A double street has a max loss of 84 spins. At 59 spins the odds are 0.0000004600931320517899 of directly failing which is 1 in 2,173,913. So chances are, that won't be happening.

Also, if we let random.org choose our double street, it will improve those odds further.

When put into the simulator, the worst downward swing was around 300 losses (+1) versus wins (-6). So consecutive loss streaks easily crushes this progression.

This whole thing looks pretty unmanageable. It seems like one-dimensional configurations like this just don't have enough of an edge to work with.