Not too many years ago, I awarded a score of 93 to the wine of Chateau Mouton Rothschild. In the same column, I gave the same score to the Nouveau Beaujolais Village of Joseph Drouhin. Within hours after the column had appeared, I received a phone call from a reader wanting to know why in the world he should spend more than $90 to buy the Mouton when for a mere $6.00 he could buy the Drouhin wine. "After all", he pointed out, "you gave both wines the same score".

What my reader had failed to grasp was that scores are not absolute, all wines being rated within their category. That is to say, the score given to a light, youthful and hyper-fruity wine made from Gamay grapes cannot be compared to that given to a deep, full bodied wine made from grapes such as Cabernet Sauvignon, Merlot and/or Cabernet Franc, the peak of drinking for which may come only five, ten or even thirty years later on. Numerical comparison between the wines of the great Chateaux of Bordeaux and those meant to be consumed within weeks or months of the harvest is akin to comparing (by means of a single number), the current status of a 1998 Lamborghini to that of a 1965 Volkswagen Beetle.

Following that, and in the names of and intellectual honesty, let me get into the body of this little essay by paraphrasing Mr. Shakespeare in stating that I come more to praise scores rather than to bury them. I hasten to add that abusing scores is like abusing wine. As wine is a beverage to be enjoyed in moderation by civilized people in civilized company and often with good food, scores too must be used in moderation, must be taken in context and should never be taken in and of themselves as critiques or complete evaluations of wines.

In brief, I find three advantages to the awarding and then reading of numerical scores. First of all, if you trust the critics you are reading, the scores they award to wines can serve as initial guides or, if one prefers, hints about their overall impression of the wine in question. Second, and again on the condition that they come from a source one considers reliable, numerical scores also give an immediate basis for comparison - of that wine to others in its category and to the same wine of the same winery from earlier years. Finally, for the at least partly knowledgeable or more sophisticated reader, such scores give valuable hints as to whether the wine in question is available at a reasonable value for one's money.

In my own case, even though numerical scores are part of my evaluations, I do not consider them enough on their own, and thus base my critiques on a combination of tasting notes and descriptions. In and of themselves, scores are valueless, for they give so little information about the wine that only fools, yuppies, incurable snobs, and the nouveaux-riches would buy a wine on the basis of a score. In other words, scores that have meaning must be accompanied by an at-least somewhat detailed review of the wine in question, one that will give details as to body, color, bouquet, length, flavor, overall style and other of the factors we seek out in the wines we choose to drink. That is to say, although scores are valuable reference points, they are neither the be-all or the end-all of evaluation.

Even if there was a perfect system for rating wines (and I do not believe such a system exists), no two critics, no matter how professional or well intentioned they may be, can be expected to use precisely the same criteria for every facet of every wine they have to evaluate. My own scoring system, which is based on a maximum of 100 points, is fairly simple (and quite similar to that of both The Wine Spectator and American wine critic Robert Parker).

95 - 100 truly great wines
90 - 94 outstanding wines
86 - 89 very good to excellent wines
81 - 85 good to very good but short of excellent
70 - 79 average but faulted in some way
60 - 69 seriously faulted, not recommended under
60 undrinkable in my opinion

One Major Reservation

Although I value scores given by individual critics, I have a problem with scores derived by committees or tasting panels. It is true that everyone, including the most devoted professionals, have their own prejudices as to what they seek in different wines. The individual taster (assuming of course that all evaluative tastings are done blind, the taster knowing only the broad category of wines he or she is tasting) can, however, let his or her prejudices be known to readers in advance, and is then free to proceed on that basis. In group tastings, however, there is no way for readers to know or to compensate for the biases of those making up the panel. More seriously, in group tastings (especially in competitions), where the scores awarded are calculated as an average of the scores of the members of the panel, , there is a mathematical and psychological process at work (known to statisticians as "regression to the mean") in which outstanding wines tend to fall to the mid-range scores and the score of mediocre to good wines tends to become higher.

To Read About How I Do My Own Tastings, Click Here

© Daniel Rogov