IHME projections

arguablywrong has taken a look at the projections the Feds are using for planning.  He has some criticism of some features of their model: unfortunately he is almost certainly correct.

First, they accept CCP data. I am not sure what actually happened in Wuhan, but I am sure that A. things were at least as bad as admitted and B. the CCP (eventually) reacted with overwhelming energy, in a way that seems to a fit a much bigger disaster than  admitted.

AW thinks that deaths are significantly underestimated in several countries (certainly Italy, Spain, and France) and that IHME errs by not considering this. Correct.

IHME  model assumes that state that do not yet have infection measures will do so with a week.  That would be nice.

They assume that once controls are imposed, the epidemic will shrink just as fast as it previously grew. If the original R0 is 3, the post-control R0 must be 1/3rd. Uh, why?

“They assume that state-wide infection controls are equivalent in effect with the controls the CCP imposed in Wuhan.” AW calls this assumption crazy, but surely a stronger word is called for.  In their first try, the CCP tried these measures, which are more intense than any yet in the US:

  • Blocking outward transportation from Wuhan
  • Closing public transit and vehicular traffic inside Wuhan
  • Compulsory mask-wearing in public places
  • Cancellation of public events
  • Self-quarantine of confirmed or suspected cases

That dropped the R0 from around 3.5 to 1.25 – which wasn’t enough.  They moved on to

  • Full quarantine of confirmed or suspected cases (i.e., extraction to a separate quarantine site), including contacts of confirmed cases
  • Temperature monitoring of all residents
  • Universal and strict stay-at-home orders for all residents

All of these errors are in the same direction.

This entry was posted in Uncategorized. Bookmark the permalink.

81 Responses to IHME projections

  1. Allen Sheep says:

    “First, they accept CCP data. I am not sure what actually happened in Wuhan, but I am sure that A. things were at least as bad as admitted and B. the CCP (eventually) reacted with overwhelming energy, in a way that seems to a fit a much bigger disaster than admitted.”

    I agree on point A. However, on point B, an alternative explanation is that the CCP knew that a weak response would be worse in the long run, so they decided to have an overwhelming response.

    Just consider as a hypothetical: if the CCP data were correct, was the CCP’s overwhelming action an optimal policy, or would they have been better off with a weaker response? If their strong response was an optimal policy, then it seems totally plausible to me that they simply used their intellect to select the correct response.

    • Rosenmops says:

      Too bad the CCP didn’t use their damned intellects to nip the thing in the bud in December instead of trying to cover it up until near the end of January.

      • reiner Tor says:

        That’s too bad, but they had no forewarning and probably lower level officials botched the response. It’s not like the initial response of Western governments was much better, despite ample information on the nature of the disease and transmission and even genetic code of the virus.

        Meanwhile, Western governments knew fully well what they were dealing with by January 23 at the latest. They should’ve prepared for the coming storm, by organizing mask production, test kit mass production, etc. Instead they showed appalling incompetence. What makes you think that if they knew slightly more or slightly earlier they would’ve done any better?

        Regarding nipping it in the bud. Do you think there is a government on the planet (any government) which would’ve shut down a city of 10+ million people to stop a virus which may or may not jump from human to human and may or may not be dangerous? Me neither. Test kits weren’t yet available in December, so the South Korean method was surely unavailable to them. (Besides, would Western governments have criticized the Chinese for muh human rights violations if they did so? I think so, too.)

        Perhaps (it’s far from sure) they were late with the quarantine by a week or two, but I’m pretty sure that it took their political leadership some time to realize the enormity of the situation. They were still way faster than any other non-East Asian government on the planet.

        As to hiding the true numbers after January 23, what does it matter? It was always clear to anyone numerate that the numbers would grow exponentially, so 100 today could mean 100,000,000 in a few months, even if their true numbers were an order of magnitude higher, that would only mean that they were lying about being 2 weeks earlier on the curve than in reality. What difference should it have made? Is the complaint basically that Western governments were innumerate, and didn’t understand that the Chinese numbers only represented an early point on an exponential curve, and that if they were shown a 2 weeks later point on the same curve, they’d have better grasped the problem? I cannot really think of a more devastating criticism of Western governments and their incompetence than that.

