I think people are continually surprised by how often I use believability as a metric when evaluating practical advice:
- have you had at least 3 successes in the domain?
- do you have a coherent explanation?
If yes, queue to test against reality. If no, discount advice.
Conversation
Replying to
How do you ask for advice/seek viewpoints about less common topics where the 3 examples might not be available? Also what about advice via failure modes? Eg ‘I did this and it didn’t work’ etc
1
4
Replying to
Oh this is a fantastic question! I think in practice believability is a spectrum. So, if you only have 2 successes, and I can’t find anyone with 3, I’m probably going to take your advice seriously and try it. (Though with the understanding it might not work out.)
1
2
Failure modes are trickier. I find that if you’re more believable than I am in the domain, I tend to overweight what you’ve tried and haven’t gotten to work and therefore ignore it.
In practice this means I might not critically consider context.
3
2
I’ve heard someone say ‘never say “that won’t work”; instead say “we tried it during X and Y and it didn’t work at that point in time”’ but in practice I think this is really hard to do. Haven’t found a good mental trick for this yet.
sees this problem as not embracing the role of randomness.
His solution: More low-downside bets.
He has more than 3 successes 😜
Quote Tweet
"The definition of insanity is doing the same thing over and over again and expecting a different result."
Whoever said this underestimates the role of randomness in life.
1