Part of my job is explaining base rates to people who are, understandably, not thinking clearly about statistics in the moment, because they’re scared about a test result, and I’ve tried a lot of different framings over the years to make the concept land without it feeling like a math lecture delivered at the worst possible time in someone’s life. The one that’s worked best lately, oddly, came out of a Thursday night argument about a card game.

The specific disagreement was with Ines, over whether a particular play was smart or reckless given the odds, and I found myself explaining, at the table, exactly the kind of reasoning I use at work, that the raw probability of a card being in the deck matters less than how that probability updates once you know what’s already been played, what everyone’s discarded, what’s visible on the table. It’s the same logic as a positive screening result meaning something very different depending on how common the underlying condition actually is in the first place, the same math, just with lower stakes and better snacks.

What struck me, turning this over afterward, is how much more naturally that logic landed with Ines in a game context than I’ve ever managed to make it land with a patient in a clinical one, even though it’s structurally the same reasoning. Some of that is obviously the stakes, nobody’s scared during a card game the way they are in an exam room, fear genuinely does shrink people’s capacity to absorb new information, that’s not a mystery. But some of it, I think, is that the game gives you a shared, visible board to point at. I can say “look at what’s already been played” and gesture at actual cards on the table. At work I’m usually asking someone to hold an invisible, abstract set of numbers in their head while they’re also processing something frightening, and that’s a much harder ask.

I’ve started, carefully, borrowing a little of the game-night framing at work, not the game itself obviously, but the instinct to find something visible and concrete to anchor the abstraction to, a diagram, a real comparison, anything besides a bare percentage. It’s not a solved problem. Some days the fear is too loud for any framing to get through, and I’ve learned to recognize those days and back off the explanation entirely until someone’s ready for it.

But I keep thinking about how strange it is that the clearest teaching tool I’ve found for a serious professional skill showed up at my own kitchen table, over a card game, in an argument I wasn’t even trying to make a point in.