        • EldnahYm says:

          Cancelling large public events, like the massive banquets in Wuhan and not covering up what happened might have made a difference. Expecting Wuhan to have been shut down early might be unreasonable, but those simple measure were not. They simply screwed up. Also the local governments used national Chinese laws against spreading rumors to shut down people who were warning of the virus. So it is false to lay blame solely on lower level officials.

          • reiner Tor says:

            Yes, they could’ve done better, but once they realized the gravity of the situation, they were quite efficient. Western governments either didn’t realize the gravity of the situation until late February or in many cases even March, or they did, but still didn’t do anything. Hard to say which explanation is more damning.

            All I’m saying is that realistically speaking (based on what Western governments were doing) the Chinese couldn’t be expected to do much more than they did. They were idiots until late January, but others were even bigger idiots for much longer.

            • saintonge235 says:

              Your conclusion of Chinese efficiency requires assumptions that are extremely dubious, such as believing the measures China took against the CCPVirus worked as well and as quickly as they said.

              Your conclusion that other countries were “even bigger idiots for much longer” seems to require that every country should have expected China and WHO to lie about the epidemic long after it was known to be serious by China. Yet you do not make that assumption.

              • reiner Tor says:

                Even if Chinese numbers were inaccurate, it only meant they were showing numbers from two weeks earlier.

                And anyway, they were publishing data such as:

                R0 of around 2-4 (later proven to be reasonably accurate)
                mortality rate of 2% (ditto)
                hospitalization required for a very long time (ditto)
                comorbidity statistics (ditto)

                So what they were lying about was the exact number of deaths and cases, but it just meant they said they were a couple weeks (or a month, or a week, or a few days, who knows) earlier in the curve than in reality. It shouldn’t have mattered if Western governments weren’t drooling morons.

                Western governments based on the R0, mortality, hospitalization, etc. statistics should’ve known there was a risk of a major epidemic, so should’ve

                closed traffic to and from China (Trump deserves some credit there)
                started to stockpile PPE for medical personnel and masks for the population in general, and organize their mass production on a scale not seen before
                started to stockpile medicine, and ramp up production etc.
                started to prepare for a lockdown to be implemented swiftly and efficiently

                That’s regardless of whether there were 3,000 or 30,000 or 300,000 or 30,000,000 deaths in China.

              • EldnahYm says:

                Considering how much money governments spend on intelligence, knowing that the Chinese were lying about the epidemic shouldn’t have been that hard. The Taiwanese were certainly under no illusions about the situation. Western governments did not want to do anything about the epidemic. Only after it blew up in their face did they start taking measures.

              • reiner Tor says:


                Also regarding the nature of lies. The Chinese didn’t lie about the numbers that matter (R0, CFR, time of hospitalization lasting several weeks, etc.), only numbers that don’t (at exactly which point they were on the curve when they started their hard measures). It didn’t matter at all whether the Chinese had 3,000, 300,000, or 30,000,000 deaths, what mattered were the R0, CFR, time of hospitalization, and other statistics, which were consistent with later statistics from places like South Korea, so must’ve been fairly accurate.

                The number of deaths were important in that most people are dimwits who don’t understand exponential functions and keep comparing deaths from an epidemic to deaths from car accidents. I still keep hearing from the iT’s jUST tHE FLu crowd that the flu kills more in an average year than Covid-19 killed so far. That SARS-CoV-2 is likely to infect way more people (because no partial immunity, and perhaps more infectious), that its CFR is higher, etc. don’t seem to matter to these people. On the Karlin blog (which I believe you also regularly visit) I keep debating a few otherwise intelligent (?) people about how the 1968 flu was supposedly worse because it killed more.

                So if China announced that actually 30,000,000 people died of this, this would be a good thing because then the iT’s jUST tHE FLu crowd would at least stop their “arguments.” But what if China were truly competent, as competent as Taiwan or Singapore?

            • EldnahYm says:

              Clearly most Western countries were and are incompetent. In any sensible country, a good old fashioned purge of public health officials would be in order after their performance. However, countries like Taiwan were taking the coronavirus threat seriously before the Chinese were. Compared to some other East Asian countries, including other Chinese East Asian countries, what China did is less impressive.

              China deserves credit for being able to take such strong measures, but they deserve blame for causing the virus to get out of control in the first place.

