Seeing Stars: Eater Blinks and So Do I

The stars are back.

Two years ago, the Los Angeles Times announced it was dropping the “star system” for rating restaurants. To my knowledge, no other publication followed the LAT’s lead.

In New York City, the Times, New York magazine, the Post, the Daily News, Time Out New York, the Observer, and GQ, all give stars. Among print publications that regularly review restaurants in New York, to my knowledge only the New Yorker and the Village Voice have yet to use stars (and they never did).

Last week, Eater.com (which had long eschewed reviews of any kind) began two new series of reviews, with critics Ryan Sutton (formerly of Bloomberg) and Robert Sietsema (formerly of the Village Voice), who will use separate non-overlapping four-star systems. (New York is the only other publication with separate star systems for low- and high-end restaurants.)

After the LAT announced its decision, the Times’ Pete Wells filed a blog post explaining why his reviews would continue to have stars:

No two critics are going to have the same reaction to a restaurant, and no two critics are going to come up with identical interpretations of the precedent. The whole process of critiquing restaurants is inherently subjective. Readers are free to disagree with the critic. Go ahead and throw the newspaper across the room if you like. That’s part of the fun. . . .

Whether you think that renders the stars meaningless depends entirely on what you expect them to do. If you hope they are going to organize the entire New York City restaurant scene into an objective and verifiable hierarchy of good, better and best, you’re going to find that the reviews are a weekly exercise in frustration. (A corollary: If you read the reviews believing that all restaurants with a given number of stars are meant to be equally good, you’re going to lose your mind.)

On the other hand, if you understand that the stars accompany a review of at least 1,000 words, I hope you’ll believe that they do have meaning. The reviews have to cover a lot of ground. They tell you what kind of restaurant is being reviewed, how it looks and feels, how customers are treated, how some of the dishes taste and often whether it’s worth the price. A star ranking from zero to four can’t do any of those things in any meaningful way, but it can try to serve as shorthand for how strongly the reviewer is recommending the restaurant.

After the LAT announcement, I introduced my own system, which not-coincidentally was a five-step scale, just as the stars had been (four to zero), but—as I then saw it—without the stars’ historical baggage. From highest to lowest, my ratings were:

• Extraordinary
• Category Killer
• Critic’s Pick
• Neighborhood Spot
• Not Recommended

These categories had always approximated my general sense of what the stars ought to mean, although not all critics used them that way. I found in practice that my new system was no more liberating than the stars. In fact, it was less discriminating, because I had used half-stars in the past, but I had allowed myself no way to designate a restaurant as, for example, “Neighborhood Plus.”

Other problems I saw with the star system seem less serious to me now. I do not fundamentally take issue with a restaurant like Roberta’s receiving three stars, as it did from Sutton last week, assuming you agree with his assessment. Restaurants are rated against the ideal versions of themselves, not against others in completely different genres. You could agree with Sutton’s rating, while concluding at the same time that Roberta’s isn’t for you. (That, in fact, is precisely my view of Roberta’s.)

There is nothing to be done about the fact that, on crowdsourced review sites like Yelp, a three-star review is terrible, while to most professional critics it is terrific. In my system, a one-star restaurant is “good,” which ought to be a compliment. I cannot do anything about the fact that some people will read my reviews, see one star, and think, “It must be awful.” The readers who say that haven’t read the review.

There remain considerable differences in the ways critics use the star system. Some publications go up to five stars, although most use four. Some go down to zero, but others stop at one. Some use half-stars; others don’t. Because of this, stars are generally not comparable across publications. For a given publication and critic, the stars, as Wells put it, “serve as shorthand for how strongly the reviewer is recommending the restaurant.”

But the system, in whatever fashion it is used, remains the lingua franca of restaurant reviews.

A few years ago, the Post’s Steve Cuozzo dropped the stars (in fact, he dropped traditional “reviews” entirely), but after a while concluded that he might as well re-instate them. I’ve now come to the same conclusion.