fivethirtyeight.com/... For those who haven't seen it yet, there's a new piece by Jay Boice at 538 about what the "expert consensus" was as of Apr 14 regarding the likely toll from Covid-19 by May 1. The main takeaway...
The expert consensus is that the U.S. will have reported around 47,000 COVID-19 deaths by May 1, with a 90 percent chance of having between 32,000 and 82,000.
Seriously?? This is what passes for pandemic predictions over at 538 these days? Of the 18 "experts" contributing to their COVID Tracking Project, all but 2 thought there would be 50,000 or fewer deaths in the US by May 1, with a "consensus" estimate of 47,000.
Granted this survey was conducted at the beginning of the week, when US fatalities were still south of 24,000 and accumulating at a rate of 1500-2000 per day, and before the change in classification of nursing home deaths that dumped close to 4000 "new" deaths onto our totals, but their "90% confidence" window went as low as 32,000, with several of the individual survey participants thinking it could even be down around 20,000! -- did they actually think this was some sort of zombie virus and thousands of the already deceased would somehow be resurrected by May 1?
Do any of these forecasters ever bother to look up from their models and take note of the actual casualties accumulating day by day and week by week? They tend to remind me of TV weather persons who keep insisting it will be a sunny day when they can't be bothered to look outside once in a while to notice the pouring rain. The author notes that their forecast aligns quite closely with the "official" IHME that was also last updated Apr 13, and which most people still cite as the gold standard of such predictions -- for reasons I've never been able to fathom (confirmation bias anyone?).
Even more ridiculous, the survey participants were also asked to predict how many states would have 1000+ Covid-19 deaths by May 1 at a time when just 3 had reached that threshold. Their consensus was that there was a 40% chance there would only be 7 or less by then (when in fact CA and CT became the 7th and 8th states to hit that mark today), the most likely result would be 8-11, and that there was only a 31% chance of 12 or more