. Making 10c a spin, you gonna need some frequency.

I was making about $12 to $15 an hour.  The 1 in 50,000 was going to happen every 30 to 35 hours. So lets say it comes after 30 hours. I’ve made $375 – the $204.70 (I lose when 15 colours in a row come) leaves me a profit of $170. Divided by 30hours = $5.67 an hour. By jove I’ve done it. I’m beating roulette. Ok $5.50 an hour is not much but I’m thinking I can get a bot made up to play it 24 -7 that’s gonna be almost 1k a week for doing nothing.

Between my wife and myself we did about 20hours or so, and each time we started to get to the $12.80 bet or higher, we were thinking, is this it?? Is this the 1 in 50,000, but alas we kept hitting our colour and kept winning. We had a few $51.20 bets but never a $102.40 bet. We were so happy, over $200 profit, this is too easy, can’t be possible.

After coming down from the clouds and reflecting on what we were doing. I started to think, this can’t be right, I’m missing something. I wanted to make sure I wasn’t missing something, before I invest in a bot to do this for me.

I know how to program so I wrote my own crude roulette simulation to try out my strategy over millions and millions of spins. This will settle it. The long term, the expectation will settle any of my doubts and fears.  If my simulation confirms my strategy, then that’s it. I’ve done it!!! I’ve beaten roulette!!! Many have tried for hundreds of years, but I’m the only one that has done it. I’m smarter than any before me. I’m a genious.

I ran it over 100,000,000 spins, a pretty good sample size, usually you want more for complete accuracy, but for my purposes it would be ample.

 I wanted to know the frequency of how many 4 colours in a row would come up over 100 Million.

Spins: 100,000,000

4 Reds in a Row: 2,723,092

4 Blacks in a Row: 2,725,117

Total 4 in a Row: 5,448,209

1 in 18

Then I wanted to see how many times 15 in a row would occur, should be around 1 in 50,000

Spins: 100,000,000

15 Reds in a Row: 1025

15 Blacks in a Row: 940

Total 15 in a Row: 1965

1 in 50890

Ok so every 100,000,000 spins i am going to win $544,624   (5448209 -1965) x 0.10

And lose $402,235 (1965 x $204.70)

So a net profit of $142,389.

You can imagine how excited I was starting to get. My simulation confirmed what I was doing. Can this be possible? Good grief I’ve discovered plutonium.

All I need to do now is code in my betting strategy just to make sure.

Below are the results when I added my betting amounts.

Spins: 100,000,000

Number of Colours in a Row before betting:  4

15 Reds in a Row: 1890

15 Blacks in a Row: 2117

Total 15 in a Row: 4007

1 in 24956

Total Profit    -$432,000

Hmmm, must be a bug in my betting logic, this can’t be right, plus it says 1 in 24,956. Somehow my coding logic screwed up the rest of the program. So after hours and hours of debugging , couldn’t find anything wrong with the logic and coding, but somehow deep down I was thinking there must be something wrong with my logic. More a blind hope.

Then it hit me like a ton of bricks. Those of you who have read this far, probably already worked it out and are calling me an idiot, but I was so blinded by my calculations and the thought of this actually working, that I didn’t see it. What a fool I am.  So disappointed in myself that I thought this was even possible. 

What is the difference between running 15 black/red in a row and betting 15 in a row?? Yep, you got it, the dreaded house number. The “ZERO”. Once in the betting sequence if a zero hit you still have to continue increasing your bet. When I was actually doing it on Titan it seemed like it was no big deal. Occasionally a zero hit but we just continued to increase our bet and still got our colour before the 15^th bet.

Surely one extra number can’t make that much of a difference anyway. Only one way to find out, run the simulation with the zero included during the betting sequence.

Spins: 100,000,000

Add zero to bet sequence count = true.

15 Reds in a Row: 1960

15 Blacks in a Row: 2043

Total 15 in a Row: 4003

1 in 24981

Hmm, ok so no big deal. Let’s see what happens for 16 in a row including the zero.

Spins: 100,000,000

Add zero to bet sequence count = true.

16 Reds in a Row: 996

16 Blacks in a Row: 1008

Total 16 in a Row: 2004

1 in 49900

Great, so all i have to do is wait for 5 colours in a row instead of 4 and it should work. Right?? Wrong?? The frequency of 5 in a row is gonna be less than 4. Let’s run it.

Spins: 100,000,000

5 Reds in a Row: 1,325,483

5 Blacks in a Row: 1,326,011

Total 5 in a Row: 2,651,494

1 in 37

Ok so every 100,000,000 spins for 16 in a row i am going to win $265,949   (2,651,494 -2004) x 0.10

And  lose $410,218 (2004 x $204.70)

So a net loss of -$123,560.

No matter what you do, it always works out the same. You lose.  Wait for 6 and then bet or wait for 7 and then bet etc, the frequency decreases but that huge loss comes and wipes you out.

When I reflect back at it now, I realise how foolish I was. The bottom line is that there are 37 numbers and you get paid 35 to 1. No matter what you do, you can never ever get past that mathematical edge.  

It was a fun exercise for me, and a lesson learnt.

Hope you enjoyed the ride as much as I did.