Kilian, I´ve got a question. Let´s assume we have a fractals (1,0,0,1,0) within ZZ leg (it should be up leg, because the first fractal is 1, right ?). We want to compare it with ZZ legs from history. We would do this like with "next bar" prediction (4-13 seq).
1,0,0,1
0,0,1,0
1,0,0,1,0
Then we add next fractal (0 or 1) and we have (1,0,0,1,0,0) or (1,0,0,1,0,1) and we have these subsequences:
1,0,0,1
0,0,1,0
0,1,0,1
0,1,0,0
1,0,0,1,0
0,0,1,0,1
We compare it with statistical results and try to predict what possibility is more likely.
BUT
We also have to consider a ZZ length. GG and you said we should look at ZZ length +-3 bars. Let´s assume that our ZZ leg looks like this:
1,-1,-1,-1,0,-1,-1,-1,0,-1,1,-1,0
It is leg with 13 bars. Therefor we should compare this leg and fractals only with legs which contains this sequences (1,0,0,1),(0,0,1,0),(0,1,0,1),(0,1,0,0),(1,0,0,1,0),(0,0,1,0,1) and their length is 13+3 = 16 bars or 13-3 = 10 bars. So we search these sequences only in 10 - 16 length ZZ legs, right ? It is important in what sequence are all 13 bars in leg, or it doesn´t matter and we look only at ZZ length ? I mean this:
1,-1,-1,-1,0,-1,-1,-1,0,-1,1,-1,0 is the "same" as
1,-1,0,-1,-1,0,-1,-1,-1,1,-1,-1,0 ?
I guess it is not
.. 
We try to predict end point of ZZ leg. In up ZZ leg we wait on ZZ leg end prediction and concurrently opposite bar direction (in uptrend we wait on down bar prediction).
Am I right, or ? I hope it does make sense what I wrote.
Thx, K
