Did doom visit the Swedes yesterday as planned?

On May 3, in “Doom for the wicked Swedes is always three weeks away”, the IHME prophecy for Sweden was a peak in ICU usage on May 22 and a peak in deaths (494/day) on May 23.

What actually happened? Yesterday’s WHO report showed 54 new deaths. The day before it was 40. In other words, the prophecy was off by a factor of 10.

They were going to need nearly 4,400 ICU beds. The actual number in ICUs all around Sweden? About 340. In other words, the “scientists” were off by a factor of 13X.

What’s the latest from the prophets at IHME? As of a May 20 update to the forecast, Sweden will have a gradually declining daily number of deaths, in more or less the same shape as still-shut-down Massachusetts. A total of 5,129 Swedes will die from/with Covid-19 (roughly one third the previous forecast). The virus will simply burn itself out, apparently, despite Sweden’s lack of shutdown. (But in other countries, the same shape decline will be attributed to a multi-month shutdown?)

How about the Fall of Saigon-type scenes at hospitals that were forecast? IHME has walked that back to forecasting about 500 ICU beds occupied as of today (compare to 340 actual). Within a month it will be down to 150.

Do the arcs of the epidemic in different countries confirm the Swedish epidemiologists’ theory that Western government policies have minimal effect? Here are a few:



22 thoughts on “Did doom visit the Swedes yesterday as planned?

  1. > But in other countries, the same shape decline will be attributed to a multi-month shutdown?

    It can hardly be otherwise now in MA. We don’t have all the contact tracers in place yet! 61% of the deaths here have been in nursing homes and assisted living facilities, not hospitals. Only in May did the Baker administration get around to assessing how they were protecting the elderly and vulnerable. Their “baseline audit” covered May 4th – May 15th, after all the numbers had started trending down. We have more open ICU beds than in-use throughout the state.


    Phase II of reopening doesn’t begin until July, and if anything starts trending back up, we will all roll back to Phase Zero. Pages 12 and 14 (deaths by age / preexisting conditions) of the MA COVID-19 dashboard are virtually unchanged for over a month now. https://www.mass.gov/doc/covid-19-dashboard-may-23-2020/download

    Boston’s mayor Marty Walsh says this will go on for another eight months to a year! There’s tremendous solidarity among governors and mayors about the strategy.

    I have a family member who is a dentist in big city in PA. He was screaming at me on the phone today that Trump is a lunatic and the whole country should be locked down in rigorously enforced quarantine for another six months! He said: “Trump should be in prison for crimes against humanity!” He thinks that even the essential businesses lists are irresponsibly loose!

  2. You’re failing to account for a feedback loop. The initial predictions may have been perfectly accurate if not for their existence, or the existence of any reporting of statistics on the pandemic. In fact, isn’t that the most legitimate and valuable reason to do predictions–to alter behavior? Whether health, weather, earthquake, tsunami, economics? You presume the scientists were very wrong, and they may have been exceptionally right, though you’re grading them superficially while they’re probably more interested in outcomes. Reasonable people likely adjust their behaviors in regards to COVID accordingly based on their own assessments given all the various inputs available.

    • @Senorpablo: That’s correct, it’s a social control feedback loop. And using COVID-19 as a pretext, it can be applied to anything. It can be done with global warming. Isn’t that the most legitimate and valuable reason to do predictions – to alter behavior? The level of overt force can be adjusted as necessary to achieve the desired outcome. Hong Kong has done splendidly with their COVID-19 response thus far, but Hong Kongers must not break the social distancing rules as they protest against the security laws:


      “Police fired tear gas at the crowds less than an hour after the start of the march, which did not receive official authorization and went against coronavirus social distancing restrictions, which ban groups of more than eight people meeting.”

      Here in the United States, Ohio is using undercover police this weekend. People there are just going out to enjoy the holiday, but it’s the same principle, minus the arrests and tear gas:

      “Along with looking for criminal activity, the officers now are tasked with watching for potential violations of the state’s health order for restaurants and bars to safely operate during the coronavirus pandemic.”


