A Tale of Two Pandemics - One Real, Another of Panic.
Here are two interviews of two experts in epidemiology (which is basically applying math and statistics to identify sources of pathogen outbreaks and predict their behavior). The first, Dr. Michael Levitt, being interviewed on the American streaming version of the China Global Television Network - not exactly what we'd call the "mainstream press" in America, and the second is a CBS Evening News interview of a person who, remarkably, is never identified by name (red flag) but is described as "one of this country's top experts from Harvard on viruses who has a startling prediction".
The first video (above) talks about there really being two different diseases - one in the Wuhan area where it all started, and one for the rest of the world. Now what this really means is, without calling out the Chinese government, that the numbers from Wuhan were clearly lies and I have already reported that several statisticians elsewhere have observed that ratios of reported infections to deaths fall within a clear statistical boundary around the rest of the world but the numbers coming out of China were aberrations that show clear manipulation or fraud. We use similar techniques to demonstrate unequivocally that certain players are colluding to cheat in online poker for example. Dr. Levitt makes precisely the right points about rates of increases being more important than absolute numbers and he used that to predict a decrease in the rate of increases in deaths which is exactly what happened. He points out that this is a serious pandemic and we all need to take responsible precautions - which given the long tail potential of this pandemic means that stopping school and other radical actions are now probably quite reasonable - and that we can see actions being effective and make it possible to see an end to this crisis in the near future. Also note the reporters questions don't presume a conclusion. This is a responsible and informative interview.
Now lets look at the CBS interview:
Norah O'Donnell is immediately identified and kicks off with a recital of scary and unsourced "worldwide" statistics and then follows with a "and that's not all!" bit of vitriolic prediction of dire dire consequences. Then she introduces the interviewer as Jim Axelrod but only identifies the role of the gentleman being interviewed and never gives his name! Jim, of course, starts off with another dire prediction of what "could" happen but is NOT an actual prediction - it's one possible estimate based on many different outcomes. Jim uses the word panic and just piles on the bad news. The guy being interviewed keeps throwing in "ifs" and ultimately says there's reason for being to be concerned and we should use that concern to turn into action. Jim never asks an unloaded question. He makes dire statements and then ends it with "do we?" or "right?" to nominally turn it into a question but it was never a question.
Now the Harvard guy never clarifies or sources his numbers or even bothers to try to talk about alternative outcomes. He sticks to unsupported presumptions and gross generalizations and provides precisely ZERO actionable data or advice. It's possible that he provides this information and was just edited out. (I've experienced this first hand.) He just says "almost inevitable" and other platitudes of dire outcomes. "Millions of people dying" is his only actual predicition. He gets asked one legitimate question about "what should we do to prepare" and he doesn't say what to do - he just says there's a reasonable chance that schools will close and public gatherings will be restricted. His one suggestion is to have business owners prepare to allow staff to stay home if they're sick. Thank you CBS & "country's best expert" from Harvard. We really couldn't have done it without you.
George P. Box said, "All models are wrong, some are useful." Well Dr. Levitt's model made predictions, and when those predictions came true it gave us more confidence that our understanding of how this pandemic will play out is correct and that certain behaviors are effective in containing it. That's quite useful if your goal is to save lives and reduce the impact on our economy to as little as possible. CBS's model has zero predictive value and is useful only in increasing fear, uncertainty, and doubt - FUD - what governments and our main stream press live and breath - much to our detriment.
Another thing this demonstrates is that information has to be judged on its own merits and not its source. Don't fall prey to the 3 "A's", Advertising, Anecdotes, and Authority. In this case CBS is pushing Advertising (they keep repeating fear to the exclusion of all else) and Authority (best expert). One would hope, given our free press, that the American news media would provide us with something truthful and useful and that a source coming from Communist China would be vague and full of propaganda. Now - to be fair - if Dr. Levitt's responses would have made China look bad then I'm fairly certain that they would never have been published. But, when the facts were with them, Chinese media had no problem publishing them and in doing so, actually helped our understanding of this thing. CBS - not so much.