              • reiner Tor says:

                China was clearly more competent than the West, but less competent than South Korea, Singapore, or Taiwan. Each of these three significantly richer than China, so it was to be expected, nothing extraordinary. (By the way Japan seems to be less competent than China, at least so far, despite its significantly higher level of wealth.)

                China being more competent than the West is something out of the ordinary, in my opinion.

              • EldnahYm says:

                “Each of these three significantly richer than China, so it was to be expected, nothing extraordinary.”

                I agree with the majority of your post, but I’m unsure about this statement. One of the important lessons of this crisis(although all of this was obvious already) is that easy, common sense measures like closing down borders with high risk countries, close monitoring of people entering the country, quarantines, shutting down large events, voluntarily staying at home, and widespread use of masks can do a lot to slow down the spread of disease. This is well within the abilities of middle income countries. I’m not sure it’s necessary to be rich. So while it may not be so extraordinary what say Taiwan did, that doesn’t necessarily mean it’s much more difficult for a country like China to do the same.

                Also, medical capacity is probably more important than overall wealth. I’m not sure China is particularly poor in that area.

        • gothamette says:

          A charming story from the old world.

          A rabbi was counseling two people with a dispute.

          He listens to the first person and at the end of his story, the rabbi say, “You’re right.” The second guy gives his side of the story, and the rabbi says, “You’re right.”

          The rabbi’s wife sticks her head into the room and says, “They can’t both be right.”

          The rabbi says, “You’re right!”

        • gothamette says:

          “Meanwhile, Western governments knew fully well what they were dealing with by January 23 at the latest. ”

          Yes – but – didn’t the Chinese allow people to leave China after that?

          Of course, Western governments shouldn’t have allowed anyone IN after that, but still.

          What is so significant about January 23? My head is spinning with so many new facts, the meaning of that date escapes me and I’m tired of having 34 tabs open at once.

          • reiner Tor says:

            Has there ever been a government on the planet which stopped outbound flights because of an epidemic within its borders? Must be a pretty rare occurrence. I don’t know of any instance from this crisis or any other.

            Though you do have a point. The Chinese kept complaining when other governments (very slowly and late in the game) started closing their borders to them. Okay, so another point where China was definitely worse than the Western governments. (The other being the police harassment of the doctor in Wuhan.) There are multiple points where Western governments were worse.

            • saintonge235 says:

              You keep saying that Western govts. were worse. How were they worse? Based on what information available when, what is it you think they should have done, and what chain of reasoning should have led them to do it, based on the evidence available then?

              • reiner Tor says:

                I wrote above. They knew the R0, they knew the approximate CFR, and based on these two they should’ve known an epidemic was coming their way.

                I also wrote above what they should’ve done. They should’ve prepared to mass produce f#%ing face masks (we’re not talking space rocket technology here) and other PPE, medical equipment like ventilators, medicine, and similar things. Did any Western government place an order for ventilators in late January or early February 2020? Okay, maybe ventilators are expensive. There’s been a face mask shortage in basically each European country since late January. Did any Western governments do anything to build up a stockpile? To build up production capacity? (Again, we’re not talking about some space rocket technology, it should be within the capabilities of most countries with a population above 1 million, even third world countries should be capable of producing some kind of face masks.) Did Western governments even think about substitutes? (Now it has unsurprisingly turned out that home-made face masks are actually better than nothing, especially if your goal is to protect others from your droplets. So making them compulsory helps slowing the spread of the virus.) Did they do anything to create plans to organize making it compulsory?

                No, instead they were spreading propaganda how face masks were worthless. That’s what each Western government was “contributing” to the solution. Basically the first time they started taking it seriously was when it appeared in their countries. In Spain they were encouraging a feminist rally a few days before ordering a lockdown, so that’s pretty similar to the huge Wuhan banquet, except the Wuhan banquet was happening at a time when it was still unclear what kind of disease they were facing (apparently the party leaders believed it wasn’t spreading from human to human, or at least not very efficiently). In Spain they knew or should’ve known (had they done their jobs) the approximate R0, approximate CFR and the other statistics.

                If you don’t find this level of incompetence truly shocking, then I don’t know how you can criticize the Chinese for what they did in December and January, when they didn’t yet know whether it spread from human to human, or how efficiently, they didn’t yet know the CFR, typical length of hospitalization, danger to medical personnel, etc.

