The Olympics are upon us, a year late and probably many dollars short. There will be no cheering in the stands, because the stadiums will be empty. All your favorite Olympic sports will be on display, even if some of your favorite athletes won’t be, disqualified by an adverse drug or COVID test. There will likely be grumbling about some of the color commentary and the officiating, because there always is.
Some sports can more easily be objectively judged — the track events, for example. First to cross the finish line wins. It’s that simple. There is not a lot of discretion or opinion involved in picking the victor. There might be some questions about a false start or a close finish, but the camera will almost always settle the issue without controversy. Certain team sports are more susceptible to hijacking by bad officiating — penalties are often judgment calls and bad calls can alter the outcome of a game. Still, the victor in basketball or in soccer is determined by who scores the most baskets or goals, respectively, which is not objectively in dispute even if there is some awful refereeing along the way.
In other sports, though, subjectivity is the hallmark of the competition. In Olympic gymnastics and diving, for example, where there are no goals or baskets to easily tabulate, judges must employ their subjective discretion to evaluate the execution and artistry of each performance. Often there is consensus about the excellence or mediocrity of a routine or a dive, and there is a scoring system that calls for point-deductions for certain mistakes. But in close cases, there is no photo that can definitively settle the matter. The final result is determined by fallible human judges. And at the Olympics, where national pride is on the line, there has long been a worry about national bias.
So how do these sports mitigate the risk of corrupt or incompetent judging? Among other things, there is not one judge; there are many. The multitude diminishes the impact of a single biased judge. In Olympic gymnastics there are five judges; in diving there are seven. But that is not considered enough to assure the fairest result. There is another mitigation strategy — the high and low scores are discarded. In gymnastics, the highest and lowest scores are ignored; in diving, it’s the two highest and two lowest. In both sports, it is the remaining three “centrist” judges who determine the outcome.
Statisticians call this kind of data analysis a “trimmed mean,” and its usefulness is recognized across many fields. A trimmed mean omits the outliers, rendering it a better fit where there might be erratic or outsized high or low values in a data set. It is a kind of compromise. It helps to reduce the disproportionate impact of an extreme data point on the group. Imagine trying to understand the wealth of a neighborhood. Suppose there is a single eccentric billionaire living in the sole mansion in a working-class neighborhood. An untrimmed mean would give a distorted picture of that community. In the same way, a trimmed mean operates to provide a more accurate consensus about the performance of a gymnast or diver.
Consider how this is the opposite of how our media-driven political world works. (You knew I was going to make this about politics at some point, right?) Whereas in gymnastics and diving, the extremes are discarded, in American politics they tend to get amplified. Two-time Stay Tuned guest, Adam Grant, who is an organizational psychologist at Wharton, talks about the inordinate attention paid to extremes in political and policy discussions. On the issue of climate change, for example, Adam notes that we pay a huge amount of attention to the full-throated climate deniers. We hear about their harangues all the time. But they are only 10 percent of the population. There are other, less extreme, views held by people who are not in favor of taking aggressive action on climate. There are those who don’t deny climate change but are skeptical that humankind is the cause. There are others who don’t deny, but may not think it worth the cost to correct or that it is not correctable. These people are potentially persuadable. But we don’t focus on them. Think how that distorts debate.
The same may be true in debates about the COVID vaccines. Yes, there are hardcore anti-vaxxers, who will resist the jab even unto their dying day. They get a lot of attention. We read about them and see them on television. But there are any number of unvaccinated people, as Adam Grant also points out, who are merely skeptical. They are merely hesitant. They have questions; they may be waiting for formal CDC approval. These, too, are potentially persuadable people.
By definition, hard core extremists in any debate or in any party are a small bunch. But a small bunch can hijack a debate by, among other things, hijacking the attention of people on the other side of that debate. Allowing this is arguably a form of self-sabotage on the part of people who are reasonable. The extreme views, because they are not trimmed like a biased Olympic judge’s score, skew any understanding of the true consensus.
This feeds feelings of polarization. The whole concept of polarization, of course, is about the poles, where few people stand, as opposed to the spectrum, along which most populations are arrayed. Acquiescing to the specter of polarization prevents consensus even when consensus can be had. It overemphasizes black and white, when most issues are gray. It overstresses red and blue, when many communities are purple.
The concept of trimming the mean, I think, is a useful one in politics. There are those, I know, who want their extreme to prevail, who would prefer to trim the boring, incrementalist center. But, like Walt Whitman, communities contain multitudes. As one statistical primer teaches, “[I]f your goal is to estimate central tendency, then. . . trimming doesn’t discard information, it actually increases the quality of information about central tendency.” And this is how elections are won — by smartly assessing the central tendency.
What else could trimming the mean signify for our politics? Well it counsels that we allocate our attention better. Focus more time on the persuadable than on the outrageous. Accept, like many recently successful politicians, that Twitter is not America.
But what about at the ballot box? No one is throwing out anyone’s votes like they do at the Olympics. Every vote counts. But there are fair and democratic ways to flatten the extremes in the way we elect our leaders. Ranked-choice voting is one example. By allowing voters to select more than one candidate in order of preference, the attention-seeking extremist might see fit to appeal to a broader audience to be ranked someone’s second or third choice. The elimination of party primaries, especially closed primaries, might also help, as voting would be opened up to a larger and less likely extreme population. Some municipalities have eliminated partisan elections altogether, which would also mitigate the impact of extreme partisans.
These are all worth thinking about. Let me know your thoughts. And enjoy the Olympics.