Quibbles with Fearless Critic's Guide to Austin?
- tom in austin Feb 25, 2007 03:25 PM
It seems to me that we Chowhounds, dedicated to deliciousness above all else, ought to set the record straight for those who may be misled by some of their more curious or inexplicable scores -- be they too high, or too low.
I brought the tome to work so my coworkers and I could make educated lunch decisions about places in the area that we had not been to. Since we started using it heavily to guide our decisions, we have noticed several glaring discrepencies.
I agree. I flipped through a self-styled _Fearless Critic Austin_ and saw several scores and descriptions that made no sense. Of course, that doesn’t mean that the reviewers are right about the things I like and wrong about the rest. Aside from the keeping-it-real schtick, however, these “experts” have nothing new to offer me. To be fair, I should admit that I generally don’t use guidebooks to tell me where to eat locally, and only consult them for maps and addresses when I travel to unfamiliar areas.
You probably wanted to get into their ratings of specific places, but I’m afraid that I didn’t read closely enough for that. Maybe someone else will post on the subject?
I’m glad that you started this thread here rather than on the Austin board, although, as you said elsewhere, some local ‘hounds may not respond here. My feeling is: If I didn’t buy this book myself, why would I want to constantly be informed of its ratings—not just in a thread like yours that was specifically about FCA, but even in other discussions that are just about local chow? Plus, let’s not forget that chowhound.com reaches a large audience. Constant casual references to this guidebook (which are almost always positive, on our home board) amount to free advertising. The publisher should have to pay for that. Somewhere else.
One of my big complaints with the FC guide is that the scores seem completely half-assed. Establishing a meaningful continuum for their scoring system is basically impossible.
Take several samples for Mexican food of various styles. At the top, they have El Chile at 9.2 and Fonda San Miguel at 8.9. So far, so good; these restaurants are obviously pretty darn good. They also score Manuel's at 8.9. I personally disagree with Manuel's being tied for second best Mexican in Austin, but I'm no Fearless Critic, so what do I know? (They also score Uchi at 9.7 as the best overall food in Austin. Again, this makes sense to me.)
Taking the score out of it, we could just say that El Chile is (per FC) Austin's best Mexican food -- 100%. Uchi is Austin's best food, period -- again, 100%. All other Mexican food can be compared against the best Mexican and the best Overall as percentages of the top score. This helps give context to the otherwise inscrutable gulf between the scores themselves.
Next, we've got FSM -- at 8.9, it is 96% as good as El Chile and 91% as good as Uchi.
Some more sample scores:
Polvo's - 8.3 - 90% as good as El Chile, 86% as good as Uchi.
El Meson - 8.1 - 88% EC, 84% U.
Chuy's - 7.7 - 83% EC, 79% U.
Angie's - 6.6 - 71.7% EC, 68% U.
Taco Cabana - 5.8 - 63% EC, 60% U.
Guero's - 4.9 - 58% EC, 55% U.
Nueva Onda - 4.4 - 48% EC, 45% U.
Cisco's - 3.4 - 37% EC, 35% U.
Maudie's - 2.9 - 32% EC, 30% U.
Enchiladas Y Mas - 1.4 - 15% EC, 14% U.
Looking at these numbers, here are some claims they're thusly making:
"When considering deliciousness only, Polvo's is three times better than Maudie's, and about twice as good as Guero's."
"Likewise, Taco Cabana is four times better than Enchiladas Y Mas."
"A meal at Nueva Onda is less than half as delicious as a meal at Manuel's."
Do these sentences even make sense? Especially in context of our experiences at these restaurants: does even one of these sentences ring true? Every one sounds blatantly abusrd to me. Then again, you know what they say about opinions...
Chief amongst this mathematical debacle is the question: How could they begin to justify the inexplicable 1.4 scored to Enchiladas Y Mas? Did one of the owners' kids beat up one of the FC team's writers' kids? Did they not get the respect they felt the Honorable FC Team deserves? Was the meal not comp'd? Did they have to wait too long for a margarita? Do they honestly mean to say (for instance) that Chuy's is nearly six times better than Enchiladas Y Mas? Six. Times.
Think of something six times more than something else. Say, you're going down the highway at 55. Someone flies past you at six times your current speed. That guy would be going 330 MPH.
The best hitter in baseball hits 46 HR in a season. Next year, someone does SIX. TIMES. Better. A simple 276 home runs.