            • gothamette says:

              Two journalists and a dissident businessman have “disappeared.”

              Communism sucks. I think that’s been conclusively proven.

              • Paul Conroy says:

                Yes, people who tried to warn the West have “disappeared”.

                China has lied about everything throughout and is still lying!

                Obviously, “Reiner Tor”, is a ChiCom! 🙄

                China is NOT managing this crisis better than the West. China is totalitarian, so can randomly murder citizens, and do things like “weld apartment doors shut”, actions not available to the West.

                China is a disgrace and ought to be sued for $Trillions.

                They have sold defective test kits to a number of European countries. Spain’s test worked only 30% of the time and gave “False Negatives” 70% of the time. China also distributed tests that were infected with Covid19 to Europe.

                China have done everything they can to spread this virus and kill Europeans.

                China has declared war on the West!

  2. Paul Conroy says:

    Yes, I knew this was serious in early January as:
    1. I watched the smuggled video from a Wuhan doctor, trying to warn the world. Then the Chinese ordered him arrested and he was set to be executed, but supposedly died of Covid19 before he could be.

    Saw video of Chinese authorities welding the doors of apartment blocks shut, with all residents inside!

    • NamelessNobody says:

      Hong Kong-based Expat so perhaps a bit closer to the fray back in January, but still we supposedly live in a global online world so I was utterly gobsmacked that some of the videos I was seeing (copied over to Twitter, mind you; so I didn’t need WeChat or Weibo to know about them) didn’t appear on global news sites or in feeds of major opinion leaders.

      I mean truly medieval stuff like people dropping dead in the street, apartment gates being welded or chained shut with inhabitants inside, provincial borders with bulldozed earth berms blocking roads, villages with signs warning off outsiders and guards ready to provide physical enforcement, etc., etc.

      I knew of it before the Jan 23 Wuhan shutdown, but the moment the CCP shut down a city the size of Wuhan it was obvious to any intelligent person that something big was happening. Then there was this deluge of uncensored content making its way through the Great Firewall … this stopped about 10 days after Chinese New Year when all the Censors and 5 Cent Army went back to work after their vacation.

      I just can’t get over the ignorant blindness of our supposedly cosmopolitan global elite. It was all there for the viewing. Hong Kong Government is a bunch of Quislings, but the local people and the Taiwan Government and People are very familiar with the way things work in China and both immediately took steps to protect themselves. So what were all our Embassy Staff, Consular Staff, foreign correspondents doing? Jaysus wept!

      • Anon says:

        Judging by the numbers in Japan and Taiwan this virus barely rose to the level of an inconvenience. Do we know why that is?

        • Pincher Martin says:

          Because they and a handful of other East Asian countries and cities took it seriously before it ever reached them. The West did not.

          • Anon says:

            Most of Asia Population Total Cases Total Deaths
            China 1,438,000,000 81,669 3,329
            India 1,380,000,000 3,588 99
            Indonesia 273,000,000 2,273 198
            Pakistan 220,000,000 3,123 45
            Bangladesh 164,000,000 88 9
            Japan 126,000,000 3,139 77
            Vietnam 97,000,000 241 0
            S Korea 51,000,000 10,237 1 183
            Taiwan 24,000,000 363 5

                             3,773,000,000   104,721            3,945

            VS. dat’s BILLION with a B folks !

            Tiny European Country
            United Kingdom 68,000 47,806 4,934

            You’ve nailed it ! As a self loathing European that explanation and many others totally work for me.

          • Gkai says:

            Japan did not, at least they were super criticized at the beginning of the epidemy… Their dealing with the diamond princess was judged abysmal… And now, nothing, not a blib. South korea, Taiwan, Singapore, even China, i can believe that maybe they took drastic measures that dropped R0 below one. But all the south east Asian countries? India? Pakistan? And Africa, that should be in a catastrophic situation since quite a long time now. At first i though it was the hot weather, but now it seems it does not change things much. Or poorly reported numbers, deliberately or not… But after so much time, unhindered exponential with the assumed doubling time and fatality observed on diamond princess should have been seen, even with poor reporting. There is something really strange there, and i have found no satisfactory explanation yet.

