Expectation

[Margaret Pilgrim]

I've found it interesting to notice how often in recent threads people's differing expectations result in completely different evaluations of similar dining experiences. Many have visited the restaurants in question on the basis of the same reviews, which have obviously conjured up totally different images of what that dining experience would be, hence wildly enthusiastic praise from one poster and withering disparagement from another. The problem is that both reports are true and valid in terms of the reality of the individual writer.

Personally, we have experienced real disappointment at some universally loved restaurants, and haved probably caused our own problems by expecting too much on subsequent visits after experiencing a near perfect first visit at others. At the same time, we frequently find treasure in food and service when we least expected it.

It has become rather obvious that someone who dines mostly in very expensive restaurants may well overestimate the value of a moderately but not extraordinarily well executed inexpensive meal, and that someone with no experience in haute cuisine may be overly impressed with an ambitious and inventive plate that lacks perfect execution.

How do we read so as to perceive with more acuity what our individual experience will be in a given restaurant? How can we more accurately compare someone else's experiences and therefore expectations to our own? Saab says it best in its commercials: "Find your own road." But it's not easy.

[marcus]

As I'm sure we've all come to realize, communicating real information via a message board is very difficult. The information is incomplete and imprecise, as well as being guided by different levels of experience, levels of discernment and subjective preferences. Occasionally one can find someone whose opinions are sufficiently similar to one's own that thay have some relevance. I know that this is not a popular view here, but when I have a real and serious new restaurant decision to make, I am still largely influenced by the better guidebooks such as Michelin and Gault Millau (not Zagat), where I believe that the results derive from skilled and trained people using at least a minimum level of objective process applied to a very broadly based set of choices. Not perfect at all, but better than most alternatives.

[Margaret Pilgrim]

Very true, Marcus. However, my reference to the forum referred to eG people's perceived experiences, not the source of our personal misadventures. The places that disappointed us were, in fact, GM coup de cour at the time of our visits, and dining rooms and hotels reviewed and honored by Michelin. My point is that by the time that the consumer reads a M or GM review, close to a year has elapsed. The venue visited is not the same one that the initial reviewer/s experienced. In addition, there is the effect of that very review on the sheer numbers of eager clients and the capabilities of the provider.

[robert brown]

Margaret, it is good to see this topic of discussion starting to get some play in more than one forum at a time. You put it in compelling terms. I wish there were compelling resolutions. It seems that there is no way that widely-differing opinions about the same establishment will ever go away. It is easy to identify the dishonest restaurant and restaurateur who is clearly playing off the diner with no knowledge and looking to maximize his profit as much as possible, As one moves up the ladder, however, it really becomes impossible to identify a restaurant that even the most experienced eGullet members would all agree upon. The cliches "One man's meat is another man's poison" and "There is no accouting for taste" certainly apply. That you have had some less than good experiences at Gault-Millau "coup de coeur" restaurants is upsetting.

No doubt we have differing ways of deciding which restaurants to visit that we have never visited before. How people chose a restaurant is an interesting topic. Lately I find myself thinking the the Guide Michelin model may be the best we can hope for--a kind of "the less said the better" approach. Then I think how the Italian Michelin guide seems less reliable that other Italian guides and the Guide Michelin for France. I have stated before on the site that there is some metaphysical haze called consensus: a haphazard coming together in each person's mind based on long-ago- read articles and recent word-of-mouth recommendations or criticisms that drive a person to chose one new restaurant over another. I can't explain it any other way. It is the way I often go about it. I would say that it works about half the time.

[Bux]

I think this is actually a much more complex issue than it seems on an initial reading of Margaret's original post, but to answer what I see as basic premises of that post, a rave review--not only of a restaurant, but a movie, or almost anything that can be reviewed or enjoyed--will often set an anticipation level so high that my enjoyment of the restaurant, movie, book, event, etc., will suffer and be less than if I came upon it without any advance reading or knowledge. Yet at the same time, there is no one's word I can absolutely trust. I've thought I've had friends whose taste paralleled mine perfectly, at least until we've had a meal together and had opposite reactions. I've thought there were reviewers whose opinion I could trust, until they gave a rave review to a restaurant I've hated or or a bad review to one I loved. Not even the reaction of my daughter whose tastes I think I played a major part in forming and who has an incredible palate with an ability to deconstruct a dish as no one else I know, can accurately predict how I will find a meal.

