This afternoon I came across a paper published in The Journal of Wine Economics (JWE) 3-4 years ago that reported a study where blind tasting was used to test whether expensive wines tasted better than less expensive wines. I was interested in this study because of my long belief that the plethora of wine raters in the world single handedly discourage the success of many really good wines and, at least to some degree, assign ratings for reasons other than quality. In fact, the paper cites a prior study (Hadj Ali et al. 2007) that claimed to find a strongly positive correlation between the price of wine and its Robert Parker rating. In fairness I have not read this paper but it is highly likely that the claim is true.
The blind tastings used to collect data in the JWE paper were well designed and the authors were generally honest about the sample's limitations. When the authors analyzed their data two key findings emerged:
1) There was absolutely no relationship between the price of wine and the quality of the wine as perceived by the tasters.
The authors were honest about the fact that even though the sample was fairly large (more than 500 tasters) most likely had greater than an average level of education and an interest in wine. What this means is that the sample did not represent "average America" but rather those who generally enjoy wine. The authors assessment of the sample is certainly true but my guess is that the sample adequately represents the segment of the population that is most likely to buy wine (at least something other than Ripple). In actuality the data suggested that more expensive wines are less preferred than less expensive wines in the sample but the regression coefficient is so small that I doubt this is meaningful. It is more likely that the variety of individual pallets in wine drinkers does not process a price variable in determining what they like or don't like. While this analysis is not directly related to wine ratings I suspect it is fair to say that if this study were repeated substituting wine rating for price you would get similar results. Face it. Those that rate wines cannot possibly speak for every enophile in the country and likely speak only for one or a few people.
2) Expert wine tasters rated more expensive wines higher than less expensive wines.
On the surface this result suggests that if you have a trained pallet that you are better able to distinguish good from bad wine. Here is where the author's data set breaks down a bit. First, the statistics associated with this conclusion are far less concrete than the first conclusion above. I won't bore you with the details but the way the regression model (yes, now I'm talking like a scientist) falls out leaves some uncertainty. The authors are honest about this. My other concern, not addressed by the authors is how the subsample of experts was defined. An individual in the sample was considered an expert based on profession (e.g. sommelier; about 12% of the total sample). However, I would contend that what one actually tastes when drinking wine is stingily influenced by genetics (the physiologist in me is talking here) as anything else. If this result is true it is equally likely that a trained person is good at recognizing an expensive wine, which adds bias to the result. Further, it is highly likely that many in the remaining component of the sample were skilled tasters with pallets developed equally to those defined as experts. Bottom line is that this separation is pretty artificial.
Another issue I would have with this study is that the wines used were broadly drawn from various varietals and countries. Thus, individual preference for varietal and place of origin were not controlled for. When this is considered the sample size no longer looks big enough to swamp out the error that could result from this omission. Since I live in the Willamette Valley I would very much like to repeat this study focusing solely on Willamette Valley Pinot Noirs (for example) and see how the results hold up. Maybe I will one of these days.
It has been said many times that wines fall into two categories: wines you like and wines you don't like. I honestly believe this to be true. I also believe that the plethora of wine ratings out there today discourage many from trying new things and broadly exploring what the growing wine industry has to offer. There is something for everybody out there because we are all different. Wineries go after ratings because like it or not it drives sales. However, if we can ever convince ourselves to ignore the Robert Parkers of the world I think we would discover wines that are more diverse that we ever imagined and also wine prices that reflected production costs rather than someone else's opinion of quality.
Cheers
No comments:
Post a Comment