The phrase is "the score that an expert disc golfer would be expected to make on a given hole with errorless play". I choose a 1000-rated player as the expert. Not the whole group of 1000+ rated players, and not Paul.
If the wind was backwards and slower than ordinary, then I wouldn't set par based solely on that round. Unless there was no other data. However, we have nine rounds of data now. If the winds were extraordinary for only one of those rounds, the other rounds will have the most influence on those rounds anyway. If the winds were backwards for 3 of those rounds, we have to start to question whether that's not just part of the range of ordinary.
This is, I think, the first time I've been able to look at so many comparable rounds to see how weather conditions and just random chance might affect the results.
Note that with just one round of data, about 10% of the holes are questionable. With three rounds, about one hole per course per year may be questionable. Which is about what I would expect. Some holes are just going to be near an edge condition and bounce back and forth between two suggested pars.
Combining all the rounds, we have almost 600 player-rounds of data. Based on that, if I were setting par, I would set it to 59. Unless someone knows for sure that the data in one or two of these rounds were under extraordinary weather conditions and should be thrown out.
(I'm not sure we got there the same way, but that's how I go there.)