You mention fuel efficiency. Figure 6 shows that in some cases, looks like the threshold is between 60 and 80% load, the story flips such that adding ethanol to the mixture reduces sfc. In other words, less fuel is needed for the same amount of power. Meaning, if it takes X hp or kw to sustain whatever test speed you're measuring efficiency at, the amount of fuel you use is lowered. I think this is very important to understand. Figure 6 shows that if your engine is sufficiently loaded, nonzero ethanol can improve the amount of energy you get out of the same mass of fuel relative to pure gasoline, which I believe would be directly proportional to better fuel efficiency.
Now very notably, if less heavily loaded, it seems better to use 0 ethanol. The ability to choose lets it get the best of both worlds.