There's a lot of data that goes into these outcomes
remember it's probable that the big man are starting out atrocious and so it's almost assured they will pop first time out thus skewing the numbers, and of course age enters into it.
This is a very good point. Age is being controlled for, so I don't think it's that, but the fact that big men likely have lower skills (and therefore may be benefitting more from high initial sublevels, or from "elastic" type effects) is certainly possible. Of course, if that were the case, I'd expect the same to be true of OD, and there training clearly decreases with height.
I just re-ran the calculation of the coefficients and updated the analysis page to see if the recent data had gotten rid of some of this, but the passing trend is definitely still there (and maybe even more dramatic). As hard as it is for me to believe that's correct, it's at least possible that passing training is faster for tall guys. Passing is one of the most trained things, so we actually have quite a bit of data on it. (Not as much as OD, but more than just about everything else).