The more complex aspect of the original post for me is that if we take two different people who are likely to have two different reactions to a restaurant, and if have them read the same review, we are likely to end up with two different people entering the restaurant with two different levels of expectation which will temper or magnify the two different reactions they are likely to have anyway. If I follow Margaret correctly, she's also saying that these two people are also likely to read the single review with different prejudices or at least with different perspectives. If I haven't, I suppose I could fall back and note that much of this is all about what we come away from an experience or restaurant, with, rather than what anyone, or a chef, puts into it.

The real problem is that if the two people eat on different nights and order different foods, there's no scientific control. How often does anyone find that two dinners at the same restaurant will bring the same reaction and memory? Over time, a pattern may develop that will allow me to predict a lot about my next meal, but invariably, my second meal at a restaurant I absolutely loved the first time, is a bit disappointing the second. I often wonder how much better a second meal might be at a restaurant that wasn't good enough to draw me back immediately.

Marcus suggests that it's not popular here for people to say they trust Michelin or any guide book over and above what they may read here. Perhaps not, but one of the reasons I think this site is succeeding as it has, it s that it's quickly become much more than just a series of first hand restaurant recommedations. I think it's not so much finding a person whose opinions are similar to one's own--I'm already convinced, even those who seem to share my opinions will lead me astray at some point--but it being able to put the opinions in a context.

[marcus]

I think this site is succeeding as it has, it s that it's quickly become much more than just a series of first hand restaurant recommedations. I think it's not so much finding a person whose opinions are similar to one's own--I'm already convinced, even those who seem to share my opinions will lead me astray at some point--but it being able to put the opinions in a context.

As I participate in eGullet and other message boards, I'm coming to the conclusion that these sites are primarily about self-expression and interpersonal communication within a community of people with like interests, rather than deriving utility from the actual information and opinions presented, although there is some of that as well.

[Jonathan Day]

If the goal is to amass data that will accurately predict one's own reaction to a restaurant, there are at least four sources of added variance to deal with:

1) No restaurant is ever on totally consistent form. Perhaps the chef is absent, or getting ready to open a new restaurant, or the sous chef has a hangover, or a supplier sent a bad batch of foie gras. As has been noted elsewhere on eGullet, it's very difficult to reach a conclusive picture from one visit.

Beyond this short-term variance, it is well known that even great restaurants change over time. The Moulin de Mougins was once great; it then became horrible; it is now somewhat better, though still a shadow of its old self. Vergé's rise, decline and partial resurgence happened over many years. Today's Grand Vefour is far better than the one I first visited. So multiple visits are needed, and these can't be spread over too many months or they have little chance of predicting a reaction.

2) Diners differ in their tastes and even in their physical reactions to food: allergies, bad memories associated with a particular food, and so on. Someone with a very sensitive palate may be constitutionally unable to eat blue cheese.

3) As Margaret points out, expectations make a difference. More generally, food is a complex aesthetic form because of the high level of interaction between the "consumer" and the "product": there is no other art or craft that we literally take into ourselves. It is striking how many reviews on eGullet and in print media begin with a discussion of the welcome, service, reservationist, and so on, and let these first impressions influence perceptions of the meal later on.

4) There is a socially reflexive aspect of "reviewing", especially on an interactive message board like this. Diner A goes to restaurant X and criticises it. B, who loved it, slams A for having an uneducated palate. Or C praises a restaurant and D dismisses C as being undiscriminating and inexperienced. Then A and C, having been criticised, strike back at their critics for being snobby, or biased to like a famous restaurant, or whatever. So not only does a diner have expectations of what she will experience, but may also worry about what other members will think of her comments.

I continue to think that we should try to make our reviewing less one-dimensional (a scale from good to bad, or no stars to 3 stars, or whatever) and more focused around specific criteria: "if you like elegant, minimalist design in a meal, you'll love restaurant X. If you are looking for fun with friends over rough but very tasty food, you'll like Y." With what other art form do we put some much attention on stars or numerical ratings? Should the Louvre start putting star ratings on its paintings?

The advantage of Michelin and Gault Millau is that their reviews are based on multiple visits, anonymous (in the case of Michelin, at least, I don't know about G-M) and are updated annually. But they convey relatively little about the atmosphere and the character of a given restaurant.

