2 ms·
This is neat but I think your derivation needs another step: a Bayesian prior that before this experiment, you thought the probability of A beating B was 50%.
by dmurray 6y ago
This is neat but I think your derivation needs another step: a Bayesian prior that before this experiment, you thought the probability of A beating B was 50%.