Actual Versus Forecast COVID for Past and Future Four Weeks

John Hopkins reported 471,567 total US COVID deaths as of February 10, 2021 and reports 514,700 as of March 1, 2021. This was an additional 43,000 deaths over 19 days. The seven-day moving average of daily deaths is still about 2,000 per day in the USA. The March 6, 2021 total cumulative deaths is tracking to 523,000.

The pandemic is tracking towards about the fourth most optimistic forecasting model of 11 major models used by FiveThirtyEight.

The March 6, 2021 projections made on Feb 11, 2021 of cumulative US deaths had a forecasted range from 512K to 546K.

The Feb 24, 2021 projections for March 20, 2021. Johns Hopkins currently reports 507,000 total US COVID deaths on Feb 24. The four middle projections had 536k-541k cumulative deaths for March 20. The four models in the middle of 11 models covered at Fivethirtyeight are forecasting 31,000 to 36,000 new deaths by March 20, 2021.

The updated cumulative US COVID death forecasts to March 27 now range from 535k-562k. This would be 23,000-48,000 additional deaths from March 1. Six of the middle-of-the-road projections are tightly bunched around 546k-555k for March 27. This would be an additional 31000-41000 deaths.

The Los Alamos and the Arizona models have a high-end ranging over 580,000 by March 27. Two of the 11 models have scenarios where the COVID variants cause a surge in cases and deaths in the US by the end of March. Three other models think high-end scenarios over 560,000 deaths are possible.

Two forecasting models (John Hopkins and IHME – Washington) out of eleven are optimistic that US deaths will be below 546k by March 27, 2021.

The end of March through April will show if there is renewed surge from COVID variants in spite of the new vaccines.

SOURCES – John Hopkins, Fivethirtyeight
Written By Brian Wang,

17 thoughts on “Actual Versus Forecast COVID for Past and Future Four Weeks”

  1. Obesity might be a factor just due to it causing more viral illness in general, but in this case I think it's the correlation with fat people not getting outside in the sun as much.

  2. Right. So if hospitalizations and deaths don't drop to negligible levels, at least one of these two things must be true: 1) vaccines don't work as advertised, or 2) hospitalizations and deaths are being overcounted.

  3. There's an obesity specialist who says we can mostly tell who's going to get hospitalized or die from Covid by their BMI level. 46% of Americans are obese. Blacks are twice as likely to be overweight and about that much more likely to get sick or die from Covid. Hispanics are more likely to be overweight, though less than twice, and more likely to get sick and die from Covid. Of course, these groups are also more likely be deficient in vitamin D too.
    This explains a lot, but hardly ever gets any media attention because it would lead to uncomfortable questions about personal responsibility and media and government support for lousy food options.
    Much better to feed the beast with expensive, risky, RNA vaccines with an uncertain future against a rapaidly mutating virus that most people will respond to like a bad cold anyway. Even among the elderly, the hospitalization and death rate is almost nil if there are no comorbidity factors.

  4. exactly… no reason to vaccinate the whole population but rather the target at risk population. The JnJ vaccine should be a huge boost too especially for populations like homeless and drug addicts that will be hard to get back for a second dose on time if at all …

  5. Can you get pizza delivered there? Do you have anyone to keep you company? If so, why would you want to leave?

  6. If you're trying to convince me that the pandemic is ending, you're preaching to the converted. My point was that, although hospitalizations may go down, they're still well above what is needed.

  7. No, you must hide under the bed until Congress passes at least $3 trillion of covid related spending.

  8. "Even if we look at only hospitalizations, that doesn't save us. A lot of governments won't hesitate to keep someone in hospital for testing positive, more for quarantine purposes than because their life is in danger."

    Hospitalization counts are a good indicator because hospitals have a real financial incentive to report people as covid cases. The hospital counts are plummeting and these counts are lagging indicators (though not lagging as much as death counts).

  9. If the people in charge wanted to end the pandemic, they'd just tell people to take vitamin D and zinc. That would pretty much do it.

  10. Rather than reciting these silly computer generated forecasts, the focus should be on developing the mean to vanquish the epidemic. Grandma knew better, Steam inhalation could reduce Covid-19 viral load. Every home that there is person quarantined should be properly humified. That and exposure to sunlight.

  11. Assuming that they get their act together and stop doing PCR tests with extremely high cycle counts. With inappropriate cycle counts, you can detect virus even in people who are immune and/or vaccinated, yet happened to breathe in some virus that day.

    Even if we look at only hospitalizations, that doesn't save us. A lot of governments won't hesitate to keep someone in hospital for testing positive, more for quarantine purposes than because their life is in danger.

    And what about death counts? Maybe. But we know a lot of people in those death counts died with covid rather than from covid.

  12. The evidence is strong that the vaccines are effective at preventing hospitalization and death, even for variant strains, and that most of those occur in the vaccine priority populations – which will be mostly vaccinated by the end of March, so a month later hospitalizations and deaths should be very low.

  13. Best to look at the second derivative of deaths (the acceleration/deceleration of the death count). In my county we are on track to have ~1 person die per day out of a population of 3.1 million. New cases in my county are < 200 per day for the last few days out of a population of 3.1 million.

    Despite numerous problems CA is managing to vaccinate ~200k people per day. This is ~100x the rate of new infections.

    I'll say it again: by the end of April things will be effectively back to normal in terms of new case counts.

Comments are closed.