One thing that has stood out from our past hurricanes in central Florida and also from watching those around the country is the real large propensity for folks to board up windows.
But it has also seemed, at least with our limited experience (with fairly weak hurricanes overall), that there was rather minimal breakage of windows left unprotected.
Certainly it's a complicated relationship, and things like debris quantity and exposure vary greatly. But nonetheless, I wonder if anyone has taken any basic statistics. Even those could greatly help in focusing preparation messages; saying "you're about X times more likely to have a window break at XXX mph versus XXX mph" would be quite useful, as would "only X% of windows break during a category 1 hurricane".
Certainly the stronger the wind, the more vital that windows endure, as any breakage could well undermine the structure at stronger wind speeds. But it'd just be great to be able to offer more informed statements if they exist!