Scoring a restaurant as well-liked and well-reviewed as EyM is (at best) a silly attempt to be provacative and generate publicity, and at worst the equivalent of a food review hit-job.
re: tom in austin
Just as a first reaction, it's hard to take FC seriously having seen this. Chuy's isn't six times as good as anything.
Not wanting to give these folks any money, I'll see if I can't find a copy at the library soon so that I can say something cogent about the reviews themselves.
You can amuse yourself w/ scores (sans reviews) on their website:
To see all their Mexican, for instance, go to "Find Restaurants" in the topnav, pull down Cuisine for Mexican and choose nothing else. You'll have to hit "Next" six times to see everything, but there you have it.
More examples of weirdness: they rank Capitol Brasserie at 9.0, tied w/ Aquarelle; both are ahead of Chez Nous at 8.2. I should point out that this means Polvo's is a better restaurant, food-wise, than Chez Nous, and El Chile is better than Aquarelle.
They score Madam Mam's at 6.8 -- tied with the bland Thai Noodle House. No Thai in the Austin area, including Little Thailand (not listed), cracks into the 8's. (Bangkok Cuisine caps it at 7.7.) Per FC, Madam Mam's is scarcely better than Taco Cabana.
Another bizarre scoring:
re: tom in austin
I don't think you can take the ratings to mean that a 9.0 restaurant is three times better than a 3.0, or 9 times better than a 1.0. If one rated using your system then most of the restaurants would be above 5.0, which wastes the entire 0-5 range. I talked to Robin Goldstein, the editor of the book and he specifically stated that his goal was to spread the ratings across the whole 0-10 range, which is reasonable to me.
That said, there are some questionable ratings. I don't know much about Mexican so maybe you're right about Enchiladas y Mas.
One example is Sambet's (7.9) which considering that Gumbo's is 7.4 and Evangeline Cafe (6.0) I'd be more inclined to rate Sambet's a 5.0.
Others are Cafe Josie (9.3) which I think should be 8.4, Fino (9.2 -> 8.5), Aquarelle (9.0 -> 9.3), Malaga (8.4 -> 9.0).
I do find the bulk of their ratings reasonable, more so than any other guide such as the Austin Chronicle. Their arrangement of restaurants by genre and neighborhood is also very helpful.
Yes, Chowhound, eGullet and other internet sources will be the most reliable but one does not always have the time to parse through all those threads to make a decision.
re: tom in austin
Well, the copies in my local library are out, so I'll make a few comments based just on the web site for now.
My biggest problem is with the name, and with the resulting attention it gets from those who seem to want to prove their edginess by following a different leader. The catch is that (with a few exceptions that serve more to prove the rule than otherwise) the general thrust of the ratings does not differ from the consensus of the unadventurous.
Since I was at the library, I pulled a Fodor's off the shelf. The authors of FC fearlessly seem to agree with Fodor's almost all the time. The only disagreement noticed was on Guero's, which Fodor's liked more than FC. It is interesting when two presumably independent sources both overrate the same restaurants (some examples in my opinion are the Driskill Grill, Castle Hill, La Traviata, and Gumbo's).
When the East Side reared its ugly head, they bravely turned their tail and fled. Searching for "Mexican" in the two regions of "East Austin" and "Southeast", I found a total of 14 places reviewed. I've seen MPH hit that many in a single week, and with free and more useful reviews. While they did make it to Angie's, El Mesón, and El Regio, they miss nearby great places like La Regiomontana, Rosita's al Pastór, and Seis Mesas. It's not just limited to Mexican food, though. They don't even bother to review Tony's Southern Comfort.
As to the results of the process, I can respect the concept that other people have different tastes that may even seem absurd from my point of view. The important thing is that I can put the review in context and thereby make some use of it.
Based on the one review that can be seen on the web site, I can't do that. The 10-paragraph review of the Blue Star Cafeteria opens with four paragraphs on the lampshades and musings on what they imply about the restaurant. It ends with another half paragraph on decor and a paragraph on service. That leaves less than half the review to cover the food. The smug descriptions of the hamburger, pizza, and meatloaf spend their words trying to be clever with unusual but unenlightening comparisons ("an aggressive spice mix that has all the refinement of a lawn statue of Michelangelo’s David in the lawn of an Italian-American Jersey suburb") and populist lowest-common-denominator appeals ("who wants to eat plain broccoli?").
Their surprise that a char-grilled burger would not be as greasy as a similar-sized fried burger tells of a lack of experience. For those who shudder at missed details, the use of "mitochondria" where "nuclei" would be more appropriate and also the missing "neither" might be off-putting.
