For What It’s Worth . . . .

| May 29, 2020

The Institute of Health Metrics and Evaluation updated their Wuhan coronavirus projections for the US three days ago. I’d guess the update was connected with CDC’s recent release of information indicating the true infection fatality rate for the Wuhan coronavirus to be around 0.26%.

IHME’s core funding is provided by the Gates Foundation and the state of Washington (they’ve received close to $400M in grants from the Gates Foundation since 2007); they’ve received project grants and contracts from multiple other US and foreign sources over their history. I hardly think they’re members of that “vast right-wing conspiracy” some seem to believe exists.

Bottom line: IHME’s projection for total US coronavirus deaths by 4 August 2020 of this year (the limit of their projection) are now slightly less than 132,000. The precise figure projected is 131,967, with a 95% uncertainty range of 115,839 to 173,528.

At that point, IHME’s projected curve for total US deaths appears fairly close to (e.g., within a few percent of) its asymptotic limit. Eyeball estimation of their projection graph puts that asymptotic limit at around 140k if the projected figure of 132k for 4 August 2020 is reasonably close to reality.

The IHME Wuhan coronavirus model is not without its critics. However, most published criticism appears to predate the major update of the IHME model that occurred in early May.

And so far, best I can tell IHME model seems to have been the most accurate to date. If someone has a better one, I’d love to hear of it.

IHME’s Wuhan coronavirus resource page can be found here. IHME’s projections for various countries (their projections seems to show the US as the default, but you can select other nations if you like) can be found here.


FWIW, the midpoint of CDC’s estimate range for deaths associated with this past flu season is around 43,000. ( 140,000 / 43,000 ) = 3.2558+.

Category: Coronavirus, COVID-19

Comments (15)

Trackback URL | Comments RSS Feed

  1. Jay says:

    Hmm. It’s almost as if they INTENTIONALLY inflated initial projections in order to grab as much power and lock down people as much as possible. Odd.

    • thebesig says:

      After the CCP virus got underway, the Chinese shut down domestic travel in and out of Wuhan… But allowed international travel to continue in and out of Wuhan. They, like leftists in the US, took advantage of the crises to further their objectives.

      The Democrats/leftists/”progressives” served as useful idiots for the Soviets, they served as useful idiots for the terrorists, and now they’re serving as useful idiots for the CCP.

      In these three instances, they took actions, and advanced arguments, that aligned themselves with America’s enemies. They don’t care about the people they claim they stand up for. They only care about power and about having things their way.

      Conservatives represent what America is, and America’s will to fight. The Democrats represent the opposite to that. For a long time now, we’ve been in a fight for America’s hart, mind, and soul. Conservatives are for preserving such. The Democrats are for destroying them in favor of their fantasy society that works in their minds but doesn’t work in real life.

      Make no mistake about this… The movers and shakers of the leftist organizations, including the Democratic Party, are already on the socialist/communist end of the political spectrum.

    • LC says:

      No, the models do complex calculations based on things like the known mortality rate and transmissibility. Early on, due to this being a novel virus and a lack of widespread testing, the best estimate you can have for mortality is quite simply how many people you have dead divided by how many people you know who have had the virus.

      We didn’t have any sense of wider spread – there were no antibody tests, no random sampling of population, nothing that would indicate ‘true’ coverage, vs. known coverage. When all you have is dead and positives, even now we get a US mortality rate of 5.8%. But that additional testing -primarily antibody testing now- shows a much more widespread virus, with only a select few getting very sick.

      Plugging 5% mortality into a model will give you vastly different numbers than plugging 0.3% into the same model.

      No intentional misleading, just going by the numbers we had available, which is the most sensible approach when dealing with a novel pandemic.

      • Hondo says:

        I argee with you here, LC – with one minor quibble.

        I do not have any idea why everyone (including myself) missed what appears to be a wild overestimate in Imperial College’s (and other) original “worst case” projections regarding the total number of infections in the US.

        My understanding is that Imperial College’s projection of 2.2M US deaths was based on a 0.9% IFR. Working backwards, that implies around 244M total cases – or about 75% of the current US population.

        Call me skeptical, but that seems WAY too high a fraction to me for a disease that’s roughly as transmissible as the flu. As I said in a previous comment elsewhere, I don’t think we ever saw that fraction of infections in the US population with the flu even before we had flu vaccines.

