Wine lovers have known for centuries that decanting wine before serving it often improves its flavor. Whatever the dominant process, the traditional decanter is a rather pathetic tool to accomplish it. A few years ago, I found I could get much better results by using an ordinary kitchen blender. I just pour the wine in, frappé away at the highest power setting for 30 to 60 seconds, and then allow the froth to subside (which happens quickly) before serving. I call it “hyperdecanting.”
Although torturing an expensive wine in this way may cause sensitive oenophiles to avert their eyes, it almost invariably improves red wines—particularly younger ones, but even a 1982 Château Margaux. Don’t just take my word for it, try it yourself.
But set up a proper blind taste test to avoid subconscious bias among the tasters. That’s a bigger problem than you might imagine. Researchers who examined the voting records of wine judges found that 90 percent of the time they give inconsistent ratings to a particular wine when they judge it on multiple occasions.
To avoid bias, use a “triangle test,” which is a scientifically rigorous way to test for a perceptible difference between wine prepared two different ways. Get as many judges as you can—10 is the minimum to get good statistics. Give each judge three identical glasses, and label the glasses X, Y, and Z.
Hyperdecant half a bottle of wine, and save the other half of the bottle to use for comparison. Out of view of the judges, pour an ounce or so of wine into each glass. The undecanted wine should go into two of the glasses, the hyperdecanted wine into the third, or vice versa. Vary the order of presentation among the judges so that not all are tasting the hyperdecanted wine first or last. Record which wine goes into which glass, and have the judges guess which two of their wines are the same.
You’ll probably find that hyperdecanting does clearly change the flavor of the wine. To determine with scientific rigor whether your tasters prefer the hyperdecanted wine requires a more complex trial called a “paired preference” test, or “square” test. But a blind side-by-side comparison works passably well, too, and requires no math.