If you don't:

A) Massively outperform compared to your peers.

B) Massively underperform compared to your peers.

You are almost certainly a mediocre programmer.

If you are in the B category, it is usually pretty obvious to _everyone_ and you've probably seen your performance reviews at _multiple_ employers be below average.

If you massively outperform your peers, you probably find your boss at _multiple_ employers going out of his way to retain you. (i.e. Massive salary bumps, bonuses, promotions, etc.)

If neither of these are true, you are almost certainly mediocre.

For instance, I made a really stupid decision for personal reasons to leave a job where I massively outperformed my peers (my boss was really bad at hiring people) and received a 30k raise a couple years before I left. But, realistically, that was the _only_ job I had that kind of outsized performance and it was more to do with my boss's skill at hiring people than my skill as a developer.

At my jobs before and after that one, I've fell safely in the mediocre range. Never been fired, never had a PIP, and generally got "Meets expectations-ish" performance reviews.

At my current job, I underperform my peers but that is largely because I'm their first hire in ~3 years. I have alot of catching up to do as a result, and I've been there ~90 days.

Mathematically the vast majority of programmers in the world are mediocre by your definitions there. Was that the intent of your post?

Actually, the definition is very localized (you could be mediocre at a big co or a rockstar at a tiny backwater company) and relative. I don't think it's a stretch to assume that the majority of programmers are mediocre, the distribution curve of talent/ability not withstanding.

Yes. He was asking how to be unbiased about self evaluation. You have to paint really broad strokes to do that.

Is it perfect? No.

Do I think the 70% of programmers that probably fit this definition are mediocre? Yeah.

Keep in mind I am calling myself a mediocre programmer.