            • Pincher Martin says:

              But all the south east Asian countries? India? Pakistan? And Africa, that should be in a catastrophic situation since quite a long time now.

              I suspect age structure is the difference.

              The age structure of Africa and South Asia skews very young. The percentage of the population over 65 in Pakistan is less than 4%. In India, it’s less than 5.5%. In Nigeria, which is Africa’s most populous country, it’s less than 3.5%.

              East Asia is the opposite. The age structure of that region’s countries skew much older. Well over 25% of Japan, for example, is older than 65 years of age. And while Japan is an extreme example, the age structure in the other East Asian countries is closer to Europe than it is to Africa and South Asia.

              • Pincher Martin says:

                An obvious exception to this rule is Iran, which has an age structure closer to Pakistan than to France, but which was also hit hard by the virus. Less than 6% of the Iranian population is 65 and older.

              • Gkai says:

                Iran is the best example indeed. Also South America. Ecuador for example seems to have been hit hard, with 7% above 65. And it’s warm there. Weather and Age can be an element, as is exponential grow and delay for first arrival/slight differences in doubling time that could gives the impression of large differences while it’s just a timing difference… but there should be other factor(s).

      • Paul Conroy says:

        Yep, totally!

    • reiner Tor says:

      Execution was never in the cards, after an hour he was let go with a warning. It was still the clumsy and heavy-handed response of a police state, but let’s not engage in wild exaggerations.

      • Paul Conroy says:

        Reiner Tor,

        You must have happened on this story late. He was set to be executed, there was a huge backlash in China over that and a smaller one in the West. After he died the CCP changed their story and Reluctantly are now calling him a hero.

        Please don’t be that fool, who parrots the CCP propaganda here!

        • reiner Tor says:

          AFAIK the backlash happened in January. The whole thing happened back in December. Could you provide a source?

  3. Steve D says:

    If the original parameters of the model were correct, which predicted five million deaths in America alone, shouldn’t we be seeing a lot more dead bodies right now?

  4. Paul Conroy says:

    Covid19 projections based on “Excess Deaths” around the world:

  5. j says:

    Was the overwhelming reaction of the CCP in Wuhan motivated by a larger number of deaths than recognized? A well organized Communist dictatorship is able to hide massive death rates, it has been done before. Apart from fake-news, I cannot find reliable testimonies that this is the case. Applying Sherlock Holmes’ theory, if all the possible scenarios are excluded, then the impossible must be true. The “impossible” in this case is that the CCP is much more intelligent and capable than all the Western governments. Pls. somebody refute me.

    • NamelessNobody says:

      Can’t refute you directly because in the West we are ruled over by idiots. This goes for the permanent bureaucracy as well as the nominally elected clown show.

      Plenty of dysfunction and epistemological cluster fucks and circle jerks in CCP Land too.

      BUT… when there’s a purely utilitarian job to be done (extinguishing an outbreak of X) then they don’t mess around. If the order goes out from the top that it’s gonna get squashed, it gets squashed flatter than flat and damn the collateral damage. You can be sure a lot of cancer patients, insulin injecting diabetics, dialysis patients, etc. were collateral deaths in the great Wuhan One Size Fits All Lockdown.

      I’d argue that our open society allows them to weaponise our dysfunctions and use them against us. We are so cucked that we cannot even conceive of doing the same to them.

      • j says:

        Thanks, NN. You are right about the collateral. Regarding the last paragraph, in this case their brutality was effective and we too are among the beneficiaries. But yes, they exploited Western dysfunction, to the point that we cannot produce even paper masks and have to buy them from China.

      • reiner Tor says:

        It’s unclear how much collateral damage was done. Apparently they moved a huge portion of medical services online, and what couldn’t, were organized into separate hospitals. The whole thing was organized well enough to visibly impress Bruce Aylward from WHO (admittedly they just might’ve pulled a fast one on him), so is there any data point suggesting that those people were sacrificed? It’s not like without the hard measures taken by China, those people would’ve been better off – I’d guess any patient visiting any doctor was in grave danger of getting infected before the hard response.

  6. mapman says:

    The weak measures taken so far in the US and the pathetic rate of public compliance to them (50% at best) guarantee that our course will be much longer. I don’t know what happened in Wuhan and at this point I don’t particularly care. What I do care about is that our policy makers, at every single point along the way, have assumed and continue to assume the most optimistic scenario. The inability to learn from the past is insane.

