DRMacIver's Notebook

Good strategies often fail

Good strategies often fail

I stockpiled food for Brexit. A lot of people were skeptical of me for that, and many people have said that the way Brexit played out proved me wrong and that it was an overreaction.

They are wrong and I was right. Stockpiling food for Brexit and not needing it was the expected outcome. It was never something that I thought I was likely to need, it was a capacity I was building in case of a low probability high cost event. The logic was that either food stockpiling wouldn’t be useful, in which case I had a surplus to use up, or it would be kinda useful in which case it was a net positive, or it would be essential, in which case I would be incredibly glad I had it.

Annie Duke calls this fallacy resulting, and I think it’s a rare instance of something escaping the fallacy fork. Resulting is the fallacy of assuming that because you got a bad outcome you must have made a bad choice, when in fact you might just have got a bad outcome because you picked the option with a 60% chance of success instead of a 40% chance but it turns out that you’re in the 40% chance world. Chance means that you can never perfectly control the outcome, and you should be careful not to let random chance inform your learning too much.

More complicated is that often good strategies you’ll never even know if you succeeded or not. A lot of mitigations for COVID-19 are like this: You’ll never know whether it really mattered that you were washing your hands, or socially isolating, because all you’re doing is stacking the odds on a numbers game. You might get the virus anyway, or you might have never got the virus, there’s no way to know. It’s still a good strategy.

Capacity building strategies in particular are often a good idea, because they can be useful in unexpected ways. I didn’t immediately dismantle my Brexit stockpile when it seemed unlikely that we’d need it, I’ve just been using it as a buffer, so now if we do end up on pandemic lockdown we’ll be well placed for food supplies while self isolating.