Flat bets and the Law of the third.

[ThomasGrant]

Hi there.
Now, as you may know.
I am bot person.
That means.
If it aint in a bot.
Then I aint interested.

Anyway...
I have read countless posts over at VLS and some in here on Flat betting.
Plus I have also read some posts on this thing they call the law of the third.

I know that the two bots that I have.
Roulette Bot Pro.
Super Roulette.

Use some of these systems.
However.
If someone could explain to me how they work in plain English.
I would find it helpful.

Thanks in advance.

[mr.ore]

Imagine you are throwing a dice (1k6) and you write down numbers which came, the more numbers you have written down, the greater the probability, that you will throw one of these written down numbers.

Number            List (before throw)  Probability of throwing number already in a list
    2                    -                                       -
    3                    2                                      1/6
    1                    2,3                                   2/6
    2 REPEAT        2,3,1                                3/6
    4,                   2,3,1                                3/6
    1                    2,3,1,4                             4/6
    3 REPEAT        2,3,1,4                             4/6
    3 REPEAT        2,3,1,4                             4/6
    6                    2,3,1,4                             4/6
    1                    2,3,1,4,6                           5/6
    4                    2,3,1,4,6                           5/6
    2                    2,3,1,4,6                            5/6
    5                    2,3,1,4,6                           5/6
    4                    2,3,1,4,6,5 (DONE)            6/6

This works for any random event, even series, and so on. Bet selection based on this suppose that for some time it would work. But because any series of results have same probability, sometimes comes series like 1,2,3,4,5,6 or so (all six numbers different) and that makes balance. You can't know when this would happen. It is just following the trend.

Law of the third states that during 37 spins (one cycle) will on average appear 24 different numbers (2/3*37) and half of them(2/3*1/2=1/3 of wheel's numbers) will be repeaters. On average it is truth. Now you can try to abuse this fact in that you wait either that too many different numbers appeared and you will bet against this trend to continue (something must repeat) or you just bet numbers as they are appearing in a hope for repeater.

It does not work in the long term, all it tells is obvious. I have got a feeling, that using "law of third" is just another Gambler's fallacy. In the end it is gambling, anything can happen, and anything will happen, and wipe out your bankroll. On the other hand, it CANNOT HURT, it just does not matter where you bet.

[ThomasGrant]

Thank you.
That was a fairly good explanation.
Could you do a visual guide?
To both the law of the third and Flat bets?
Show some screen shots of bets on roulette table.
To make your point more clearer.
If you could.
Then that would be great.
As I am a very visual person.
That is why you see many videos from me.

[Bayes]

Just to back up mr.ore, there's a nice post at the other forum which covers this in some detail:

Some math, and the "law of the third"

Just a minor point, I haven't actually checked the formula given at the end of the post:

1 + N[1 + 1/2 + 1/3 + 1/4 + ... + 1/(N-1)]

If N = 1 you have 1/0 which isn't allowed. To be mathematically correct, Bliss should have added the constraint that N ≠ 0.

[mr.ore]

The law of the third is more a principle you can use for creating your own bet selection. Imagine you are flat betting EC, than you write down for example all triplets as they come, RRB, RBR, BRR, ... and you bet that some of them repeats or after some time you can rule out the last one to appear and play against it.

[mr.ore]

Each three spins there are 2^3 = 8 combinations of RB (ignore zero, it's tax you pay for playing), and if you have 7, then you have 7/8 chance, that another three spins one of the written down combinations repeats. You can see that the probability does not change, any combination have same chance, p = 1/8, and it applies for the last one missing. If you base selection on this, than you are betting that it will not appear for more spins than is an average. It is just illusion that something is more or less "due", the reason that it seems working is that the lower the chance of negative event, the longer runs of positive trends. On average they balance out, and zero kills all.

[mr.ore]

Now antithesis to "law of third": imagine that you write 7 random combinations of (RB)(RB)(RB), like RRB, RBB, BRB, BBR, RBR, BBB, RRR. Now you start progression against BRR, so you play RBB as your bet selection. The probability that you lose all three steps is still 1/(2*2*2) = 1/8, actually little lower because of zero(es). You are just gambling.

If you like to use bet selections based on gambler's fallacy, you could wait for time when for say 3*6 spins there would appear say only three combinations, then you would randomly pick up one of those which did not appeared yet, because you suppose that it will have more hits soon for balance to be reestablished, or at least it will just start hitting regularly. That's just another version of gambler's fallacy I think. But it does not hurt, and logic behind this is that event is statistically rare enough, that there would be long enough runs where it obeys what you want. In the long run it will balance out, but you are betting that it won't happen.

[esoito]

I've just added The Law of The Third (PDF file) to Downloads.

It's compiled from a study group at the 'old' VLS.