  7. Couple of followup comments:

    The uncritical acceptance of CCP data is key because it’s the only data that’s actually flattened out, letting you look at the downturn of the epidemic. Note that the actual # of deaths isn’t the key bit here, since that’s floated as a local parameter for each locale. It’s the shape of the death curve over time that matters. Evidence about CCP undercounting is there as evidence that the CCP is lying about this, not that the undercounting itself is material to the error.
    Parameterizing against the CCP data is likely why they took the barking mad decision to fit with the Gaussian error function. Wuhan data did have an R0 before controls roughly equal to the reciprocal of that after controls. Some moron saw that erf fit the data with the smallest least squares, and they didn’t stop to think about what assumptions that made about the projections in other places.

    I’m furious about this. My wife’s a doc on the backlines right now, doing telemedicine. She got called up this week to volunteer on the frontlines when they need more docs, likely a couple of weeks from now. We got to have fun conversations about making sure our life insurance was paid up and how to set up a quarantine zone in the spare room to keep her away from me and the kids.

    What the fuck is wrong with these people?

    • j says:

      Interesting insight. In a planned economy, somebody may have calculated the curve and set quotas, and the CCP told the locals: “Make it happen!” or “Make it look like it happened!”.

      • NamelessNobody says:

        Some guys in Twitter fitted a curve to the Chinese figures on about day 6 and it tracked perfectly for about a fortnight. Those published numbers came straight out of a spreadsheet model.

        • dearieme says:

          That’s been my assumption all along. There are enough numerate and expert people in China to confect fake data that look real. But as all the old Westerns would have had someone say “Too real, Jeb, too real”.

        • j says:

          Your worthless servant plotted a logistic curve in February and voila! reality complied.

    • gothamette says:

      “The uncritical acceptance of CCP data is key because it’s the only data that’s actually flattened out, letting you look at the downturn of the epidemic. ”

      I’m looking forward to reading John Massey’s response. He says the Chinese numbers are accurate.

      When he does decide to show up again, I want his source for claiming that the first five documented cases in Wuhan showed no connection to the wet market. I’ve asked him six times to source this and he ignores me.

      • mapman says:

        China data, even if inaccurate, are not way off: roughly on the same page as in other Asian countries and other Chinese provinces. Two-fold in deaths would be my wild guess.

    • mapman says:

      No need for a complete flattening. Taiwan, Korea, Japan all slowed it down. Shanghai and Bejing metros did. If the data are available, they are just as useful. Still think that the craziest part is the assumptions that the measures in the US are comparable China’s measures. I am in the same boat, BTW, just sitting here and waiting when it starts. Although I suspect that there is almost no way to avoid infection when sharing a full household with kids and pets.

    • ASR says:

      In the upscale suburbs of Boston, stores are limiting the number of customers getting in at one time and keeping those in line six feet apart. (Six is a magic, voodoo number based on outdated research from the 1930s.) It’s an exercise in futility. Maybe one out of every six persons I’ve seen is wearing a mask of any type that covers both face and nose. Once in the store customers are constantly bumping against each other as they scramble for the goods they want.

      One factor that no model I’ve seen takes into account is that viral load probably varies significantly from one person to another. This may account for the wide variance in symptoms, lethality, and probably infectiousness. In China, Turkey and other countries, back before the 19th century, direct inoculation with pus from a smallpox lesion was the standard method of inducing immunity. Skilled practitioners were required since even a small variation in the pus’s viral load could either confer immunity with limited side effects, or result in a bout of mild small pox that lasted about a week, or cause a full blown case of smallpox that often resulted in death.

      I raise this issue partly because current policies seem to encourage the concentration of Covid-19 patients, especially those with severe illness. The same policies then put first line medical staff in close proximity with these patients, virtually guaranteeing that these doctors, nurses, medical technicians, and others will spend a large part of each day in environments that are saturated with SARS-COV-2 viral particles. I cannot imagine a better policy for spreading serious cases of Covid-19 among our front line medical responders, the very people we will most need when, as appears likely, the Covid-19 epidemic will reappear sometime late this year.

