Authorities and Amateurs

I have been stewing about this question in general for a while. When I look back at my long (so far) life, I think of the many times I have been misled by so-called experts.

From which I learned that experts, even real ones:

• Are subject to massive cognitive biases, without realizing it. One common one is the filtering of data based on prior beliefs, not updating when new evidence comes along. Science advances funeral by funeral.
• Are often influenced by mercenary motives, and are frequently oblivious to it.
• Often defer to out of date, wrong or incompetent but powerful figures. Another reason why science advances funeral by funeral. Medicine is particularly prey to this problem, due to the strict hierarchy in medical organisations.
• Frequently optimize something other than truth. Publication, career advancement, money (as mentioned above), status, etc. Ask "What is the success metric?".

On top of that a lot of self-proclaimed and even highly credentialed 'experts' don't actually have much of a clue. Because:

• Fields often have huge blind spots. E.g. I frequently see studies of the influence of childhood poverty or education or SES etc on people's lives, in which the model explicitly assumes that there is no influence from parent to child via genes. Knowledge of statistics and of mathematics, key tools for understanding pandemics, are particularly weak in many fields. In my country it is typical of doctors to have mathematics only up to year 11.
• People are experts in a far more narrow domain than they realize. My own country has a Chief Medical Officer, who seems to have little grasp of the management of epidemics. Long ago he was a medical specialist in a largely irrelevant field, for several decades a bureaucrat/political player. Dunning Kruger Syndrome.
• Training often induces in people a hefty dose of arrogance - medical training being a particularly unfortunate example; law is another - and this arrogance is transferred to areas beyond the person's sphere of competence.
• Fields of study are often set up with safeguards and barriers which may or may not be well intentioned, but which prevent outsiders with good ideas from having any influence. In endocrinology, a field that impacts me personally, the practitioners in my country appear to have lost large swathes of knowledge and nothing can be done about it (e.g. of how to understand complex systems, so that endocrinological problems are typically assessed in terms of "is this individual blood level 'normal'" rather than looking at the system as a whole).
• Ideas, beliefs and practices that were formed based on little or no evidence become entrenched and remain in place, while anyone trying to overturn them is held to extremely high standards of rigor. Have a look at the evidence behind the original recommendations to avoid saturated fats, and to eat "healthy" trans fat laden margarines for example.

Important **None of this is to say that an amateur with google and ten minutes to spare can do better**. be cautious. It is very hard to do better than flawed experts.

Personally I have worked out, over time, some heuristics which have proven useful to sort out actual experts. Some things that mark out an actual expert:

1. They can make surprisingly accurate predictions. Better than most people, and better than simple techniques like linear extrapolation.

2. They can fix things that are broken. Whether broken machines, or dysfunctional social systems, or sick people.

3. They can explain things in a way that is as simple as possible, illuminating, and gives one clues as to how things might be better.

Not only that, but they have evidence for this. An example of the opposite: After thirty-five years of Freudian psychoanalysis, someone thought to do a study of whether they actually helped people get better more than doing nothing. No, they did not.

Things that do not mark out an expert:

1. Status among peers. The peers may be equally clueless or useless.

2. Great confidence. This is more a sign of arrogance than of competence. In "A Mind for Numbers" it is pointed out that claims of skill or competence or knowledge not accompanied by proof are actually far worse than acknowledged incompetence.

3. Ornate certificates on the wall.

4. Having attended high status institutions.

5. Having been successful, after taking huge risks. They may be a lucky idiot - look closely.

6. A few lucky breaks.

In the current context, I am willing to listen to experts who have a proven track record, who have relevant experience, and who have the skills needed to do the job. Even then I look hard for biases.

I welcome any additions/corrections/clarifications to all this.

It's not hard to find amateurs who got it right, after the fact. Amateur advice is of lower expected quality, but of much higher variance than expert advice.

Apart from filtering out the obvious crazies, can we identify high quality amateur advice ahead of time?

Neil Ferguson and Ian Lipkin, heads of their respective prominent epidemiological institutes, have caught or are suspected of having caught COVID-19. Being a top epidemiologist seems to be a strong risk factor for being infected, given that fewer than 1% of the English speaking world has probably been infected so far and there are not that many epidemiologists who are as prominent as these two. (This seems worth noting here as a surprising/interesting fact to update on. I'm not suggesting we should draw strong conclusions from this.)

ETA: See also my previous comment [LW(p) · GW(p)] about a couple of authorities on economics.

