Wednesday, September 4, 2013
Peer pressure brings wine scores toward the middle
As is often the case with wine competition results, one of the culprits was human behavior.
People occasionally complain that the so-called 100-point scale for wine critics is actually about a 15-point scale for publications, as the Wine Advocate and Wine Spectator rarely publish ratings below 85.
In wine competitions, the scale is a little wider, but it's still more limited than you'd expect from looking at the scoresheet. Scores below 70 are rare, but so are scores above 95, which is why Grand Gold medals rarely happen in Europe without statistical help.
I always try to follow the written rules of wine competitions: to score each wine on each attribute where it belongs. But there are unwritten rules also, and the chart below will show you that it didn't take me very long to learn and follow them -- even though I don't necessarily agree with them.
The implication is that if you're far from the rest of the panel, you're wrong.
There is some logic to this. The point of having multiple tasters is to get a consensus. One taster might like a funky, unusual wine, but if others hate it, it won't win a medal, and that's probably best for consumers.
What struck me, in getting the quick feedback, was how much I was influenced by my peers in keeping my scores in a narrow range. I like to think while I listen to reason, I stick to my convictions if I continue to believe I'm right. Clearly, I do not.
As you can see, I hated the third wine, a German Pinot Gris. It was buttery, low-acid, unpleasant. I gave it 67. But I didn't pull this score out of thin air. I marked the individual attributes on the scoresheet at the top.
If you look closely, you'll see that a wine in the middle of every category -- average in every way -- should get a 74. If you decide that it's below average in a couple of categories, it could slip below 70. Probably should slip below 70.
Think of the famous example of a classroom full of Nobel-winning scientists, graded on the curve: Some of them will get C's and D's. But we're not talking about tête de cuvée Champagne here. On this day, on the chart above, we had German Pinot Gris and Portuguese reds from Setúbal. Some of them were pretty good. But shouldn't some of them be below average?
What happened was, I gave that wine a 67. And two other panelists, both from Germany, complained. One told me that the wine was unflawed, and that only a flawed wine should get a score that low. That's not what the scoresheet says.
But I liked Mundus Vini, and I didn't want to get disinvited for my scores. So look how quickly I went to the norm. I went below 70 again two wines later, and once again got flak for it. I didn't drop below 70 the rest of the day.
Looked at another way, if 74 is average, I graded none of the final 40 wines as below average. My classroom was full of Nobel Prize winners.
How does this affect the absence of gold medals? Our avoidance of extreme scores worked in both directions. Nobody wanted to use the second-from-right column -- below average -- because the scores would be too low. But we also didn't want to use the far left column -- excellent -- because the scores would be too high. And if you didn't go into the far left column, you couldn't score a wine above 87 -- a silver medal, also known as kissing your sister.
In fact, in three days, six of us judged 150 wines: 900 judging opportunities. And only one person (me) one time gave one wine 95 points, just enough for a Grand Gold. The jury chairman remarked on it. But just as when my scores were too low for the norm, I had to defend my very high score, which was also slightly uncomfortable.
The second day, my graph was much closer to the group as a whole. In my self-image, I am a proud independent American, not afraid to speak my mind and stand up for my beliefs. In fact, I'm demonstrably as malleable as everybody else. At least my belief that I'm willing to take my lumps in public when necessary is accurate.
Posted by W. Blake Gray at 6:00 AM