Continuing from the last post.
As explained by Mladen:
Adaptive lookback stochastic
On the picture compared to "regular" (14,3,3) stochastic. It seems that it really does reacts much faster than the regular one (but I leave the exploration and conclusion of it to all of you)