Their predictions of future human behavior are bunk. First, look at the heterogeneity in behavior across the epidemiologists - indoor dining, hanging out withou

Their predictions of future human behavior are bunk. First, look at the heterogeneity in behavior across the epidemiologists - indoor dining, hanging out without social distancing and going to gyms! This lack of consensus indicates that epidemiologists do not think alike and hence that field is not a settled science. Second, life has mostly returned the normal in the far east where they have successfully contained the virus so far. This is before a vaccine. Human beings have short memories otherwise humans would not indulge in repeated destructive behavior on an individual (alcoholism) or societal level (bubbles, mass hysteria). Finally, after past pandemics like the Spanish flu, life did return to normal. The great depression more scarred human behavior than the Spanish flu ever did. Epidemiologists are essential during a pandemic to help us understand the virus, the properties of its spread and build models of human behavior along with risks/costs. We need to factor in their input along with advice from others. However, the notion that Epidemiologists are the best mediators of our behavior during and after this epidemic has been falsified. Why? Exhibit 1: Masking The consensus at the beginning of the pandemic and through most of the summer was that masks are not essential because we lack the evidence of it. Folks like Taleb et.al, Zeynep, Balaji, Howard et.al in the US who understand complex systems argued for masks back in Jan/Feb. This skepticism of masking because of lack of evidence probably bumped up death count by 50-90%. Exhibit 2: Epidemiology Models The models have been greatly off and as a modeler in an unrelated field, I could see the heterogeneity in the model predictions across models and intra-model volatility. When a bunch of models differ, the question is not which is right but are they all wrong? A lot of the models like the one from Imperial had predictions that moved by a lot and never converged to the observed behavior.