Radu Craiu is being serious about the lessons to be learned from comedy:

I had a dream in which Poisson’s cousin claimed that the 21st century will be either humorous or poisonous. While I still look for a relevant reference, I confess that my 21st century is a lot more tolerable because of stand-up comedy and, of course, statistics. The two are not as unrelated as they seem, and I argue that the idea of getting some inspiration from comedians stands up to scrutiny. I hope you don’t think that I am going to write an IMS column suggesting that a statistician’s work is a joke—in fact, I argue, quite the contrary: humor is a funny business that can seriously inspire.

Comedic success is built on years of working small rooms, sometimes with only a few onlookers. Even after doing the work, results can be a mixed bag. Some days one goes away with the image of three puzzled audience members; other days one watches as the room slowly empties, dreams of success going up in smoke before punchlines have a chance to land. If this sounds unlike anything you have experienced, remember your early contributing talks at a large conference, say JSM? Or, more recently, that online talk where the only feedback was the echo of your voice asking over and over whether there were questions or suggestions?

Failure for a comedian is unambiguous—heckling and critical reviews are as “in your face” as it gets—but it can also be illuminating. A devoted disciple of the craft will gracefully take the punches and use the time-honored alchemy of combining transpiration and inspiration to turn them into successful punchlines. Honing a joke or a story takes many iterations, just like our papers require multiple rewrites and iterations before publication. Self-citing often brings to mind Amy Poehler’s dictum: “I am big enough to admit I am often inspired by myself.”

There is a delicate balance that is kept in comedy circles between cumulated prestige and current output. It is often said that the great stand-up comedians continue to be judged by the quality of their latest show, which is perhaps more aspirational than realistic. The prior probability that the same is true for statisticians or not is almost surely influenced by the size of your professional clique.

A research presentation is story-telling constrained—by time, logic and ability to interest. It is a monumental optimization task and sometimes I wish we could find and study another group where their members spend time alone on stage speaking about their obsessions… The gigantic challenge of giving the conference’s post-dinner talk while keeping people in their chairs is on par with late-night live comedy where the audience needs to be lured into paying attention and buying drinks. The successful cocktail of a strong stand-up act goes beyond a good joke, just like a successful talk often relies on more than a good idea. Timing and an absence of inapposite details, powdered with subtle humor, can raise the levels of entrancement.

Comedy that tackles daily events feels perhaps more relevant to the contemporary zeitgeist but tends to have a short shelf-life. In opposition to that, the comedy gold inspired by some of the more perennial issues that plague us has a way to avoid oxidation, as recently pointed out by a recent New York Times article (“What’s the deal with adulthood? 25 years later ‘Seinfeld’ feels revelatory”) featuring one of the best comedy shows. The comedic trade-offs implied when choosing between ephemeral and perpetual topics are not unlike the ones we face in the midst of this data science revolution/involution.

Perhaps more fundamental inspiration comes from the comedians’ reputation of calling it how it is. In coffee-fueled interviews, Seinfeld & Co. talk endlessly about the constant pursuit of truth, and the incessant exposure of frauds and hypocrites. In the past, I called statisticians the “science ninjas” and the “universal scientists” because the scientific method has placed our discipline squarely at the center of innovation. We are the ones who are supposed to be “keeping it real” by not fawning over fashionable theories and convenient intellectual shortcuts. Whether we are doing it or not depends on our proclivities, but the tools and the mission have been clearly described to all of us.

Differences remain, thank goodness. If statisticians raise the pulse of people witnessing their acts, I cannot tell for sure, but I am fairly certain that, so far, no one has slapped someone else over a misplaced lemma. Unlike the mathematical lore that postulates greatness (only) for the young, the world of comedians seldom sees success early on, perhaps because one cannot scrutinize, much less make fun of things one doesn’t properly understand. As Robin Williams put it, “What’s right is what’s left after you do everything wrong.”

On the face of it, both professions benefit from variation, and both have developed methods to tame it. One could think of a joke as a (very) low-dimensional representation of reality (this is an existence result, and a general constructive proof is not available) where the less canonical the projecting space is, the funnier the joke. In fact, it can be argued that a comedian’s job is to mix signal and noise, while we try hard to do the opposite.

Those who dream of being in the club of funny statisticians will do well to remember Groucho Marx: “I refuse to join any club that would have me as a member.”