This is one of the most stressing NYC dilemmas: You’re in a rush to get somewhere and the train is nowhere to be found. You’re underground so you can’t check the train status on your phone and you have no idea what’s happening, whether the train hiding is just around the tunnel bend about to pop out or it’s been rerouted to the land of wind and ghosts forever. How long should you wait for the train before going above ground again and trying some other way to get where you’re going?
Now there’s finally an answer: An engineer analyzed the MTA’s public data to determine how long you should wait before giving up: 11 minutes. After that, the odds that you’re facing a serious delay are high so you might as well swim your ass home.
This is probably the closest thing we’ll get to countdown clocks, which are still waaaay in the future for most stations. Erik Bernhardsson, who the head of engineering of Better, who worked on the analytics team on Spotify for years, helping develop the Discover Weekly feature, lays out his strategy for figuring this out in a blog post here. There’s a lot of complicated analytics involved, but the breaking point according to his data is 11 minutes; if you cross that point, your wait time is a sunk cost, after which you’re likely facing a huge delay, so you might as well go above ground and take a cab, bus or bicycle home.
So when should you give up waiting? One way to think about it is how much time you think it’s worth waiting. The time you already waited is “sunk cost” so it doesn’t really matter. What matters is how much additional time you are willing to wait. Let’s assume you want to optimize for a wait time that’s less than 30 min in 90% of the cases. Then the max time you should wait is about 11 minutes until giving up (this is at the point where the yellow line cuts the 30 min mark).
He also has this factoid about the L train:
Interestingly, L stacks up pretty well against the other subway lines, despite its notorious delays (and websites such as is the L train fucked). The median waiting time is the smallest out of all the lines, and even the extreme case compares favorably.