I wonder if weather model pages are pulling the HRRR ptype right from its own algorithm. Because it pretty much shows sleet and freezing rain as overlapping in all areas where there is either...possible that the graphics are pulilng fzra if it's present. Some areas that it shows fzra would definitely be sleet-heavy. For example, the ptype graphics show this as fzra on TT, whereas in reality it is probably a solid mix of ip/fz.