        I might buy that number if Imperial College also assumed the Wuhan coronavirus had transmissibility roughly equal to either measles or chickenpox; both are believed to be 5+x more transmissible than influenza. But I haven’t seen anything to indicate that’s what was assumed.

        • thebesig says:

          In addition to what Hondo said, as time progressed, and numbers grew, these models could’ve been updated with the numbers that had been the case. More hard numbers became available as time went on. This means that more solid information could’ve been utilized… Changing the actual projections.

          This could’ve been done sooner and the “oops” our models were initially wrong moment would’ve occurred much sooner rather than later.

          It’s like tracked ship’s plot in the ship’s combat information center. Initial projections, based on limited data on the detected ship, provides a projection. Since fewer numbers are used, the error that creeps in ends up being large. But, as more data is utilized in the tracking system, courtesy of time passage, the tracked ship’s plot matches actual and projected track.

        • MI Ranger says:

          Don’t forget that these are the same group of people that had rehearsal in October 2019 to war game how they would react to a global pandemic.
          Although, all they seemed to focus on in that scenario was how to get the most funding, attention, and buy in for their products that their members sponsored…wow that seemed to have worked, they got almost everything they asked for except someone found a cheap decades old drug that does a better job then their new patented ones…all they had to do was smear its use! Turkey did not fall for it, and they came out quite well!

      • timactual says:

        “No intentional misleading, just going by the numbers we had available”

        Horsehockey. As you say, the numbers were bad. They knew it (they are “experts”, right?) and they made recommendations based on them anyway. That’s intentional. And it is certainly NOT sensible recommending the economy be shut down indefinitely.

        Even non-experts here and elsewhere knew and loudly criticized, correctly, the numbers as BS.

  2. Commissar says:

    I think thiese numbers are pretty solid. Some of the past prediction have been ridiculous.

    60,000 by August? That was a fucking idiotic prediction. Though it came out of a university and not the CDC. Still it was nationally reported.

    It was clear the day they released their numbers they were wrong.

    I cringe for the poor grad school slobs that got roped into working with the moron who led that team. They had to know something was wrong wi5 their model.

    • Hondo says:

      Actually, the number projected by the late-March IHME study was 81k, not 60k. See the Wikipedia article on IHME linked in the article.

      Yes, the March prediction of 81k deaths by 26 July was off substantially. But it was actually not much farther off, percentage-wise, than the 220k projection of total US deaths released at about the same that was contingent on implementing rather draconian measures. As I noted above, the asymptote for the most likely projection curve by IHME appears to be around 140k. (220k/140) = 1.571+, or a bit over 57.1% high. Assuming the current IHME projection is close to reality, (131k/81k) = 1.617+ – which means their March prediction was a bit over 61.7% low.

      My point? Calling someone a “moron” because an early projection is off by roughly 60% for a completely novel situation being modeled is unwise. The model IHME used in their late March study may have been completely sound. But early on, you simply don’t know enough yet to make reasonably good assumptions. You thus have to make guesses – and not particularly educated guesses, either. Make a bad guess, and you could be off far more than 60%.

      • Commissar says:

        I was not talking about the IHME prediction,

        I was talking a out the University of Washington model that received widespread press attention when Trump touted it.

        However, on April 9th the IMHE did indeed revise their 74,000 prediction down to 60,000. Which received criticism because it was seen as trying to please Trump since he quoted the University of Washington 60,000 number the day before

        • Hondo says:

          Thanks for clarifying your original comment.

          Haven’t been able to find any mention of that UW study (most searches I’ve done only mention the IHME projections), and I wouldn’t mind looking it over. A link would be appreciated.

  3. 5th/77th FA says:

    Lies, damn lies, and statistics. Bottom line to all of this is the Chinese Communist, aided and abetted by persons known to be pushing for a New World Order, unleashed a very nasty bug on the world. It is that simple. Bouncing numbers around just to bounce numbers around, who calculated what and when, don’t mean squat. The bug is here, their hands are dirty and now they know how to hit us and the wanna be dictators of the future know how far they can push the sheeple of the world.

    My sympathies to the ones who lost their lives is strong. My desire to string up the ones that caused this is stronger.

    Connect the dots…follow the money.