The Washington Post has an interesting article on two economists, married to each other, who share Democratic politics but disagree about why Biden is polling poorly on economic matters. One, the woman, takes what people tell her seriously and tries to incorporate it into her work. The other, seeing good economic data, dismisses what people say about their own conditions and thought process and blames poor polling all on social media and Fox News. He is the person who really interests me, because he is making the same mistake that sports stat heads make constantly: substituting data for expertise.
What I am about to tell you is an oversimplification, but I believe it gets the major points correct.
Long ago, in the before times, when Emmit Smith and Barry Sanders and other running backs roamed the NFL, some number inclined people noticed that passing was a more efficient means of moving the ball in professional football. If you spread the field and forced smaller defensive backs and lumbering linebackers who made their bread and butter stopping running backs cover tall, fast receivers, you had an offense that the defenses of the time could not reliably stop. This perspective gradually took hold in the NFL to the point that running backs came to be seen as interchangeable.
But the change in offenses also lead to a change in defenses. They got smaller and quicker, and thus better able to defend against the pass-oriented offenses. Recently, however, a couple of offenses have changed. Faced with smaller, lighter defenses they did the unthinkable: they oriented around running the ball. Not to the extent of the past, but the best offenses, such as those of the Dolphins, Ravens, and 49ers all drive their offenses through the run. This perplexed many people — after all, the numbers said that such behavior was inefficient and counter-productive. But those people had made a critical mistake: they confused data with expertise.
Data is not reality. It merely reflects reality within the limitations of your ability to measure meaningful inputs. And it is predicative only to the extent that the conditions remain the similar or the data can be adjusted for new realities. This s not to say that data is not helpful or useful — far from it. The opening some people saw during the age of the running back was real. But people who did not notice that the reality had changed under the data, they were unprepared for what people who examined data through their expertise discovered: that NFL defenses were again ripe to be exposed by a change in tactics. People who used expertise to understand the data built championship quality teams. Those who thought that data and expertise were the same thing did not.
Which brings us to the bickering (the article slyly mentions how their teen and tween kids wish their parents would shut up about the matter at dinner) husband and wife economists. As noted, she thinks that what people tell us about their own lived experiences needs to be incorporated with the data. He thinks that the data is all. He is making the same mistake, then, that the people who missed the return of the running game did.
The husband likes to point out statistics about business owners expanding, people buying at a good clip, and wages increasing faster than inflation and say that people are being conned into thinking the economy is bad. The wife, however, notes that people have been falling behind for a generation and that there have been several shocks that affect people in very real ways. So, to the husband, the fact that, say, a wage increase doesn’t go as far as it used to just a year or so ago does not matter — wages are higher overall than inflation, so people should be happy. Similarly, the loss of childcare subsidies and the return of student loan payments must not matter because people are spending at a good clip. The answer cannot be that people are correct that their circumstances aren’t as good as the data say they are — the answer must be that people just have bad vibes from TikTok. Data over all.
The wife, however, notes that people have had several shocks and that there is no reason to believe they are lying about their own lived experiences and perceptions (taking into account the hyper-partisan times we live in, of course). She applies her expertise to the data and notes that there are thig sit may not be presently explaining well because the underlying reality is different today. She would make a great NFL coach. Her husband would be left wondering why his defense is getting run over every game.
This dichotomy matters because far too many people seem to want there to be a simple, data driven answer for everything. Pop some figures into a spreadsheet and that must be reality. It might be nice if the world worked like that, but in most situations, data is only an approximation of reality. If you do not apply expertise to the picture data is presenting to you, you will be confused about what is reality and what is merely numbers. The husband in our story, much like many NFL coaches and stat people, cannot seem to shake the belief that numbers are reality. The wife is wise enough to understand that numbers reflect reality, and that to truly understand them you most incorporate non-numerical expertise in your thinking. Otherwise, you ignore people who are actually struggling and do not notice the conditions that make return of the run game possible.
Leave a Reply