    • saintonge235 says:

      “What the fuck is wrong with these people?”

      They are human beings, that is, apes that evolved speech. But they still have the instincts and reactions of apes: a) Nothing important ever changes. B) The most important thing in life is to get more power in the troop, which gets you better food and more sex partners. C) If an idea or statement of alleged fact gives you an unpleasant emotional reaction, it is false. If it gives you a pleasant emotional reaction, it is true.

      So, believing the Chinese Communists will routinely lie, both because they are Chinese and because they are Communists will be rejected. It causes unpleasant emotional reactions. Believing that a top down culture of extremely centralized power will do good things for the majority causes favorable emotional reactions, and will be believed—they hope to be among the rulers, and even if they aren’t, a society where people make their own choices discomfits them at a deep level (Isaac Asimov’s enthusiasm for totalitarian societies is typical of this). To admit that they are acting like monkeys causes such unpleasant emotional reactions that hardly anyone will believe it, even if they believe that we are evolved from primates (the late Stephen J. Gould’s belief that evolution stripped us of instinct and left us as purely rational calculators is a prime example).

      A pity, but that’s how it is.

  8. rncarpio says:

    For what it’s worth, the IHME model seems to be massively over-predicting hospital and ICU usage in New York. Deaths are on point, so far. In a few days we’ll see if there is indeed a peak, or it just keeps on rising.


    • Balckagroop says:

      Yea I think Greg is wrong on this one.

      Massive underutilization of hospital and ICU capacity relative to what the models say for many states. Look at the models for Alabama and NY right now. Ridiculous.

      • jacquesongoo says:

        Don’t worry, you’ll get there

      • Balckagroop says:

        Projected vs actual:

        This model is very very wrong, but not in the way Greg thinks.

        • gothamette says:

          Berenson is a huge COVID minimizer, although arguably right on certain points.

          One of his quirks (using that word kindly) is to claim that the CDC is overcounting COVID deaths. Anyone who dies w/the virus in their body is a COVID death, says he, and he disagrees with that. Argue with him.

          Now, everyone here is an expert, except me. I’m an ignoramus. So, does anyone die directly of COVID, by Berenson’s definition?

          • DaveNYC says:

            What’s amazing about almost everyone in this comments section is your total commitment to the absolute worst case scenario.
            Are you actually looking at how off those projections were?
            You people don’t want to see good news, even when it’s staring you in the face.
            Hospital admissions have halved in NYC in recent days, and the Navy vessel docked on the West side is essentially empty.
            But go ahead and keep acting like your hair is on fire.

            • gothamette says:

              I’m aware of those data and I’m cautiously optimistic that we’ve turned a corner in NYC.

              Also, not every hospital is overwhelmed. NYC has a huge medical infrastructure that is coping well, with assistance.

            • gothamette says:

              But: epidemics come in waves. I could easily see a situation in NYC where we ease off and get swamped again.

              • glenndc says:

                The only hard data we have is deaths. All else are models. And Box said in his textbook, “…all models are wrong, but some are useful.” In this case, the modelers made testable predictions, number of hospital beds in use, by day, and number of ICU beds in use, also by day. Their forecast is notably wrong, nay, catastrophically wrong… And at the cost of incinerating the economy, and the future of our children and grandchildren. The requisite humility is notably absent.

              • Pincher Martin says:

                The only hard data we have is deaths.

                No, it’s not. As other people here have already pointed out, deaths are being heavily underreported in many hard-hit countries.

                And the economy is not being incinerated.

                In fact, I’m rather tired of people arguing that the economic cost only runs one direction. What about the cost of not taking care of this virus now and instead letting it run through the population over many more months and possibly years?

                There are many economic activities which require the participation of older people (50+) which might never recover. Been to a movie theater recently? How about a cruise ship? Restaurants which aren’t fast food joints? People who are older than fifty are a large segment of those markets.

                Nursing home residents don’t spend lots of money, but people older than fifty who are still a long ways from the nursing home spend a great deal of money. And this group is not going to return to their old spending ways if the virus is still out there.

                So good luck restarting the economy based on nothing more than a politician’s order with COVID-19 hanging around.

    • Didn’t talk much about their hospitalization model. It’s not crazy, though with sketchy parameters, just like everyone else’s. But it’s dependent on their death curves, which are crazy. Basically, they’re taking the death curves they come up with and trying to walk backwards: how many critical patients would you have to have to produce that many deaths.