    • “scientists … they’re probably more interested in outcomes.”

      What a beautiful example of commie thinking ! Means justify the end, very Gruberish. One would not call those modellers scientists though, politicians perhaps for lack of a better word. Besides, in the case of Sweden they did not seem to have managed to alter people behavior much by their attempts at “science”.

    • Senorpablo: That’s a great theory, but it is easily tested. https://www.google.com/covid19/mobility/ shows that the Swedes, despite the doom forecast for them by IHME, did not adjust their behavior via additional cowering in May (nor did government policy change). The biggest difference in Sweden for May seems to be that they were out in their parks more.

    • Philg- I haven’t heard that parks is where they keep the COVID, nor that being mobile will give it to you.

  3. Don’t use the log scale as it is very deceptive. Use the linear scale of deaths per million and you will see that the some lockdown countries have far more deaths per million. https://imgur.com/o2rxCjs

    • I like the log graphs because they make it easy to see that every country’s curve is basically the same shape, and in every country represented on the map the trend towards zero cases is obvious.

  4. > But in other countries, the same shape decline will be attributed to a multi-month shutdown?

    Beth Cameron from Obama Pandemic thinks hard:


    “Asked if Americans need to be prepared for ANOTHER THREE YEARS of social distancing, Cameron said she believes it could be longer than a year and emphasized how difficult it is to make predictions when WE KNOW SO LITTLE about the virus.”

    • Wuhan was locked down for 74 days and seems to have gotten their coronavirus cases down to quite a low number as a result, albeit not quite zero. Here in the US, Governor Whitmer just extended Michigan’s “lockdown” to 80 days, which she believes to be necessary because of the ineffectiveness of American-style lockdown. Too bad we can’t work together and follow the rules as efficiently as Chinese people. I think the political will to avoid seeing our friends and family for extended periods of time is now gone, so we’re going to have to let hundreds upon hundreds of thousands of people die as a result.

      I’m pissed that I followed the rules for two months in New Jersey, and as far as I can tell didn’t catch coronavirus or give it to anyone else, but I’m surrounded by peers who ignored the rules and contracted it along with a bunch of their friends. Since the New Jersey and New York state governments are less competent than those of places like China, South Korea, Taiwan, and Singapore, we still don’t have any contact tracing so none of these yahoos have to bear any responsibility for any of the deaths they may have caused. And for those who are most worked up about the “lockdowns” and economic devastation, of course the lockdowns have to continue longer as a result, and the economic devastation continues longer (in part because businesses are still closed and in part because people are still scared to visit them).

  5. I think the predictions for hospitalization rates and the length of time it will take for the pandemic to sweep through the population have been off way (for the latter, possibly in part due to mitigations) to a way larger extent than those for the infection fatality rate. The available evidence suggests we’re still seeing roughly the amount of death and destruction expected. E.g. Fauci said on March 31 that 100,000 to 240,000 Americans could die, and the University of Washington’s Institute for Health Metrics and Evaluation predicted 84,000 deaths by the beginning of August; we know for a fact that the IHME estimate ended up being a lowball since we’ve already hit that number in May, and it’s unclear at this point whether or not we’ll fall into Fauci’s range (I think it’s not unlikely we’ll blow past 240,000).


    • Pressed by a reader, in a comment on https://philip.greenspun.com/blog/2020/03/26/why-do-we-care-about-covid-19-deaths-more-than-driving-related-deaths/ , I predicted (March 27) 150,000 U.S. deaths tagged to Covid-19:

      Hooded: why can’t we extrapolate from Lombardy to get a reasonable estimate? Out of 10 million people, COVID-19 has killed 5,000 to date. Assume that becomes 10,000 by the end of the year. That’s 1 in 1,000 people. Applying that to the U.S. we get 330,000. Horrific, of course, but about the same as the driving deaths expected for next 10 years (which didn’t seriously concern the nation). I don’t think this can be off by 100X. 33 million is too much. 3300 is, sadly, likely to be exceeded (1,301 as of right now). Maybe we can do 4X better than Italy due to advances in knowledge and drug therapies that are available. That brings us down to 82,500 deaths, not too different from what the Imperial College folks are predicting per capita for the UK. We’re not as competent as other countries when it comes to health care, so estimate 150,000 COVID-caused deaths through February 2021? Let’s come back to this post on March 1, 2021 and see if the mortality was, in fact, within 0.25X to 4X of 150,000.

