The wines I’ve made have been reviewed infrequently over the years. Different reviewers, different wines but the results have shown a predictable monotony: 84-89 points, nearly every time. Same with competitions: the wines always medal — frequently gold, but never double-gold or best of show. Seems that I show a remarkable consistency.

But hey… I’m used to it. I’ve long reconciled to the fact that my winemaking style does not meet the most common expectations. Scores and medals aside, the wines do very well in ranked blind tastings, but are sometimes a “love it or hate it” proposition. And I’m thankful that they sell well… even without high scores. People taste them, they like them.

That’s not to say a great score would not help my brokers and salespeople sell more wine, and in the current economic downturn better sales would be most welcome. So these days we are sending our releases out for review more frequently. Recently we scored another couple of… no surprise… 89′s. The reviewer had great things to say about the wines, descriptions which I plan to use in upcoming marketing efforts. But I’m not going to mention the scores. Why not?

Because an 89 is really a 63 — mathematically, anyway, when the ranges of scores are normalized. The publication I’m discussing here gives out scores on a scale of 50-100, split into six categories: from “undrinkable” to “superior.” I arrived at a graph (see original blog posting) by assigning the six categories a value of 1 to 6, calculating the fractional values between 0 and 6 represented by the 50-100 point scores, and then normalizing these fractional values back to percentages. So if “above average” is 3-4 out of 6, an 89 is 3.8, or 63% of 6. I’m exploring this particular system just as an example — pretty much every wine ranking publication uses something similar, explicitly or not.

Now if someone wanted to argue that my recalculation is bull, I’d be OK with that — even I think it sort of is. A proper analysis would evaluate the actual frequencies that scores are awarded; Leo McCloskey could probably clarify (though I doubt he would) or perhaps Peter O’Connor has these numbers. Bogus or only partly so, my analysis does point out some funny stuff about 100-point wine scoring systems. Like, that just 30% of the points available in the scoring range are assigned to wines that are below average, and 33% of the points are devoted to wines that are outstanding or better. Talk about grade inflation.

This is unlikely to be the intention of the rankers who employ a 100-point scale — but there it is. Realistically, everybody — from the dewiest newbie to the most experienced and educated palate — is rating wines at just three levels.

The top level is: “According to my taste, this wine is really good. I mean really good. You should spend a lot of money to buy this, because it’s worth it. As an investment, as a date-impressor, or simply as a measuring stick to tell people ‘mine is bigger than yours,’ this wine is the schiznit. About a third of what I write about fits this category.”

The middle level is: “I’ve tasted many, many a wine I have enjoyed more or less as much as I have this specimen. You should not be disappointed even if you have to spend more than the average American’s weekly wine allowance for a bottle of this juice. It’s good. Not ‘blow the kids’ 529s to invest in cases of it’ good, but ‘good enough’ to be right in the middle of the range of wines I would bother to drink. About a third of what I write about fits this category.”

The lowest level of rating is: “Meh. This wine did not impress my palate today. It is OK as these things go — not too overtly flawed, or thin, or hot, or too out of balance — but I’ve tasted better. You should thank me — I just spent an hour of my life tasting this bilge… er, I mean… ‘wine’ so that you don’t have to. Seriously. I mean, what were they thinking, sending me this crap? Yes, I wrote about it but if you have the budget, you will be better served going for something I’ve ranked in the middle tier. About a third of what I write about fits this category.”

There is a fourth level — everything that does not get ranked at all. These wines represent, I’d guess, something like 85%-90% of the total volume of wine sold in the world. For the most part these are wines that are not worthy of review, but some fraction of them are wines that just don’t get reviewed because they are too quirky or too rare or too… something.

This post has been incubating a while and in the meantime we’ve received another set of results from a different publication. News flash! — some 87s and an 84. {yawn} Like I said, it makes no difference to me. I am not driven by any desire for external validation. I’m not going to make these wines differently just to get higher scores — I believe that would be the wrong play in the short term, and I’m positive it would be wrong in the long term.

But still… there’s no doubt — and no denying — that better scores would help our sales. Joseph Heller cribbed this plot.

For comments to this blog see the original posting on John Kelly’s blog: notes from the winemaker.