Pundits are not forecasters. Even though we occasionally make quantitative predictions, it’s not really our comparative advantage — that’s the job of people with big data sets and statistics packages. This makes it hard to hold us accountable for our mistakes. Deciding whether pundits were right or wrong involves a lot of personal judgement — both in choosing which predictions to evaluate and in evaluating how right they were. And so attempts to gauge pundits' prediction records will inevitably be exercises in extreme subjectivity.
That said, I think it’s still important to note where pundits got things wrong. This is not necessarily so we can separate pundits into “smart” and “dumb” categories; most opinion writers have some valuable point of view about something, and everybody will make some mistakes. Instead, looking at our records is a way of understanding our thought processes — our areas of expertise, our blind spots, our bouts of motivated reasoning. Understanding how a thinker thinks allows us to more easily separate the signal from the noise.
I noticed the New York Times did a recent series in which they asked their columnists to list something they got wrong. The responses are all pretty interesting (though some are less than humble). So I thought I’d list a few of the things I’ve gotten wrong since I started this Substack in late 2020. This will be especially useful for people who only subscribed lately, and may have missed my earlier mistakes. The four things are:
Missing the rise of meme stocks during the GameStop bubble
Thinking Covid relief spending wouldn't cause a consumption boom
Underestimating Covid variants
Overestimating the appeal and effectiveness of cash benefits
In each of these cases, I’ll try to explain what I got wrong and why, and where possible I’ll also try to point to someone who got it right. It’s good to diversify your sources of opinions and analysis!
1. The Gamestop bubble (January 2021)
Keep reading with a 7-day free trial
Subscribe to Noahpinion to keep reading this post and get 7 days of free access to the full post archives.