All in all, the sample review is a poor effort. As someone who has not been to the restaurant in question, i have gained nothing of use from having read it. If I read more reviews from the same author, I might glean a little more information from their track record, but since the "FC" really seems to be more than one person, that might be a little like counting into a 6-deck shoe.
By the way, anyone who is really looking for comments on the restaurant can find much more meaningful comments from several points of view here: http://www.chowhound.com/topics/305678
These "critics" presumably have an audience. Their book is lighter than the yellow pages, and includes more restaurants than national guidebooks. I really can't see, though, how it rises above them in helping either visitors or locals find good chow. I hope my fellow 'hounds will continue to share their own experiences. That remains the real hope for deliciousness.
I'm with you, Knoblauch. I’ll take my categorization of Austin’s restaurants without the unimaginative Fodor’s redux, please. This guidebook is the same old conventional wisdom with an added touch of snarkiness—all packaged in a way that makes readers feel “fearless” just for buying it. Brilliant marketing of a mediocre product.
Frankly I LOVE the book and think they're pretty spot on. I love how they don't pander to the places around town that everyone is supposed to like just because they're "Austin institutions". Sure there are a few ratings that are a bit off, but I have to agree with most of their reviews.
I enjoy the book as well: it really helps open folks up to new places they haven't tried. My coworkers generally rotate between places like Subway for lunch; using this book's reviews, I've coaxed them into trying places they otherwise wouldn't.
But some of their ratings are so far out there they can only be called absurd, and they have several clear prejudices regarding the restaurants that have nothing to do with how delicious the chow itself is.
One of many:
"...unless it is that the almost entirely white clientele really does prefer that aforementioned ... to anything more authentically Mexican."
Nice. Race-baiting. Real strong way to make your argument, FC. Aren't white people lame? They don't even like authentic Mexican food!
Never mind the comment isn't even remotely true, by the way -- the place is owned by hispanic people, staffed by hispanic people, and has many, many hispanic people amongst the clientele. I've been there probably fifteen times.
I'm sorry, but I cannot trust any "published reviews" in Austin (i.e. media OTHER than Chowhound) except for the FC. I think the FC is entirely accurate. It was so nice to read an honest assessment of the places here, and even quoted some of the same things and my husband and I have discussed amongst ourselves. This has been said, but can we please get a new batch of critics in this town? It seems the only choices we have are "critics" that love everyone and never really focus on the food or service, or bad tex-mex love fests that gush over Maudie's (please, please- why???)
I do like menu postings on Foodhawk, but an online site that give reviews of all restaurants here (www.guidelive.com in Dallas, www.nymagazine.com and www.menupages.com come to mind), and divided into the correct categories, like not just "Asian" or "Indian/Mediterranean") would fill a huge gap.
and so I wait...
I'm not saying that Fearless Critic is worthless or terrible. Simply that some of its reviews are obviously inaccurate, and blatantly so, as if their intent was to be deliberately provocative and titillating.
I'm also saying that their point system for scoring restaurants is farcical and absurd.
I find sometime FC to be spot on and sometimes terrible but that's the nature of reviews. What I like about my copy (and seems to be on the website) are their lists. Restaurants open after 11pm on a weekday? Covered. Chinese food close to South Lamar (though have you to scan the type of cuisine column for South Lamar section) Covered. I haven't seen as comprehensive list of restaurants (though I haven't tried the Chronicle's website since they've redesigned). Lists of places that are kid friendly, have good patios or live music. All have made it worth the money for me.
re: Carter B.
I agree with you completely. The value that the book provides has little to do with the all-over-the-place reviews or the absurd scoring. The actual legwork they did: a decently comprehensive list of restaurants, categorized in useful ways for reference (by region, by cuisine, by hours of operation, etc.) -- very nice.
Hi, as an author I wanted to briefly respond to a few of the points above.
On the Mexican restaurants or apparent discrepancies: there were three reviewers, who, like any three chowhounds, each have our own biases. I think this came out most with respect to Mexican and Italian restaurants. I did my best (which as some of you say, might not be good enough) to benchmark for these, but as it was impossible for any one of us to visit all 390 restaurants, this is an imperfect science. My best advice would be to look at the initials after each review and get used to each of our personal preferences insofar as they still filter through.