    • Just saying says:

      If the theory that viral load is related to disease severity is correct

      And the hospitals are treating people in a way which significantly increases viral load

      Then we might see a high death rate for medical workers and those who are hospitalized

      And a lower than expected hospitalization rate and death rate for those who get a mild case and recover at home

      (Of course, some people get hit hard by even a low viral load case, so the model is more complicated than that.)

    • rncarpio says:

      IHME has updated their resource usage model. More realistic for NY, but still predicts a lot more usage than is observed.

      Notable is that Cuomo no longer talks about a possible shortage of hospital beds. Also, the IHME model now says that WA state is past its death peak, which was 4 days ago.

      • They say they’ll publish the details of the changes tomorrow; will take a look at it when it’s available.

      • rncarpio says:

      • kpkinsunnyphiladelphia says:

        The IHME model so far seems pessimistic. It turns our as of today, April 7 in New York, IHME predicted a range of hospital beds required at 25,482, with a range between approximalely 14,000 to 40,000. On Cuomo’s slide today, the number of hospitalizations is 17,493 — off by about a third from the projection.

        Or maybe the IHME error band is so large as to be useless

        It’s frustrating that the IHME does not overlay actual numbers against their projections across all geographies. I don’t know of any such comparison that exists out there, and I am certainly not going to through the tedious process of doing so. for all their numbers. But somebody ought to.

  9. Scott Novak says:

    Maybe we could crowd fund a few people that aren’t liars and/or morons to make their own model – You know, like that book review you did. Maybe involving Quillette or something.

  10. Joel says:

    We are still seeing the “we tried” response right now from American politicians. In 3-4 weeks, we can expect the “oh shit, too late now” response.

  11. Pingback: IHME projections — West Hunter – Nicht-Linke Blogs

  12. Martin says:

    Greg, have you considered that COVID may be an anti-DNA agent? It kind of has that MO with its well-established “false recovery”: an initial onslaught of cells killed outright, then the real blow later when damaged cells can’t replace themselves. Compare radiation damage, amanita poisoning.

    If that hypothesis is true wouldn’t COVID survivors face a cancer threat?

  13. H I says:

    Greg, what do you think of the possibility of a cheap antigen test? Immunoassay, not PCR. When?

  14. Steve D says:

    The claim was made that it was silly to think this virus would become less virulent. A comparison to smallpox was made. But smallpox is a large complex virus that mutates relatively slowly compared to flu and cold viruses. That’s one reason the vaccine for smallpox works so much better than a flu vaccine.

    Routine measures, perhaps stepped up a bit, might very well have accomplished the slow down we see now, without having to ruin millions of lives.

    I think a lot of people sensed this, but the catastrophe cheerleaders were using scientific notation and dead Latin phrases to make it sound like they were really smart.

  15. gothamette says:

    Australia’s done a bang up job – about 40 deaths to date.

    Any ideas as to why?

    • saintonge235 says:

      The population of Australia is one thirteeth that of the U.S., and four fifths the area. I think that population density of 9% of the U.S. may have a great deal to do with it.

  16. Anon says:

    Deep dive by Nate Silver – helpful if we want to know why we don’t know anything.


  17. pyrrhus says:

    One of my kids lived in China and still knows people there…they say the skies were grey in Wuhan due to crematoria running 24 hours/day, and everyone assumes that the deaths were 50 to 100 times what was reported.

  18. Anon says:

    Deaths to April 10

                            Population           Total Cases        Total Deaths      % Pop = dead

    All Europe 747,536,619 820,109 70,069 0.009373
    United States 330,586,808 502,146 18,747 0.005671
    Canada 37,669,306 22,148 569 0.001511
    Australia 25,434,536 6,238 54 0.000212
    New Zealand 4,813,617 1,283 2 0.000042

    Totals 1,146,040,886 1,351,924 89,441 0.007804
    All Asia 4,632,325,100 275,041 10,224 0.000221
    All Africa 1,333,397,000 13,539 697 0.000052

    Totals 5,965,722,100 288,580 10,921 0.000183

    First world dying 40x faster 78 per milion vs 1.8 per million

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s