It's true that stepping away from something for a time allows you to see that thing in a clearer light. The grind of weekly restaurant criticism is one of the most pleasurable grinds around, but it is a grind. It's difficult to take pause and think about the bigger picture of the job, but a recent six-month book leave gave me that time. And on return to the weekly routine, I found myself increasingly frustrated with my own star rating system. 

If I'm frustrated with it, I can imagine it also frustrates some readers, and really frustrates some chefs and restaurateurs. I put a star rating system in place six months after taking the job at L.A. Weekly, in part because it was something I was accustomed to as a critic and in part because the L.A. Times had done away with its rating system. I thought someone ought to have one outside of Los Angeles Magazine, which only produced starred reviews monthly. Stars are a help to readers, and often chefs want something concrete, something more easily marketable than a quote. With no stars, readers are left to decipher the level of joy or disappointment on words alone. There's an argument to be made that a good critic should be able to easily provide that judgment without the crutch of stars, but it does leave things a lot more open to interpretation. 

Over the years, I've tried to keep my star ratings in line with the things at which L.A. really excels. Top ratings are not reserved for fine dining, and the stars are more about how well a restaurant does what it sets out to do and less about how fancy the restaurant is. This is not unique to my ratings — I think most critics in most cities are moving in that direction. But it does pose some issues, and it's those issues that began to bother me in recent weeks. 

Let's take the example of two recent three-star reviews, Salazar and Gwen. Salazar is a taco joint for wealthy hipsters, Gwen is one of the most ambitious restaurants in Los Angeles. So why do they get the same rating?

Looking in on Gwen; Credit: Anne Fishbein

Looking in on Gwen; Credit: Anne Fishbein

Well, because if you take that idea, of “how well are they doing what they set out to do,” Salazar is killing it and Gwen is falling a little short. Salazar is basically the Platonic ideal of a taco joint for wealthy hipsters, whereas Gwen is an incredibly expensive restaurant that leaves many diners feeling slightly let down. If we were to put them both on the old-school scale, where fanciness accounts for everything in a star rating, Salazar is a two-star restaurant that's way better than it needs to be and Gwen is a four- (or five-) star restaurant that isn't quite living up to its promise. Hence, three stars for both. 

But wait a minute. If Salazar is doing exactly what it set out to do, shouldn't it get five stars on a scale that's determined exclusively by “how well are they doing what they set out to do?” And here's where the theory and philosophy and math of the thing come in. Salazar is not one of the very best restaurants in Los Angeles, and although it's achieving everything it wants to achieve, there needs to be some room in a ratings system for comparative ranking. The rating has to take into account the basic question of: how good is it compared with everything else? 

With that in mind, these two three-star reviews are especially vexing. Because when it comes down to it, Gwen is objectively a better restaurant than Salazar. The service is better, the food is prepared with more thought and care and carefully sourced ingredients, the cocktails are more meticulously put together. This isn't a dig at Salazar — the owners there aren't aiming for BEST RESTAURANT IN THE UNIVERSE. They're just doing what they're doing and doing it very well. 

So what gives? The answer is that any system is imperfect. Some systems have more potential for nuance than the brute pointiness of four or five heavy stars. Certain newspapers award half stars. In my hometown of Melbourne, Australia, the daily paper ranks on a 20-point scale, with “hats” (rather than stars) given for food, ambiance, etc. None of these things seem as if they'd make distilling the complexity of a restaurant into a number (or star or letter grade) any easier. 

The best way I can explain it is very much like what I described above in the case of Salazar being an overachieving two-star restaurant and Gwen being an underachieving four-star restaurant. Think of it like when you were learning about adding and subtracting positive and negative numbers. Remember that scale the teacher drew on the blackboard with zero at its center? I start with where the restaurant might be on the old-school ambition/fanciness/technical ability scale. Then I go from there: I add stars or subtract them based on how well the place is doing what it is trying to do. I take cost into consideration. It's basically one theory of criticism layered on top of another, with a dash of value thrown in. If the aim is to be one of the best restaurants in town and that goal is being met, it might get four (or, one day I hope, five) stars. If it's aiming for that and falls short, down the scale it slides. If it's trying to be a fun taco spot and cocktail garden and it's doing that exceptionally well, up it slides. If it's trying to be a taco truck and manages to also provide one of the best, most interesting bites of food available in California, there's no reason it shouldn't slide all the way up that scale.

This explanation isn't fully satisfying (even to me), but it's the best way to tackle the problem that I've come up with. 

For all its flaws, for how much I fret about it, I still think a star rating system is worthwhile. One day dining may change so thoroughly that it truly becomes impossible to rate using something so simplistic. Until that day, we'll stick with the imperfection of these blunt, pointy, imaginary objects. 

Advertising disclosure: We may receive compensation for some of the links in our stories. Thank you for supporting LA Weekly and our advertisers.