On the issue of ratings: I wanted to use the whole scale from 1-10. It's curved around a 6.5 with a standard deviation of about 1.8, so an 8.0 doesn't mean a restaurant is twice as good as a 4.0; it just means that within Austin, the former restaurant falls in the top quintile (or, really, quartile), and the latter falls in the bottom quartile (or whatever, I don't have the exact numbers in front of me). I don't understand why in all these 1-10 ratings scales, everything falls between 7 and 9.
On the issue of Mexican restaurants in East Austin not being included: you're absolutely right. I'm working on expanding it a lot. I wish I had the time or budget (or stomach :) ) to review 600 or 1000 restaurants.
On the issue of Thai restaurants: I really believe Austin doesn't have very many good ones. Show me a kaffir lime leaf in the greater Austin area (outside the new Chinatown supermarket), and I'll show you a Thai restaurant that deserves more than an 8...
Amidst a lot of helpful feedback, the one comment I vehemently disagreed with was the insinuation that we took free meals. On the contrary, we spent an enormous amount of money (paying for incognito meals and refusing advertisements from restaurants) to remain completely independent from the restaurants, and I hope it shows.
Robin, thanks for responding. That takes guts! The fact that you peruse Chowhound only elevates your stock, in my opinion.
I wasn't insinuating you took free meals. I was simply brainstorming how you could have put Enchiladas Y Mas at a sub-2 score.
I hadn't considered the scales you used for scoring might work in that manner -- probably a good thing to cover in the book, or did I miss it? Now that I understand your scoring a little better, I'll work up a spreadsheet tonight to see how much sense things make. If what you've said works out, I should be able to report normalized scores that have no hidden curve (i.e. scoring where 10 is ten times better than 1).
I've looked up the numbers, and if it helps, the mean (for food) is 6.48, the median is 6.70, and the standard deviation is 1.70. For experience, the mean is 6.89, the median is 7, and the standard deviation is 1.80. We don't use numbers below 1.0, so were we to do a perfect normal distribution of ratings from 1.0 to 10.0, the mean and median would be 5.5 for both food and experience. We're a bit above that, which reflects the fact that I think Austin is a bit above the national norm in both categories (and a bit more in experience than food). Also the jumps (e.g. Uchi being rated 0.2 above any other restaurant in Austin) reflect what we perceive as gaps in quality between one restaurant and the next. Within food, distributions are as follows:
There's also a (perhaps not helpful enough) explanation of the ratings system in the introduction to the book (p. 2). Hope this helps. I'll be interested to see what you come up with...
Robin, I've played with the numbers and tried to make them work.
Only 1.07% of your reviews are in the same point range as Enchiladas Y Mas (for example), and approximately 4% are within the same quintile.
We can warp the numbers into a more reasonable shape using logarithms. I'll make an easily followed example of what I mean: If we call each quintile grades (A, B, C, D, and F), we can make a little more sense of your scoring. This is a ridiculous amount of contortion to do, I might add: essentially, we're saying that "1 isn't 1 -- rather, the range of numbers between 1 and 2 are actually a larger range that starts at 0% and ends at 59%".
For this model to continue, each quintile would map to a "secret score" that would be progressively smaller in distance from the previous "secret score". (Example: 0-59% = F, 60-73% = D, 74%-84% = C, 85-93% = B, 94-100% = A)
Again, forcing your numbers into a sensible paradigm requires mathematical acrobatics; can we at least agree that your scheme is flawed?
Also, would you respond to my other criticisms? The are:
* the written reviews often don't match the scores
* some reviews lend themselves to fallacious construction: for example, the Enchiladas Y Mas review uses race-baiting
* some of the scores (and reviews) fly in the face of everyday experience at the reviewed venues: for example, the Enchiladas Y Mas review's race-baiting is completely inaccurate
More details on these points can be found in my comments above. If you have any specific questions, let me know.
Finally, thanks for being a sport and responding. I think that is an impressive way to handle criticism.
re: tom in austin
tom, sorry for the delay in my response.
the chowhound moderators have asked me not to comment on specific restaurant reviews, because this is a board intended for consumers to discuss food media with each other, but in any case i haven't been to enchiladas y mas so it would be difficult to comment on it -- i'd just go back again to the point that the book was written by three different people.
