The most difficult thing about forecasting ice is that it's such a narrow variable space to maintain that as the dominant ptype. I typically lean heavy on sleet because of that.
The problem I'm having with some of the forecasts I'm seeing is like for BGR: Roughly over the rest of the event there a forecast of 1 inch snow, 1 inch sleet, and 0.33 inches ice - off 1 inch QPF.
If you take conservative ratios of 2:1 sleet, 10:1 snow, and 0.7:1 FZRA, that's already over 1" QPF. I just don't think there's enough QPF for all these things to happen, either sleet robs from FZRA or vice versa.