    • The 150,000 total mortality seems likely if we assume that no more than 5% of the US population will be infected by the next year February given the 1% IFR estimate that does appear reasonable based on seroprevalence sampling (Spain 1.1%, France 0.7%, Sweden 0.6%, New York State 0.8%, Boston 0.8%). The numbers are averages, not age adjusted.

      The infection rate is a big unknown of course.

    • I assume you mean Table 1? A symptomatic case fatality ratio of 0.004 times 65% infections being symptomatic = 0.26%? That’s ludicrously low. 21,216 confirmed and probable coronavirus deaths have already been reported in New York City and after 420,000 residents fled the city [2], about 8 million remain, which already gives you 0.2652% of the city dead right there.

      We don’t have any reason to believe either that coronavirus stops spreading once it hits 5% of the population, nor that it’s close to 100% already (the most recent antibody studies in NYC I’m aware of were about ~20% positive). If the entire country eventually gets hit as hard as NYC, that’s about 870,000 deaths right there.

      [1] https://www1.nyc.gov/site/doh/covid/covid-19-data.page
      [2] https://www.nytimes.com/interactive/2020/05/15/upshot/who-left-new-york-coronavirus.html

    • Ryan,

      “If the entire country eventually gets hit as hard as NYC, that’s about 870,000 deaths right there”

      Let’s hope not. NYC, Boston, Spain do look like a bit of an outlier. Or maybe not.

    • @philg
      >Applying that to the U.S. we get 330,000. Horrific, of course, but about the same as the driving deaths expected for next 10 years (which didn’t seriously concern the nation).
      Ugh. The difference is that the deaths due to COVID-19 were 0 in 2018 in the USA. Probably 0 in 2019 in the USA. Now we blithely regard 150000 to 330000 as “Meh, its about (pick your existing statistic) as many as 10 years of traffic deaths.” So it is _incremental_ deaths that get our attention, something of which you are no doubt aware.

      Is there a large component of overreaction and “virus avoidance theater”, of course. That is unfortunate as it provides cover for those flouting the common sense rules: work from home if you can, avoid crowds, wear masks (for OTHERS, not you) when in public.

    • The Coronafaithful here are unhappy with the CDC’s 0.26% IFR estimate, as are they elsewhere. Sometimes, it turns out the science isn’t settled!

      It’s true there are grounds for objecting to the CDC’s number. It’s based on an asymptomatic rate of 35%; that rate is more likely 81%.

  6. We’ll get the Re < 1 and the numbers will keep declining so that this becomes a "long slow burn" with pockets of new infections cropping up from time to time, combated with invasive surveillance, contact tracing, rolling shutdowns and forced quarantine before a vaccine emerges. If we cannot keep the hotspots under control, the people we know are the most vulnerable will continue to die despite our best efforts to protect them. We will do what we can, it will do what it does, and in the end the Lord must surely judge us all.

    I'll state for the record here, separately, that Trump has completely lost whatever vestiges of trust in his judgment and character I might have reserved. He has crossed the Rubicon by putting the Klausutis family through hell again for absolutely no reason other than the fact that he's a sociopath. First the left put them through hell (Truthout, Daily Kos, Michael Moore, et. al.) but I expected that from them. Now Trump has been kicking them again. He won't let it go, and he deserves absolutely no mercy for that.
    He's a disgrace to our country.

  7. “He’s a disgrace to our country.” , Alex

    My nomination for understatement of the year!

  8. @toD: I’m being consistent. I thought it was disgraceful when the above-named Left did it. I really didn’t expect him to keep this up. It’s the family’s business to do that if they wanted to, and they would have a lot of support in that instance. But they haven’t. It’s not Trump’s business, POTUS or not.

Comments are closed.