as for the scale issue, i think what your point indicates is really that my ratings were too high, rather than too low, in that if i were serious about using the whole 10-point scale then the lowest quintile should be 0 to 2 (or 1 to 2.8 or whatever if i was using 1-10 rather than 0-10), while in fact it's more like 1 to 4.8 (i think you miscalculated it as 5.9 or whatever because you were assuming there were 390 restaurants rated, whereas there were only 376 plus 14 without ratings, so you should run the numbers again). well, perhaps. but consider the fact that i didn't include fast-food restaurants, crappy neighborhood pizza joints, etc., so i guess you could say i nudged the median up somewhat to account for the fact that much of what would make up the bottom quintile of all places to eat (e.g. applebee's, pizza hut) didn't even end up in the book at all.
as for written reviews vs. scores: i think it's possible for a restaurant to get a low food score but a positive review for other reasons (e.g. magnolia cafe), or for a restaurant to get a relatively high food score but a negative review because it is so overrated or overpriced (e.g. jeffrey's). i don't think a specific score needs to correspond to a specific degree of negativity or positivity -- rather, the score is communicating one thing (the overall quality of the food as compared to other restaurants) and the review something else (everything that can't be encapsulated in a number).
as for the point about race-baiting: perhaps the statement was poorly worded in the sense that it implied that *just because* a lot of non-mexicans went there, the place was inauthentic. it was merely one bit of evidence pointing toward a larger correlation between authenticity and clientele. i don't think it would be inaccurate to say that, in general, mexican restaurants that are frequented mostly by mexicans tend to be more authentic than those frequented mostly by non-mexicans. it's telling, for instance, that while there are of course plenty of non-mexican chowhounds who seek out the authentic places, there are also a lot of people--most of them non-mexicans--who are still going to taco bell on south congress and oltorf and spending $5 for "chalupas" or whatever when for $2 they could have amazing tacos al pastor two blocks away at la moreliana meat market on south congress.
Robin, thanks again for your continued gracious responses.
Your points are well taken, and while I think the methodology of intentional review/score discord is confusing, I get what you're saying.
The final point you make is, however, fallacious. I do not know how I can carry on this conversation w/o examples, so I'll use some. Enchiladas Y Mas is (for instance) owned by two Hispanic gentlemen (brothers, I believe), the staff is largely Hispanic, and many of the customers there are Hispanic. Counterexample: El Meson is often, as far as customers are concerned, gringo city. Yet El Meson is absolutely more "authentic" (a term I try to avoid with regards to Tex-Mex, as in my opinion, classic "Americanized" Tex-Mex is a completely legitimate cuisine of its own, worthy of its own appraisals of authenticity), at least, it is more like food you would consume made by residents of Mexico than EyM.
In other words: You have identified a trend that you think is telling as it appears to map to another trend. Like many statistics, you can create fallacious correlations. Note how crime increases in the summer, as do ice cream sales. It would be absurd to assume that ice cream sales and crime rates are somehow related; yet this is something you have done. I'm sure many 'hounds will disagree, but in my opinion saying that food quality and customer ethnicity are somehow linked is thoroughly ridiculous.
As a different point, no "classic Americanized Tex-Mex" restaurants get extremely positive FC reviews. From this, I can only deduce that you and your fellow reviewers just don't enjoy this sort of food. In such a case, that sort of information should probably be shared up front. How silly reading broad wine reviews would be by someone who despises merlot based solely on principle; all readers would get an impression that merlot is apalling, simply because the reviewer is prejudiced.
Oh, to conclude: I really appreciate your bold dialectic on these issues.
re: tom in austin
Correlations are not rules, and just because there's a correlation between Mexican clientele and authenticity doesn't mean there aren't plenty of exceptions to that rule (both false positives and false negatives). But I do think it puts you in the small minority if you really believe that there's *no* meaningful correlation between Mexican clientele and authenticity when it comes to, say, tacos.
That said, I think it's a really interesting and fuzzy debate on the topic of Tex-Mex vs. Mexican. On the one hand, Tex-Mex is not Americanized Mexican: it is its own cuisine, something that sprung up in Texas and along the border sometime in the 1940s, and at its best, this cuisine can be extraordinary. At their best, queso, cheesy enchiladas with that brown meat sauce, and margaritas are some of my favorite comfort foods. Often, they're not at their best.
But it's also not so simple to divide restaurants between the two categories. This is actually something I really struggled with when putting together the book. Should I have one cuisine category called "Tex-Mex," and another called "Interior Mexican" (wherein the word "interior," in Austin at least, has strangely become synonymous with "authentic")? OK, it's obvious that Baby A's is the former, and La Michoacana grocery is the latter. But what would Curra's, which serves both cochinita pibil and brown-sauce enchiladas, be? What would Polvo's, with their pipian and their famous queso, be? What about Habanero, famous for fajitas (an American invention) but also known in the Mexican community for their menudo? What would Fonda San Miguel and Manuel's, both famous for their margaritas (which, by most accounts, were invented in the US), be? It's nothing more than a sliding scale. Where do you draw the line?
