Very minimal severe weather reports today considering the high probabilities and strong wording on the Day 2 and 3 outlooks. SPC did catch on by Day 1 and walked back the probabilities some, although even the small moderate risk maintained on the initial outlook was probably too much.
Obviously a good thing from a life and property standpoint, but it sure seems like their track record with high-end forecasts is not real good for several years now. For how rarely the high risk has been used since 2014, none of them in recent years have produced particularly noteworthy events (the type for example, that would prompt several posts of photos and discussion in our "Significant Tornado Events" thread). The most significant tornadoes of any given year (such as Beauregard and Lawrence-Linwood in 2019) have generally occurred on "sneakier" days.
Of course severe local storms forecasting isn't easy, and I'm just an amateur observer. This isn't meant as an attack on the SPC, because the entire weather enterprise has generally been on board with significant events occurring on high-risk days, and there have certainly been significant tornadoes on the high risk days. For example, that one in January 2017 had the southern Georgia long-track EF3. However, it was also the only high risk ever issued encompassing parts of peninsular Florida, and that area was entirely devoid of high-end severe weather.
The last time a high risk area truly "verified" with no reservations, in my opinion was April 28, 2014. Even then, that risk area wasn't introduced until 20Z with the outbreak already underway. The last time a morning-issued high risk area was really raked top to bottom by tornadic supercells in classic outbreak fashion was November 17, 2013 (Washington, IL EF4 among others). Since the nearly perfect placement of the rare 45% tornado prob area on that day in 2011, the last two times it has been used have been particularly underwhelming.
Again, not trying to attack anybody, but more looking for thoughts on the reasons (both meteorological and possible issues with modeling, etc) why event after event after event for years now fails to hit its ceiling.