https://www.latimes.com/science/story/2020-03-22/coronavirus... He seems to be a perfect example of the axiom that deep knowledge in one field doesn't necessarily give you expertise in others. Seems like he should leave Epidemiology to the epidemiologists.

Since when is that an axiom? He carefully analyzed the data and presented a falsifiable prediction along with his methodology. If he was wrong we should look at why and learn from it. (Did he fail to account for the virus mutating, could different regions have different co-morbidities or immunities?)

We should not rely on appeals to authority, those expert epidemiologists didn't do much better.

He's been talking about this a lot - the (lack of) communication and debate in the science community.

People interested in truth invite criticism: "Please give me contrary evidence! Show me I'm wrong!"

"Show me I'm wrong" is not the starting point in the scientific method.

It kind of is. Hypothesis rejection is the primary means of advancing knowledge.

"We tried to prove this hypothesis wrong and could not" is the main thing you want a study to do. Having everyone try to prove your hypothesis wrong, and failing, is the main way that science advances.

No, really, it is not. In your own example you're already jumping to testing a hypothesis. "Scientific method" isn't just a vague phrase, there literally is a specific method of steps to go through.

The first step is Observation. Sometimes split out into Observation, then Research.

Then comes the formulation of a hypothesis.

After that is constructing a way to test the hypothesis. Not to prove it wrong, not to prove it right, but to test it.

There's more after that, it's an iterative process and a single test of a hypothesis isn't always definitive, often it is not. But "Prove me wrong" is not the starting point.

No, science proceeds on the basis of trying to prove a hypothesis is correct. As in, I hypothesize A should B, evidence XYZ supports this.

You accomplish very little trying to prove a hypothesis wrong, because most are.

While you are correct conceptually, most scientific research using statistics does in fact attempt to provide evidence against no-effect (the null hypothesis).

The p-value of a scientific study is the probability that the given data would have been observed, given that there is no effect. Hence why small p-values can be associated with the success of the alternative hypothesis (i.e. what a scientist actually thinks will happen).

We should remember Popper (1).

You can't prove an hypothesis is correct. All models are temporal until they are disprove.

(1) - https://en.wikipedia.org/wiki/Falsifiability

I know some people who tap a can of Coke before they open it, because "that prevents it from spilling". Each time they do that, they are more convinced that their hypothesis is correct.

But is it correct? To find out, we need to try to invalidate it. Try opening the can without tapping it first. If the Coke doesn't spill, the hypothesis is clearly wrong. It the Coke does spill now, it's quite strong evidence in favor of the hypothesis.

Lol no ! You prove a hypothesis to be correct.

It's something that's felt by scientists who've developed a hypothesis they want verified. Keen interest in explanations of data, etc.

> those expert epidemiologists didn't do much better.

How did you arrive at this assessment? I've been following this since the end of January, and every expert I've heard gave sound and very good advise based on the available evidence at that time. Some of them might have made some disputable trade-offs, for example the Swedish lead epidemiologist, but overall there was a lot of agreement and the recommendations were excellent.

Okay, maybe there are a few outlier countries like the US and Brazil where there might have been political muddling of the expert messages. Be that as it may, in general the advice from experts was not only good, most countries also succeeded in controlling the spread of the virus based on it. (And not every country needs to implement the same measures to be successful in that, it's more about a mix that works in that country.)

Those experts at the WHO predicted, in March, that if Sweden does not enforce a lockdown as recommended by the WHO, it will have 96,000 deaths on 1-July.

The non-expert Michael Levitt, predicted at the same time, by fitting 3 parameters of a Gompertz distribution (And no other knowledge), that Sweden will have 5000 deaths on 1-July if things continue as they were when he made the prediction (that is, no WHO recommended lockdown).

I'll leave it to you to lookup the Swedish death count on 1-July to see if the expert epidemiologists from the WHO had a better idea than non-expert Levitt. (But I'll give you a hint: It is mid september, and Sweden has ~6000 deaths).

Nonsense, and also irrelevant. You're picking out one predictive model that got it wrong. I'd like to see this model, by the way. I've seen numerous models that got it right, and in any case I was talking about the advice experts gave. Nobody doubts that Sweden's death toll would be much lower if they had done a lockdown like almost every other country. In fact, the Swedish lead epidemiologist admitted that mistakes were made and that they failed to protect the elderly.

The WHO advice was much better than what Swedish advisors came up with.

Besides that, I was not talking about WHO experts but about experts in general. Every country has them, and they supply the data to WHO.

Last but not least, you cannot evaluate a model on the basis of a singular prediction.

bingo. Maybe why actors should stick to acting ?

It is encouraging that real scientists (not just "pro-science" laymen) seem to be listening to these ideas and producing studies on them while the media ignores and ridicules them.