First of all my mathematical training is nothing. I am trying to get these relationship tables properly for a practical system considering RB series in roulette.
If I will monitor series of 4 there are 16 possibilities:

BBBB
BBBR
BBRR
BBRB
BRRR
BRRB
BRBB
BRBR

RRRR
RRRB
RRBB
RRBR
RBBB
RBBR
RBRR
RBRB

Now comes my question: it is assumed that all spins are checked. not 4 in 4 spins but after every spin recent patterns of four are evaluated so I must consider all possibilities. Right?
For a series of 4 to be labeled a sleeper I must consider 16 spins monitoring for the full 16 combinations that can go in each of the sets.
Am I doing the calculations right so far for continuous play? A pattern of four should be called sleeper only if it goes over 16 spins without showing.

Quote from: wannawin on June 02, 2014, 08:35:50 PM
not 4 in 4 spins but after every spin recent patterns of four are evaluated so I must consider all possibilities. Right?
For a series of 4 to be labeled a sleeper I must consider 16 spins monitoring for the full 16 combinations that can go in each of the sets.
Am I doing the calculations right so far for continuous play? A pattern of four should be called sleeper only if it goes over 16 spins without showing.

Not really. I will try explaining why.
For example lets say a sequence RRBB BBRR RBBR. Now if you are planning to base your play on probabilities of patterns occuring and not occuring, you cannot overlap them. The moment you overlap them it becomes a continous set of occurences and is as good as randomly selecting a pattern as a sleeper.

One way you could improve on playign every spin is consider these are four different tracks of spins. So the above sequence will be read as

First track - RRBB BBRR RBBR
Second track - RBBB BRRR BBR
third track - BBBB RRRB BR
Fourth track - BBBR RRBB R

This way you could still play every spin without impacting the probability of patterns that we spoke about.

Quote from: Pockets on June 05, 2014, 09:47:20 AM
Not really. I will try explaining why.
For example lets say a sequence RRBB BBRR RBBR. Now if you are planning to base your play on probabilities of patterns occuring and not occuring, you cannot overlap them. The moment you overlap them it becomes a continous set of occurences and is as good as randomly selecting a pattern as a sleeper.

One way you could improve on playign every spin is consider these are four different tracks of spins. So the above sequence will be read as

First track - RRBB BBRR RBBR
Second track - RBBB BRRR BBR
third track - BBBB RRRB BR
Fourth track - BBBR RRBB R

This way you could still play every spin without impacting the probability of patterns that we spoke about.

Pockets thanks friend. Very good contribution. Maybe it is possible for you or someone else to confirm if these statistics are correct to call any RB pattern a sleeper? We all know that no matter what the pattern is but the number of elements.

Pattern - Spins
2 - 8
4 - 16
8 - 32
16 - 64
32 - 128
64 - 256
128 - 512
256 - 1024
512 - 2048
1024 - 4096

It will be quite appreciated and educational for the readers.