Notalgebraist posted a reasonable critique of Flattening The Curve Is a Deadly Delusion, explaining how it's incorrect to assume that flattenings won't reduce the total number of cases, and that it doesn't make sense to assume a normal distribution. I think that the the piece's main point was correct, though: that the curve would need to get crazy flat for hospitals to not be overloaded.

3 days later, the Imperial College study made the same point in a much better and more rigorous way (among lots of other good points), but it was less widely shared on social media.

I'm not sure what the conclusion here is. Non-experts will sometimes make false assumptions and get the details wrong, but they're still capable of making good points that only require you to multiply numbers together, and will do so a few days faster and in a way that's more memetically fit than papers from experts?

Since the very beginning of LW, there has been a common theme that you can't always defer to experts, or that the experts aren't always competent, or that experts on a topic don't always exist, or that you sometimes have to do your own reasoning to determine who the experts are, etc. (E.g. LW on Hero Licensing [LW · GW], or on the Correct Contrarian Cluster [LW · GW], or on Inadequate Equilibria [? · GW]; or ACX in Movie Review: Don't Look Up.)

I don't think this post makes a particularly unique contribution to that larger theme, but I did appreciate its timing, and how it made and referenced its examples.

Hmmmm. I am wondering about the information transmission/propagation process and how that might effect outcomes.

The "experts" are know to be connected with some subject area. When people (particularly media and government) the "experts" are brought in. LW has had at least one thread on this situation and what some of the problems might be.

The "armchair" people are probably two forms -- either they are some form of known "celebrity" type of smart person or they are unknowns to the world generally. In either case the utterances from these people run through a different filter before their claims become part of the general information about X.

In this particular context -- a pandemic, we don't have too many real experts in the sense of "I've done this before and I have see this playing out". Yes, we've seen outbreak and understand the transmission processes and models pretty well. However, some of this seems to be different than SARS, MERS and similar more contained or localized events.

As someone mentioned, we can find a bunch or pretty silly analysis or recommendation from the armchair side (even that it's a hoax, just like a cold/flu or only a problem for the really old and already sick not the healthy). We when we make the claim about how the armchair crowd has done much better than the experts I think we gloss over how those good armchair positions came into the general information set. They were the cream of the crop and filters via a number of social filtering mechanisms.

We should not compare the best armchair position against some average expert position (where the experts may in fact not really be experts).

This might however suggest the selection mechanism used by both media and government in situations where we are dealing with something somewhat new may have some weaknesses that we want to review.

On this particular topic, some actual (official) experts are starting to come forward with good material.

I like this one, with the new insight that quarantine in isolation was very important. Beyond "please stay home".

Presentation

https://zoom.us/rec/play/v8Ytceqqqzs3GNzB4gSDB_59W9TsK6Ks13RI_6cLxB62BSUAOlumZeRAZLC7e1vif7xIyy6HL_uXyNHw?startTime=1584118874000&fbclid=IwAR3VvAr7vZg2M0ZHEHYeGFsRrC0RvL3vwtCcUTmnZTccWW644x-nJt7FxyI

Paper 

https://www.medrxiv.org/content/10.1101/2020.03.03.20030593v1

Slides
https://drive.google.com/file/d/1QFfBbs-7-qCqDJ93_Ww6QQUQlMVp77Pl/view

Here is an expert trying to reverse engineer the government's strategy. [I do have some concerns with his approach, he seems to have an exaggerated view of how accurately you can manage R0. There is such a narrow zone for R0 between "it blows up fast and high" and "it rapidly declines close to zero", that IMHO it is futile to try to finesse how much slackness we can get away with.]

https://theconversation.com/coronavirus-modelling-shows-the-government-is-getting-the-balance-right-if-our-aim-is-to-flatten-the-curve-134040

The specific evidence you’ve cited is weak. (1) You write that “The argument that we should be listening to experts and not random people would make a lot of sense if the "armchair" folks didn't keep being right.” It is extremely easy to be right on a binary question (react more vs less). That many non-experts were right is therefore more-or-less meaningless. (I can also cite many, many examples of non-experts being wrong. I think what we want is the fraction of experts vs non-experts who were right, but that seems both vague and unobtainable.)

(Note that this is importantly different, and stronger, than the claim you made in the final paragraph. I agree with that claim.)

(2) For many, but probably not all, of the policy failures you describe, there is little reason to attribute them to experts. The United States is not a technocracy.