The Tasting Triangle Test

[Rebel Rose]

As I mentioned in another thread, The Act of Description, How do you thin-slice your wine? I’m reading Blink, by Malcolm Gladwell, author of The Tipping Point.

In taste tests involving Coca-Cola and Pepsi-Cola, researchers learned that Pepsi did so phenomenally well in taste tests because it was sweeter, and slightly citrusy. But when tasters were given a case of each cola and asked to report back, they preferred Coke. Unfortunately, this was discovered long after the Pepsi Challenge drove the venerable Coca-Cola Company to produce New Coke—a marketing and product disaster.

Also, to really differentiate between two products, Gladwell says we need three samples. Have a friend pour three samples of two wines for you—two of one wine, one sample of the other. Your task is not to identify each glass, but merely to point out the wine that is different.

According to Gladwell, “You will find this task incredibly hard.”

He tried the triangle test on a group of his friends, all “well-educated, thoughtful people, most of whom were regular cola drinkers.” None of them got it right!

He goes on to say, “With three glasses, we have to be able to describe and hold the taste of the first and then the seond cola in our memory and somehow, however briefly, convert a fleeting sensory sensation into something permanent—and to do that requires knowledge and understanding of the vocabulary of taste.”

So, given two very similar wines, can you do the Triangle Test?

[Marc Olson]

Sounds like a fun test.

In one of my regular tasting groups we've done this by purposely putting the same wine in the lineup from two different people. Sometimes the same wint from two bottles will show differently, but this idea should eliminate that.

[Boris_A]

Sounds like a fun test.

←

Fun test? Not at all.

The triangle test is a standard professional test (well, at least here in Switzerland) for sensoric detection of small differences:

- checking the difference whether the wine was bottled with a cork or a screw top

- checking for consequences of different storage temperatures

- checking for the effects of some added acidity

- checking for different degrees of malolactic fermentation

etc. etc.

Edited April 27, 2007 by Boris_A (log)

[Don Giovanni]

However you test is better than not to have tested your skills or others... I have to tell you that as long as it's fun and not belittling in any way...then you can really enjoy yourself.

[Daniel Rogov]

No less complex than the "tirangle test" is the professional's need to "double up" wines at a blind tasting. What this means is that at least one wine in every tasting is presented in two glasses, each with a different number, so that the taster cannot have any clue as to which wine he/she is tasting twice. At the end of the tasting, after notes and scores have been assigned, to discover which of the wines was doubled, to check your notes one against the other.

A certain amount of leeway is allowable of course. Wine tasters are not machines. But over that amount, leeway is not tolerable. In my own case, when I do my tastings wines are always doubled up and if my tasting notes vary dramatically or if the scores are more than two points apart the notes of that entire tasting are discarded, for this proves that either my concentration was not intense enough, my mood odd enough, or some physiological aspect influenced my tastings that day to the point where they are not reliable.

Fun? No way. Does it happen often? Thank heavens, not!!!!!

[ludja]

Interesting post, Rebel Rose and comments by others.

I also just happened to read Gladwell's "Blink". Besides thinking about the triangle tests and variants designed to detect small variations I also found it interesting that people apparently preferred Pepsi in sip tests but that Coca Cola won out when drinking a larger portion. I definatly do modulate my wine choices depending on whether I will drink just one glass or more and if food will be available while drinking the wine. I find that some wines (even nice ones) seem to fatigue my palate more quickly if I don't have them with food.

I should re-read parts of the book and specifically think more closely about when I taste and evaluate wine and how I "thin slice" it... In addition to the fact that he gives some beverage and food examples in the book I think the first example he gives via the experts who can immediately tell that the greek statue acquired by the Getty was a fake even though they can't articulate the reasons immediately. I've had the lucky opportunity to taste a fair number of wines including some of quality and distinction so I feel I've learned but often I can't put it into words. "Balanced" and "long finsish" are the words I feebly turn to most often after tasting a great a wine. (I know these are ivalid and mportant decscriptors vut it's difficult to go beyond that sometime for me besides describing particular flavors.)

[JohnL]

What is often confused in these "tests" is the difference between preference and evaluation.

Untrained consumers are often confused here IMOP.

Taste tests are an unnatural environment. People make purchase decisions using a far more complex set of criteria under far more complex circumstances. Consumers do not make decisions based upon quick taste tests under the conditions these tests are executed.

Professional tasters are trained to assess a wine (or coffee or tea etc) based upon an agreed set of criteria. They are further trained to communicate their assessments using an agreed upon set of terms/descriptors.

This is mainly why I do not bother with the sea of amateur notes found on the internet. It is also why, far too many people believe that wine tasting is a totally subjective endeavor and that notes and scores depend upon a particular tasters preferences (their palates). The notion that one's "palate" is a totally unique and individual entity is basically wrong.