Personally, I try to read multiple opinions on eGullet and then to triangulate with the guides. I use Michelin, Gault Millau and Gantié, as well as Le Bouche à Oreille (click here). Even then, it is rare that I encounter exactly what I expected in any restaurant. Of course the opinions of locals are often useful, but never on their own.

Does this mean adopting a pessimistic or relativist view? I don't think so. It does imply a recognition that reviewing restaurants is very difficult; and abandoning the view that any of us, no matter how "well dined", can write the definitive statement on any restaurant.

[marcus]

The advantage of Michelin and Gault Millau is that their reviews are based on multiple visits, anonymous (in the case of Michelin, at least, I don't know about G-M) and are updated annually. But they convey relatively little about the atmosphere and the character of a given restaurant.

From my perspective, the reason that I lend credence to Michelin, and hopefully Gault Millau does something similar, is that they follow a defined process that creates some level of objectivity and hence comparability.

[Jonathan Day]

This is how I think of Michelin. I wonder about G-M, though. Don't they adopt "agendas" from time to time, and use those agendas to shape their ratings, e.g. their endorsement of nouvelle cuisine?

If you haven't looked at Le Bouche à Oreille, I recommend it. The writing (in French, unforuntately) is often very funny. They rate restaurants from five chandeliers ("Exceptionelle") to five zeroes ("Scandale"), and have, in addition to the cuisine, separate ratings for each restaurant's welcome, service, quality-to-price ratio, decor, bread, coffee and toilets.

[marcus]

This is how I think of Michelin. I wonder about G-M, though. Don't they adopt "agendas" from time to time, and use those agendas to shape their ratings, e.g. their endorsement of nouvelle cuisine?

Christian Millau sold the company, probably 10 years ago, and the days of overt agendas are over. They have tightened up significantly, the ratings are much more strictly applied and overall have been significantly lowered on average. If I were to characterize my perception of the differences between Michelin and GM, over and above Michelin being more consistent, is that Michelin seems to place a relatively high value on execution, whereas GM places a higher value on conception.

[Bux]

The advantage of Michelin and Gault Millau is that their reviews are based on multiple visits, anonymous (in the case of Michelin, at least, I don't know about G-M) and are updated annually. But they convey relatively little about the atmosphere and the character of a given restaurant.

From my perspective, the reason that I lend credence to Michelin, and hopefully Gault Millau does something similar, is that they follow a defined process that creates some level of objectivity and hence comparability.

GM states that they awarded Veyrat 20 points on the basis of one meal.

[Bux]

I think this site is succeeding as it has, it s that it's quickly become much more than just a series of first hand restaurant recommedations. I think it's not so much finding a person whose opinions are similar to one's own--I'm already convinced, even those who seem to share my opinions will lead me astray at some point--but it being able to put the opinions in a context.

As I participate in eGullet and other message boards, I'm coming to the conclusion that these sites are primarily about self-expression and interpersonal communication within a community of people with like interests, rather than deriving utility from the actual information and opinions presented, although there is some of that as well.

I suppose "utility," like beauty, is where you find it, or in the eyes of the beholder. I find a lot of the communicatin here useful, but perhaps not in the direct utility sense you mean. Nevertheless, I also find that interpersonal communication helps me to interpret reviews as well. The news that's posted here is also informative and we often scoop my other sources.

[marcus]

GM states that they awarded Veyrat 20 points on the basis of one meal.

Yes, but this is really an anomaly, possibly done for marketing purposes as has been previously discussed. This doesn't change that fact that they have substantially lowered the average scores of a large number of restaurants that they have been following over a long period of time.

[Bux]

GM states that they awarded Veyrat 20 points on the basis of one meal.

Yes, but this is really an anomaly, possibly done for marketing purposes as has been previously discussed. This doesn't change that fact that they have substantially lowered the average scores of a large number of restaurants that they have been following over a long period of time.

Agreed and for that reason I didn't quote your later post about lowered scores. I was following up on Jonathan's comment about Michelin's multiple visits and I thought your reply about objectivity was worth including in my response. GM, if a single visit suffices at the top, is unlikely to make enough visits to be truly objective in any range. Is Veyrat better than Passard or Gagnaire on the average, or just on the specific day GM visited each. The 20 for Veyrat brought them a bit of publicity, but in my opinion, it pointed up their great weakness by publicly announcing the 20 was reward for a single meal.

[marcus]

I suppose "utility," like beauty, is where you find it, or in the eyes of the beholder. I find a lot of the communicatin here useful, but perhaps not in the direct utility sense you mean. Nevertheless, I also find that interpersonal communication helps me to interpret reviews as well. The news that's posted here is also informative and we often scoop my other sources.

