Okay so the pandemic was predictable and it was predicted. I'm not interested in semantics quibbling if you want to say it wasn't predictable, I'm interested in the contradiction between claims that the pandemic was not predictable and yet other similarly hard to predict things are predictable.
> (I’m a fan, but I don’t think he did a ton of proofreading on this.)
I suspect you (and parent) may have misunderstood him. IMO, he meant a pandemic (like COVID-19) - just like WWII - was predicted by a lot of people. What people could not predict (for both scenarios) was when it would happen.
>but lots of people were predicting a pandemic, even covid-like if I remember correctly.
If you look at [1] the January 2020 - predictions for 2030 thread here on HN, and look for hints of the Covid pandemic, there are only 3 mentions of any type of pandemic, and none of those was close to an actual hit. I did the reading, all of the reading.
Meh, I don't think pandemic is a binary term. If the perpetual flu pandemic is the baseline, for the forwardgoing 1 year I don't think COVID is going to be, e.g. 10x worse in deaths in vaccinated countries.
I don't understand that comment. I keep up to date on the latest scientific papers on the coronavirus pandemic, and the one thing I wouldn't call it is obvious. We repeatedly see things happening that run contrary to what the consensus expectation was. Obviously some of this is due to improvements in our understanding over time, but we are still very much in the dark about how it is currently spreading, who is most likely to suffer fatal or severe consequences, how we best treat the disease, etc.
Charitably, it is neither as bad as we feared nor as innocuous as some hoped, but that was bound to be the case and easily predictable. I don't feel at all confident predicting what comes next, however.
Agreed (though in the US we do have a yearly nation-wide vaccination program, imperfect as it is, and largely staff and equip hospitals to handle the high-water mark of cases).
> I think making predictions on how many people "could have died" when we're doing extremely limited testing isn't helpful.
I disagree with that. The alternative to making imperfect predictions (predictions are imperfect by nature) is to not make predictions. Without predictions you can't make reasonable decisions to direct your future.
Of course, when your predictions are too far off you can't make good decisions to direct your future either, but that just suggests you should try to make the best predictions you can (hence the science and study of epidemiology), not that you shouldn't make predictions.
If you don't make predictions (or you make them based on hunches or by other low quality means) you're just throwing yourself randomly into the future and are unlikely to get good outcomes.
> But conditional on being in a pandemic the chance that the virus that arises to cause it is like SARS-CoV-2 is super high!!
This will be misinterpreted by everyone who doesn't understand the Bayes rule (which is 99.999% of the population). What she's saying is just a fancy version of "this is the kind of a virus that would likely cause a pandemic", nothing more.
> At the very least, it was a half truth, and unrealistic hyping.
It was what the data showed at that time. Calling if half-truth because people couldn't predict course of a pandemic year in advance seems over the top. Do you know what will happen in a year?
No idea who is your president, so I can't comment on that specific communication.
> because the true severity of the outbreak isn’t yet known, it’s impossible to predict what the impact of that spread would be
This seems to be a key quote.
When you're making a bet, you should know the odds. Influenza is a known thing. I'm not going to think about that facemask knowing that influenza is going around, etc.
I am not playing games. This stuff is far too serious to play games over.
> To extrapolate over the general population with the 1.4% case fatality of serious hospital cases, would be wrong and would be fear mongering. The general population does not have a infection rate of 100% nor a case fatality rate of 1.4%.
No, but the Spanish Flu for instance had an infection rate of about 25%, and the general infected population could very well be above that 1.4% (in fact, there is every indication so far that it will end up somewhere between 2 and 3%).
> So the premise is wrong, the assumptions are wrong, and the result are wrong.
The premises isn't nearly as faulty as you make it out to be, it might be off by a factor of four, hopefully much less than that. But so far - taking Wuhan as our baseline - unchecked this thing will spread very fast and kill lots of people. The only reason the onslaught was halted is because the Chinese took a bunch of draconian counter measures, the likes of which very few countries in the world will be able to copy.
> Qualifying a statement ending with ”means 2 million deaths in that single bracket alone.” with a qualifier that is false is what I would call fearmongering.
