CronoDAS comments on Multiple Factor Explanations Should Not Appear One-Sided - Less Wrong

29 Post author: Stefan_Schubert 07 August 2014 02:10PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (31)

You are viewing a single comment's thread. Show more comments above.

Comment author: Stefan_Schubert 08 August 2014 04:05:00PM 3 points [-]

That's interesting. A colleague of mine raised a similar issue, namely that in a popular science book you don't necessarily want to complicate things by including countervailing factors. In your terms, you settle for the briefest possible explanation. Diamond's and Pinker's books are directed both towards the scientific community and towards the general public, so it's a bit of a tricky case, but since they are such high-profile scientists and since their books have been so influential, I think it is legitimate to criticize them on this score.

A perhaps more glaring example is this. Man City won Premier League 2012 on goal difference thanks to a 94 minute goal which put them ahead of Man Utd. Afterwards, a Swedish pundit was asked to explain why Man City won the Premier League. This is in a sense absurd, since it's clear that if a 38-matches league is settled on overtime of the last game, there is very little that distinguishes the two team in terms of quality. But the pundit's reaction was also absurd: he went on to provide 4-5 reasons for why Man City was better than Man Utd, to which my reaction was, well, if they're better on so many scores, then why didn't they finish like 20 points ahead? The "briefest possible explanation" defense doesn't work here, since it would have been easier just to give one reason, and more adequate given the small difference between the teams, than 4-5. Instead, I believe that he did so because of a deeply felt urge to tell a "story". I think that the halo effect is at play here. Our system 1 wants to tell one-sided stories where the winning team had all the advantages and the losing team was worse across the board.

Now Diamond and Pinker are obviously better than football pundits, but I don't think that the examples are fundamentally different. They, too, are most likely to some degree engaging in story-telling.

Comment author: CronoDAS 10 August 2014 10:11:06PM 2 points [-]

Apparently, the outcome of soccer matches between closely matched teams tend to be more unpredictable than matches between closely matched teams in other sports. So yeah, the only accurate answer the pundit could give to "Why did Man City win the final match?" would be along the lines of "On that day, things went right for them."