zartsnarf

I. The Observation

The mid curve is something people have been noticing for a while, usually passed around in meme form. Competence does not rise in a straight line as you learn more about a subject. It dips in the middle. Someone who knows nothing about it often behaves sensibly, someone who knows a great deal also behaves sensibly, and the person in between behaves worse than either of them.

That is a strange thing to claim, because it means knowledge can be actively harmful rather than merely insufficient. It says the person in the middle would have been better off never studying the subject at all.

II. Why The Two Ends Agree

The two ends agree because they reach the same answer by completely different routes.

The person at the low end has no theory about the subject. They do whatever is obvious, which means they are responding to the situation directly rather than to a model of it. This works more often than people expect, because for most ordinary problems the obvious answer is close to correct.

The person at the high end has a theory and has worked with it long enough to know where it applies and where it does not. In a lot of fields, working through the theory properly leads back to something close to the obvious answer. The difference is that they now understand why it works, and they can recognise the situations where it stops working.

The person in the middle has a partial theory, and that is the difficult position. A partial theory is confident enough to override the obvious answer but not accurate enough to improve on it. They have learned enough to distrust their instincts and not enough to replace them with anything better.

There is a social element to it as well. The person in the middle has the most incentive to demonstrate what they know. Someone who knows nothing has no reputation to protect, and someone who knows a great deal has nothing left to prove, but the middle position is defined by knowing more than the people below it, and that is a claim which has to be shown rather than assumed. Complicated answers are the easiest way to show it, which is part of why the answer from the middle is always the long one.

III. Where This Shows Up

The same pattern turns up across a lot of unrelated areas.

Investing. Someone who knows nothing tends to buy something and leave it alone. Someone who has read the research also tends to buy something broad and leave it alone, because the evidence on active management is not encouraging. The person in between trades, times the market, rotates between sectors, and pays a cost on every one of those decisions.
Relationships. The straightforward approach is to say what you want and end things that are not working. People who have read a great deal about relationships tend to arrive at similar advice. The middle position accumulates frameworks, attachment theory, and rules about how long to wait before replying to a message, and often does worse with all of it.
Health. The basic version is to eat reasonably, sleep, walk, and lift something heavy on a regular basis for years. Most of the supplement stacks and elaborate training splits layered on top of that produce very little extra for the effort involved.
Work. The obvious task in front of you is usually worth doing. People who have thought carefully about productivity mostly end up doing the obvious important task as well. The middle position builds a system for deciding what to work on, then spends a real share of the week maintaining the system.

In none of these cases is the person in the middle being lazy. They are usually working harder than either of the others. The effort is real, it is just being spent on the wrong part of the problem.

IV. What I Think Of It

I think the observation is correct, and that most of the ways it gets used are not.

It held up when I checked it against my own record, which is the only reason I take it seriously. The years when I was most active were not my best years, and the gap was not small. I was not being beaten by the people who understood more than I did, although they beat me too. I was being beaten by people who had bought something years earlier and never looked at it again. That is an unflattering thing to find in your own numbers.

The first problem is that nobody applies it to themselves. Everyone who repeats this puts themselves at one of the ends. The middle position is partly defined by not being aware of it, so anyone's confidence about where they sit is close to worthless, including mine.

It also treats the two ends as equivalent when they are not. They produce the same answer, but they are not in the same position. The person at the low end is right by accident and has no way of noticing when circumstances change, because there is no reasoning underneath the answer to update. The person at the high end can tell when the simple answer has stopped applying. Taken carelessly, the whole thing reads as an argument for not learning anything, which is the opposite of what it actually shows.

The middle is also not optional. It is the only route from one end to the other, and nobody reaches the high end without spending a few years being confidently wrong in complicated ways first. Passing through it is not the failure. Settling there is.

The last problem is that calling somebody a midwit is not an argument and cannot be answered, and it usually appears when the person saying it is losing. Whoever is making the point also decides what the two ends say, so the ends always agree with them.

What I have kept from it is a habit rather than a rule. When my answer to something is more complicated than the obvious one, I try to check whether the extra complexity is buying any accuracy or whether it is just there. Often it is just there, and I go back to the simple answer. That is a smaller conclusion than the idea implies, but it is the part I can actually use.