Hi KFB!

Thanks for your post, I really appreciate the HOT/NOT consideration. From this I'll try to expose our ideas.

Whatever the betting selection and patterns considered (random walks), itlr the NOT will overwhelm the HOT in frequency and not, of course, by a valuable proportional quantity.

There's no difference whether we're playing towards the NOT or the HOT, both are destined to get more probable short patterns than long patterns, then what happened by a huge/moderate density sooner or later will reverse in opposite probability.

The "sooner or later" instead of raw values could be better estimated by the "isolated/clustering" effect acting per every pattern examined.

A NOT shows up only when two things happen:

a) that pattern didn't show up once so far;

b) a previous pattern stopped its apparition.

A HOT shows up simply when a same pattern or more patterns we've decided to gather will continue to come out.

In addition, we know that each shoe is asymmetrically placed, meaning that clusters of something are slight more probable than everything else and now even an isolated/isolated pattern will get the same value of a clustering or clustering/clustering effect.

Theorically the best plan to adopt is to utilize two or three different random walks and wagering toward an isolated/clustering and clustering/clustering situation by progressively betting, better if we'd start the action after waiting some isolated/clustering situations.
If the random walks used get a different rhythm and besides some rare colliding situations, our probability to win is restricted into astronomical low sd values.   

Obviously and knowing that we're using a 0.75 probability, we are aware that isolated successions must show up, it's just a matter of conditional permutations issue.

The CFS and the asymmetrical patterns formation are just the icing on the cake.

Such effects are more than sufficient to erase and invert the HE by a simple FB scheme.

Here's an example extracted from our real shoes data.
 
We'll examine double and triple (exact 3) streaks as opposed to superior streaks (4 and higher).
C= cluster (at least two or more 2-3 streaks before a superior streak happens), I= isolated (a double or a triple streak came out intertwined by two superior streaks).

Each shoe features a first line (the common BR, and a second line being our main random walk)

For simplicity here we do not take care of additional factors.

C-C-C
C-C-C-C

C-C
C

C-C
I-I-C

I-C-C
C-C-I

C
C-C-C

C
I-C-C

C-C-C
C-C-C

C-C
C-C

C-I-I
C-C-C

C-C-C-C
I-C-C-C

C-I-C
I-C

C-C-C
I-C

C-C
I-I-C

I-C
C

C-C
C-C-I-C

I-C-C
I-C

C-C
I-I-C-C

C-C
I-C

I-C-C
C-C

C-I-C
C

C-C-C-I
C-C-I-C

I-I-C-C
I-C

C-C-I-C
C-C

I-C-I
C

C-C-C
C-C

C
C-C

C-C
C-C-C

I-C-C-C
C-C-C

I-C-C
C-C

C-C-C
C-C-C-C

C-I-C
I-C-C

C-C
C-C

I-C-C
C-C-C

C
C

C-I-C
I-C-I-C

C-I-I-C
I-C-I-I-C

C-C-C
I-I-C

C-C
I-C

I-C-C
C-C-I

I-C-C
C-C-C-C

I-C-C
I-C-I-C

I-C-C
C-C-I

C-C-I
C-I-C

C-C-C
C-C-C

C-C
C-I-C

I-C
C-C-C

C
I-C-I-C

I-C
C-I-C

C-C
I-I-C-C

C-C-C
C-C

C-C-C-I
C-C-C

C-C
C-C

C-C
C-I-I-I

C-I-C
C-C

I-I-C
C-C-C-I

C-I-C
C-C-C

C-C-I
I-C

C-C-I
C-C-C

C-C-C
I-C

C-I-C
C-C-C

C-C-C
C-C

C-C
I-C-C-I

C-I-I
C-I-C

I-I-C-C
C-C-C-C

C-C-I-C
C-I-C-C

C-I-C-I
C-C

C-C-C-I
C-I-C

I-I-C
C-C-C

C
C-C

C-C-C
C-C

C-C-C
C-I

C-I-C-C
C

I-I-C
C-C

I-C
C-C

C-C-C
C-C-C

C-C-C
C-C-C

C-C
C

I-C
C-C

C-C-C
I-I-I-C-C

C-C-C
I-I-C

C-C
C-C-I

I-C-C
C-C-C

I-I-C-C
C-C

C-C-I-I
I-I-C

C-C
C-C

I-C
C-C

C-I-C
C-C

I-C-C
C-C-C-C

I-C-C-C
C-C

C
I-C-C

C-C
C-I-C

C-C-I-C
C-C-C

C-C
I-I-C-C (6950)

Besides, notice the relationship between the two rows of any shoe, then the fact that a single C (not being so infrequent) means an entire shoe getting all doubles and triples.

Finally there's no way that the I/C ratio will stand strongly deviated toward the left for long as being slightly oriented toward the right side.

as.