Rice and beans is a parrticularly interesting area of overlap. Almost all Tex-Mex and "interior" places serve rice and beans, but the Tex-Mex places often prepare the beans without lard and the rice without stock. They deserve to be downgraded for that. In my view, there is simply no excuse for making rice and refried beans bland when they can be so incredibly delicious, even if bland is the Texas style.
And then there are the Tex-Mex places that serve horrible tacos al pastor, dry chopped pork overwhelmed by pineapple, even while serving good queso. They should be commended for their queso and their cheesy enchiladas, but shouldn't they also be penalized for their al pastor, even if you know better than to order it?
One advantage the ultra-authentic Mexican grocery-store taqueria-type places have in our rating scale, you might say, is that they don't attempt two different cuisines and execute well on only one of them. They attempt just a few things that they know how to do well--barbacoa, carnitas, and al pastor, say--and they do it right. And that is something that I do think they deserve to be rewarded for. Even the best Tex-Mex places often seem to suffer from a peculiar inability to keep their menus limited to what they know how to do right. If they took the grilled chicken breast with a side of veggies and the shrimp al diablo off the menu, their score might well improve.
P.S. One more note of explanation: I don't really even know what it means for one restaurant to be "twice as good" as another. I think the only sensible way to rate restaurants is to say, OK, A is the best restaurant you've ever been to, and B is the worst, and everything else is somewhere in between and belongs somewhere along that scale. Many food critics make A 4 or 5 stars, and B one star; Zagat, for instance, makes A a 30 and B (for some reason) about an 12. Others make A 100 and B (for some reason) about a 68. and I made A a 10.0 and B a 1.0: 1.0 is a meal that is not only disgusting but gives you food poisoning. 10.0 is the pigeon and red currant lasagne at Ambasciata in Quistello, Lombardy; a porterhouse with creamed spinach at Peter Luger; a corn tortilla with pollo en mole coloradito I had one time on the roadside somewhere on Highway 175 outside of Oaxaca; a fried snakehead fish with sour curry and som tam in Nakhon Phanom, Isan, northern Thailand; and a spit-roasted Magellanic lamb on the Peninsula Valdes in Patagonia. Everything else is somewhere in between. Maybe someday I'll taste something that moves everything else down another 0.1...here's hoping for another 40 or 50 years of chow on Earth :)
I just happened upon this thread cuz I am going to be in Austin in mid May. I always find it interesting when you try to put an objective score on something subjective. Just think of ice skating in the olympics. I spent enough years in biostatistics heck in grad school to realize that the best you can hope for is a general guideline than an absolute. we will never know the criteria of the reviewers. to some people, applebees is heaven on earth, and they are not wrong, because they love it and they will rank accordingly. while others will ding even the best place for the slighest perceived fault. People will also rate a place on what they think it should rate, not on their actual experience. Thats why I like a review rather than a plain number. I am leery of going to a restaurant just based on say a 22 for food and skipping the place with a 21, when it may be the more honest restaurant in terms of food (authenticity, taste, and love placed in meal)
every opinion is subjective. so is a trusted friend's opinion. when i come to trust a friend's palate, i come also to believe that he or she is more likely than not to be right about which place is better -- A vs. B -- and so i would ask him or her, not having been to either, "Which should I go to, A or B?" having faith that his or her trusted opinion is better than nothing. Think of a ratings scale as nothing more than an ordered set of hundreds of answers to the question "Which should I go to, A or B?", written by a food critic that has earned your trust by agreeing with you (or not, as the case may be :) ) about restaurants that you have been to. I think in that sense it can be helpful.
If I told a friend that I was considering going to Beluga, Din Ho, Hudson's on the Bend, or Taverna, and they replied (as FC's scoring suggests) that they were all equally delicious, that would not be helpful. That's one reason I agree with littlestevie's point on numbers being less than useful. (Of course a friend who rated the chain of Rudy's BBQ joints higher than Ben's Long Branch and Sam's and who placed the Salt Lick at the top of barbeque deliciousness would not be a source of food recommendations for long.)
With a trusted friend we could discuss what I felt like eating that night and come to a conclusion. Some people who post on chowhound can help in the same way. The book always says the same thing.