Critique of IHME model predicting coronavirus spread in the USA will fall rapidly

Institute for Health Metrics and Evaluation at the University of Washington has a model.

It assumes stay-at-home orders in every state in the next week.

You can look up each state:

Summary:

This model predicts the maximum deaths per day.
Florida: 136 on May 3.
Minnesota: 49 on Apr 19.
South Carolina: 30 on Apr 26.
We'll see.

Context:

I agree with this Carl T. Bergstrom thread that the Institute for Health Metrics and Evaluation at University of Washington model is too optimistic. It assumes social distancing is practiced and it is flawed by being based on fatalities so far.

The Institute for Health Metrics and Evaluation at University of Washington people are trying to get Florida's governor to order social distancing.

I have not seen any expert who thinks the IHME projection modeling is excellent. It is based on reported covid deaths and it is regularly redone to fit curves in other countries. It was just invented in March to help the state of WA. It is an estimate based on Wuhan and Italy.

But there is no other public updated national model. So, it is useful to give the public some idea since tens of thousands is hard to picture. But it fluctuates based on reported deaths. It does not take into account poor testing.
See threads below to learn more.

Huge: It only models the first wave and assumes only 3% will be infected. "Our model says that social distancing will likely lead to the end of the first wave of the epidemic by early June … an estimated 97% of the population will still be susceptible."

We see evidence that in hard-hit areas, 14% are affected in the first wave.

The point is that a huge number of people (97-86% of people) are still vulnerable to outbreaks after the first wave is slowed due to lock-downs.

Also the second wave begins not long after the first wave are supposedly stopped.

Other models:

GLEAM national model: gleamproject.org/covid-19#model April 11 peak time for deaths in the US.

LANL state model: covid-19.bsvgateway.org Cases will peak mid-April to mid-May.

CU Epi county medical capacity: columbia.maps.arcgis.com/apps/webappvie…