Codie Sanchez

He's interested in cool problems, and it's just a series of cool problems.

Well, I try and yeah, if I'm doing the interview, I try and stop that before it goes too far.

Yeah, well, pulling the goalie.

So Cliff Astis wrote this brilliant paper in which he was talking about the phenomenon of in hockey, there is a tactic that's used if you're losing late in a tight game.

one strategy available to coaches is take the goalie out, put in an extra offensive player, and that will increase your chances of scoring a goal, but also,

increase your chances of being scored upon.

And if the other team scores an additional goal, your chances of winning quickly go to zero.

But if you didn't pull the goalie, your chances of winning were pretty close to zero.

So the only way you can actually give yourself a decent chance of winning is to take this dangerous step, this risky step.

And Cliff's point is that people wait way too long to pull the goalie.

They wait until the game's almost over.

In other words, when the payout of the strategy is minimal.

The real...

opportunity with pulling the goalies if you do it with, you know, five minutes left in the game, not 30 seconds left in the game or a minute left in the game.

I forgot what his exact, he actually kind of calculated if you're down two goals, you know, how many minutes you should, before the end of the game, you should pull.

He did the whole, but his finding was consistent, which was that people who know hockey really well, coaches,

consistently misuse this strategy, render it useless by being far too timid.

And they're timid because they're scared of looking bad.

that if you pull the goalie and it blows up in your face and the other team scores an additional goal, then everybody says, you should never have pulled the goalie, right?

And you look like a foolhardy decision maker.