Replying in parallel, because I have further thoughts to share. Ever since this comment thread started I've been thinking a lot about those times of hyperproductivity I've described, and I've identified couple things that were true in both cases, and thus are hypotheses worth checking out:
- In both cases I had a hard deadline and a well-defined set of goals, but there was a big gap between "minimum acceptable outcome" and "best possible outcome", with rewards scaling with the outcome. Something back then made me set my sights on the high end of that spectrum.
- I wasn't really blocking other people - if I didn't meet my deadline, I'd suffer some limited (and mostly non-financial) loss, but nobody else's work would be destroyed by this. I.e. in one case my work was irrelevant to anyone but myself, and in the other it was an optional (though strongly desired) dependency.
- I had people to impress. In one case personally, in the other I wanted to make our group look impressive in the local community.
- Both cases have had a significant "like hell I/we can't do it, just hold my beer and watch this" factor. I.e. the deadline was tight, and relevant people didn't believe it could be done in the time remaining.
These factors are a bit harder to reproduce. Most of situations I encounter these days don't have truly hard deadlines ("there's a release next week" is almost never a hard deadline in practice), and yet at the same time involve potential losses to other people (if the release actually happens and I don't deliver, I might derail it). There's little space to go above-and-beyond, in a way that doesn't feel like setting yourself fake goals for the purpose of self-gamification.
I'll be thinking about ways to test these hypotheses and, if any of them turns out to be a significant factor, to engineer my life to make use of it.