That is what gets me curious in the first place. The fact Mythos scored so high, IMO, exposes some issues with this model: it is able to solve seemingly impossible to solve problems.
Without cheating allegation, which I don't think ANT is doing, it has to be doing some fortune telling/future reading to score that high at all.