[paggle]

Usain Bolt deviated colossally from optimal sprinting structure/style and beat the world record by a crazy number of standard deviations. And that actually happened. So these things are not just false because they're improbable.

[reascenda]

Poker is semi solved, not like chess is of course but to the point where good bots always beat good humans over a large enough sample.

There are objectively better and worse ways to play poker if you make some assumptions about the behaviour of the other players. In a small sample those assumptions may be false, but in a large enough sample it's extremely unlikely to be the case.

So it's extremely surprising to see a winning player employing crazily non optimal strategies (like betting 54o AI pre flop in a three way pot). That is just a losing play in the absence of information about the other players holdings.

The idea that he's reading people is absurd, even independently of the evidence of him not actually looking at people while betting. Live reads (to the extent they help at all) are incorporated in the win rates of the best players already.

[jimktrains2]

Bolt's record, while impressive, is by no means unexpected or "crazy". https://en.m.wikipedia.org/wiki/Men's_100_metres_world_recor...