How to rank 100 restaurants
Making an impossible task possible
Let’s say you have a list of 100 restaurants. Your collective group of 2.5 colleagues has decided they’re the best restaurants in a major city like … New York. Let’s say you even went through the pain of ranking the top 10. Cool. Fun. You’ve given people something both useful and interesting to argue about. But then let’s say you get stuck. It feels impossible to rank any more restaurants. Oh my god. One person could do it, sure. One critic can just make a list and rank it. But 2.5 folks agreeing on a single top-100 ranking? That’s crazy.
Well, I’m here to help. I oversaw hundreds of rankings projects as an editor for a digital media company. We relied heavily on the format to drive traffic and subscriptions. One time, a famous TV talking head even went on the air and demanded I be drug tested for a ranking project I led. That’s a good ranking.
Rankings are amazing. Dumb and arbitrary. But amazing. A list is easy. A list is even kind of boring. But a good ranking — one with authority and taste, one with a bit of an edge, one that was excruciating to put together but one the author or authors stand by wholeheartedly — is the Internet’s best art form. It’s so much fun to laugh at a bad ranking and so rewarding to invest in a good one.
If you find yourself crippled by a restaurant list you simply can’t imagine turning into a ranking, here. Try this:
Do whatever you did to rank the first part of the list and then … keep going.
Have everyone involved in the list-making do their own, individual ranking and then average out a composite score. If someone only counts as half a ranker — yikes — weigh their scores with 0.5x points. Tough but fair.
Just re-order stuff until the vibes look right. (This is actually fine).
Turn it into a bracket.
Or, even better, use the Pairwise system, where a panel of expert voters are given two restaurants head-to-head and have to choose the better one over and over again until an aggregate ranking comes together.
Create a point system across different categories. Decide which categories matter for a stellar restaurant. Apply different weights to the categories. Argue over the category scores and then add up the weighted point totals.
Lock yourself in a room for the day. Accept that it’s hard and nearly impossible. Don’t leave until you have your list ranked from No. 100 to No. 1. (Always rank down to No. 1, not up to No. 1).
Make the list then do a fan vote. (Don’t actually do this. No one likes a fan vote).
A friend dunked on me the other day saying:
I laughed. I deserved that. And sure. They’re not wrong. You can’t rank 100 restaurants. Not really.
Like, oh, how do you compare a taco stand to a tasting menu? Are you ranking just the food or the whole experience? Unlike a movie or a book, a restaurant is constantly shifting under the weight of a thousand variables — how do you capture that? Also, won’t people … get mad? Is it insulting to rank a great restaurant where people work their asses off every day “last” at No. 100? Is that mean? Isn’t restaurant life hard enough? Wouldn’t it be better if people could make little social graphics about the wonderful places on this inviting list they’ve already visited and the ones they want to try? Wouldn’t that help celebrate this struggling industry? Wouldn’t that be nice? Then everyone can just share cute posts on Instagram with equal levels of excitement about being included?
Or would it be better to treat people like adults. Get over it. Have a point of view. And just rank the restaurant.
All hail the pairwise