That's fine by me. But it isn't. As a whole it isn't and in the details it isn't either. We actually have precedent for this whole situation, it is the 1918 pandemic and the world apparently hasn't learned its lesson fully. Which may mean we are about to re-learn it the hard way. Maybe this time it will stick. But then - as now - there were plenty of people like you who were willfully - given the way this conversation is progressing I am assuming you are doing this on purpose - ignoring the cold hard facts. And that made the pandemic much worse than it had to be.
Large events were not called off and led to the disease spreading much further than it had to.
Absent an effective anti-dote to this virus and with an overwhelmed health care system (note the qualifiers!) we might as well be in 1918.
> No, it’s not an “implied assumption” it’s faulty epidemiology.
Not as far as I can see. People talk about hypotheticals all the time. You can't really come out like this against a pretty clear comment and at the same time make up stuff you rail against that never happened in the first place. Fix your own issues first.
> First, it is faulty to assume that everyone who is exposed will be infected.
Agreed, so let's divide the number by 4 and call it a day. That's the worst this could possibly get. Right? Or do you think that given a sample of '1' the 1918 flu might not be the best guideline for precision and that it might be better or worse? We don't actually know the answer to that.
> Second, it is faulty to apply the 1.4% case fatality rate of serious hospital infections to the entire US population.
That's true. But for all the same money we end up with a higher case fatality rate. So that may end up affecting the equation in a negative way.
But assuming 10% of the country becomes infected and 5% of those end up with symptoms serious enough to warrant ICU care there is a serious problem.
And that 10% could very well be low. So as far as I'm concerned you should start taking this a bit more serious instead of trying to shout down the discussion. Stay dispassionate and try to argue your case with actual numbers rather than nay-saying. That way you can contribute rather than just irritate. Think of it as an opportunity to educate, and you just might learn a thing or two yourself.
>>The fact that the pandemic is effectively done in Sweden is one.
Acting like this is a foregone conclusion is the height of arrogance. Diseases spread in waves. They could just be at the low point between two waves.
We won't know until years from now. Just asserting you're right when there are still this many unknown factors smacks of someone looking for data after coming to a conclusion.
>If you replace a big party with 100 people, with ten smaller parties with ten people, you probably end with a similar amount of infections
I don't think that is a correct conclusion. It doesn't behave linearly.
"We can think of disease patterns as leaning deterministic or stochastic: In the former, an outbreak’s distribution is more linear and predictable; in the latter, randomness plays a much larger role and predictions are hard, if not impossible, to make. In deterministic trajectories, we expect what happened yesterday to give us a good sense of what to expect tomorrow. Stochastic phenomena, however, don’t operate like that—the same inputs don’t always produce the same outputs, and things can tip over quickly from one state to the other."
> Currently my model is predicting a peak of around 140,000 average weekly new cases right around Thanksgiving with a steep decline due to herd immunity thereafter. It is also predicting that the Pandemic will be pretty much over by the end of February.
No one else is predicting anything like this. Makes the whole article a little suspect.
> but I think it's pretty likely they'll try and cast doubt on the results if it goes against them.
That is certainly true whether or not there is a pandemic. I don't think that inevitability really has anything to do with pandemics; they'd blame it on the phase of the moon if nothing else came to mind.
> For example this report from March 2020 was highly influential and predicted 2.2 million deaths in the US in an 'unmitigated epidemic' scenario:
You just said yourself that the 2.2 million was the "do nothing" scenario, and as we've seen in India it could've easily been that. You also just said said we did many things — "unusual and unjustified" things, in your opinion — therefore mitigating that worst-case projection. I personally lost the point you were trying in these self-contradictions, but I am curious where you were going.
There certainly was a pandemic, the 2009 H1N1 one. [1] In the last few months I've started to suspect that that was what killed my grandma in July 2010, she died very suddenly because of some respiratory issues (I live in Eastern Europe).
Okay so the pandemic was predictable and it was predicted. I'm not interested in semantics quibbling if you want to say it wasn't predictable, I'm interested in the contradiction between claims that the pandemic was not predictable and yet other similarly hard to predict things are predictable.
reply