Same goes with CEOs
Show me the equivalent of advanced metrics in sports which can give statistically significance performance estimates. A handful of CEOs which make it their business to guide companies through reorganizations have track records which allow something close to performance analysis you can use for athletes, but that's an outlier. Even if they make strategy decisions other CEOs in the same market don't and have success with it, it doesn't necessarily mean anything (confirmation bias, until he does it a dozen time he could have been lucky ... the differences between bold and foolhardy can only be determined statistically).
or do you think the board just blindly sets the price?
Pretty much, market conform rates combined with some voodoo and nepotism.
And professional athletes do fail up more than CEOs in general.
They might not perform as expected after a transfer but if they fail to perform for a season it will be exceedingly obvious and they won't get a large transfer unless they can otherwise generate profits through marketability (also much easier to judge and measure than CEO performance).
Performance metrics are for people on the floor and the field ... not for CEOs.