I don't know... Maybe higher pressure = lower airspeed + force of gravity on the water and they cancel each other out?
OK, perhaps that's a stretch.
I just recall reading in Hucho that, absent recesses, modern cars have attached flow on their windshields. Also I'd be more inclined to draw conclusions about airflow from tufts rather than much more viscous, massive and surface tension-y water droplets.
(Sorry for hijacking your thread, Christ.)