Conversation

I find survivorship bias arguments really weird sometimes. “I did such and such and then we succeeded. And then we did it again and succeeded.” “Blah blah survivorship bias.” Really?

Quote Tweet

Ken Kocienda

@kocienda

May 25

I’ve had several people cite survivorship bias arguments at me over this tweet, and I suppose if you want to believe that the original iPhone was the result of luckily and perfectly dodging all the random bullets that could have shot us down, be my guest. twitter.com/kocienda/statu…

2:21 PM · May 25, 2022Tweetbot for iΟS

Replying to

“I talked to a person who was an expert in that thing, and they told me they did X and Y.” “Survivorship bias!” Ok.👌

Cedric Chin

@ejames_c

May 25

There’s a thin, grey line between “survivorship bias” and “smart thing I can say that repudiates a factoid that clashes with my existing beliefs” and I’m not sure where it lies.

Replying to

Haha yeah! Like a million companies created smartphones, and Apple succeeded coz of luck of the draw. Probability 101 monkeys designing chips something something

Replying to

Yeah... but... twitter.com/benchmann42/st in the original thread nails it. It's not survivorship bias, but it's a similarly limited view. The apparently true "Apple didn't need Agile in the mid 2000s" and the OP's "none of it matters" are *very* different things...

This Tweet is unavailable.

Replying to

Hmm, that tweet seems to have been deleted?

Show replies

Replying to

Whats the cognitive bias where people tends to use "a bias" as a way to avoid learning truth? A variant of confirmation bias?

Replying to

I think it’s called the bias bias. 🙃

Show replies

Replying to

My guess is that the advice being given, what worked for the expert, isn't distinguished from the exact same work and behaviors that resulted in failures It would be interesting to know why the approach they took allowed them to succeed when the same approach elsewhere failed

Replying to

Survivorship bias is precisely not about what succeeded but about what failed. I don't know the specifics of this argument but I think it is right to call something that worked for the iPhone - and failed elsewhere - as a prime example of survivorship bias.

Replying to

Those arguments confuse necessary with sufficient. X behavior may be necessary -- but not sufficient, for success. But Ken's tweet is interesting because he's saying X behavior is UNnecessary. S. bias seems irrelevant; he isn't prescribing anything, just pointing out waste.