I originally came to this site looking for the lowdown on restaurants, new ones to try, the latest info on those that I was already familiar with. One of the first things to hit me was Cabrales' review of Auberge de l'Ill, a restaurant that I know and like very well, in which she detected an off taste in the foie gras, although she conceded that other diners seemed to be eating their's quite happily. Although it took a while to really sink in, this was probably my first indicator that I wasn't goin to achieve my primary objective to the extent that I was looking for. In fairness, I have found a couple of posters whose palates are similar enough to my own that I can read their reviews with some degree of confidence.

However, I also do find value in the communications aspects of the site, and that's why I stick around.

[marcus]

GM, if a single visit suffices at the top, is unlikely to make enough visits to be truly objective in any range.

It's hard to really know, but I interpreted this somewhat differently. I don't think that they were making the statement that their ratings, in general, were based on a single visit. Rather, they were saying that they had never ever previously had even a single perfect meal until this one meal at Veyrat, and that this was so unusual that it was sufficient for them to raise his score to 20. One can disagree with this, but it doesn't imply to me that their other ratings are also based on one meal. La Ferme de mon Pere remained at 19.

Edited November 13, 2003 by marcus (log)

[vmilor]

Over time, a pattern may develop that will allow me to predict a lot about my next meal, but invariably, my second meal at a restaurant I absolutely loved the first time, is a bit disappointing the second. I often wonder how much better a second meal might be at a restaurant that wasn't good enough to draw me back immediately.

I think the critical test is to eat a SECOND time in a favored place preferably during a different season and then, once the restaurant passes the test, one can be reasonably comfortable to recommend it, at least in the short term before the stagnation and decline dynamics outlined by Jonathan set in. As far as I am concerned I rarely liked a place the second time if I was disappointed the first time and this holds true for different categories ranging from the hot dog stand to neighborhood bistro to H.C. But somehow the reverse does not hold: I am often disappointed when dining a second time in a place I had liked very much earlier. Why should this be so? I have some inchoate thoughts on this and may be worth discussing if you share similar observations.

I also found out that my disagreements with Michelin often are resolved at my expense after repeated experience in a place. More often than not I think Michelin is too stingy as, say, place x deserves 2 stars not one. But after 2 or 3 meals I arrive to the same conclusion as Michelin so I was wrong. Disagreements are often resolved in time as Michelin reacts belatedly. But they react.

Applying similar standards to others, I trust reviews based on one meal less even if the reviewer is somebody I think highly of. The more the reviewer talks about the QUALITY of ingredients and place them in a comparative context, the more I trust the reviewer. Hence a less "sophisticated" diner who is aware of, say, taste difference in line caught sea bass versus farm raised one or a churra breed lamb versus Sonoma lamb may be more trustworthy than somebody who has dined in all 3 stars. But, more often than not this is a false opposition as quite a few high end eaters are also knowledgeable about ingredient quality.

I also derive utility from a well written review in ways perhaps not intended by the writer. Take O Combal review by Bill Klapp and it is very well written by somebody whose knowledge of Piemonte cuisine is second to none . It is also timely as I am selecting 2 out of say 10 or so I am considering for 2 days. I gathered from this review that the food is too precious, the chef is hardly mature as he is imitating Adria, he(Scabin) is overly obsessed with texture and is likely to be very stingy when it comes to grating whife truffles if he offers them at all as part of the 120 euros menu. So I scraped him off the list. What I am trying to conclude here is that if the reader puts the review and the reviewer in a context(Bill Klapp spends considerable time in Langhe and good examples of traditional stuff will astonish him less than me and perhaps he has more than 48 hours time span to indulge in that incomparable aroma), the value derived from a well written review increases dramatically.

[vmilor]

Not even the reaction of my daughter whose tastes I think I played a major part in forming and who has an incredible palate with an ability to deconstruct a dish as no one else I know, can accurately predict how I will find a meal.

How did you do this Bux? We keep hearing that as our daughter(who is now 20 months old) starts pre school she will acquire bad taste and habits from classmates(you know how Americans eat---even upper middle classes) and we will have only minor impact. So I am truly alarmed to the prospects of her devouring badly fried southern food and disgusting pizzas. Please tell me the strategies that worked for you and your wife. Thanks.