Stand-up comedy—like teaching, and book writing—requires “a collaboration with the audience.”
Statistical Modeling, Causal Inference, and Social Science 2026-08-04
In our bathroom we have this book, “And here’s the kicker: Conversations with 21 top humor writers on their craft,” edited by Mike Sacks. It’s well suited for the throne, as you can dip in and read as little or as much as you want.
The gimmick is that you’d buy the book if you wanted to enter the comedy field. Which doesn’t seem very likely, but, then again, who reads books either? So maybe I’m one of the few people reading this one for entertainment value alone.
One of the interviews is with Woody Allen collaborator Marshall Brickman, coauthor of the 1970s classics Sleeper, Annie Hall, and Manhattan. Lots of interesting stuff in this interview, including this:
Marshall Brickman: [Woody] found a whole new area of insight: relationships of a certain kind, psychoanalysis, and the creation of the so-called loser—mostly with women. To some extent, the lower-with-women character was someone Bob Hope would play, but in a much more general and mainstream way. Woody’s character was more ethnically and culturally specific.
Mike Sacks: It takes true genius to develop a comic character like that.
Brickman: It does, but it also requires a collaboration with the audience. It’s the only way you can do it. You have to get out there and do a variety of material. Over times, certain things, statistically, will continue to work, and other things will drop away, and the audience will tell you what seems correct for you—for what you project onstage as a personality.
Sacks: But even with that said, you can work for twenty years and never connect with the audience half as much as Woody Allen.
Brickman: That’s right. That’s the genius. Creating something that somehow resonates with an audience that strongly.
If only Allen had died in 1986 (after releasing Hannah and Her Sisters) or maybe 1992 (after Husbands and Wives), just think how high his reputation would be. Even setting aside his personal life, releasing a series of flawed projects takes a toll. Not everyone’s a Scorsese or Spielberg who can keep it up into their old age. If Woody had gone in 1986, we’d still be speculating about the amazing work he didn’t have the opportunity to do.
But the thing that really interested me in the above-quoted interview is what Brickman said about collaboration with the audience.
This is true of teaching too! Teaching a class, even writing a book, is a collaboration with the audience. A book doesn’t read itself. With books, the challenge is that the audience doesn’t see it until it comes out. To continue the analogy, a live class is like stand-up; a book is like a movie.
When writing books, we always have the audience in my mind. The trouble is that the book might never reach the audience. That’s what I think has happened with Active Statistics. I’d like to rearrange it and rerelease it as a book called Statistics Stories. Or maybe, Important Concepts of Statistics in Story Form. Also I wish we’d given Regression and Other Stories the more serious and descriptive title, Applied Regression and Causal Inference. Maybe we can rerelease it under that title.
Anyway, yeah, “collaboration with the audience”: definitely.