The Great Wine Experiment I
Wine-os! Ya gotta love us! Is there another group (except maybe audiophiles) where opinion and pontification trumps objectivity? For sure, much of this is due to the subjective nature of smell and taste. It’s just tough to describe the smell of a good (or bad) wine. Is that a touch of tobacco on the nose--or is it cedar-box? On a scale of one to ten, how have the tannins mellowed in a ten year old Cab? Objectivity seems appropriate only for answers to irrelevant questions (e.g. which bottle of wine weighs the most; what is the tallest bottle of Riesling). All questions of taste, smell, and to an extent even visual appeal seem beyond our objective reach. Furthermore, from a rather exhaustive internet search, it would seem that even those questions that lend themselves to objective testing can incur the wrath of oenophiles if the "wrong" answer is suggested (more on this later).
So, where does this leave objectivity in wine? Are we forever reduced to subjective impressions and (worse) bloviating experts? Is it possible to ask any questions at all about our favorite libation and expect a truly objective answer? Sure, many of us have hidden a recent find in a brown bag, asking the tasters various questions (e.g. what is the year?; what is the grape variety?; what is the origin?). I have noticed that even those that dare attempt to answer will often give avoidance responses (e.g. It's very extracted; It's not a Zinfandel; It sure has a nice nose; etc.) rather risk a wrong answer.
In the midst of the mire, I wade in--at least as far as the ankles. The impetus for this experiment came from comments and conversation at nearly every wine tasting and indeed almost every time a good bottle of wine is enjoyed. I'm sure you've heard or said some of them: "This is young (or old) and should really have been decanted." "Wow! This has really 'opened up' since we poured it." "Give it a few minutes-swirl it a bit in the glass-let it open up." So I'm wondering: Does wine benefit (or even change noticeably) from being opened and decanted for an hour or so before drinking? Using more typical wine parlance, does wine 'open up’ (whatever that means--use your own definition) as it stands in the glass for a few minutes?
In searching for a definitive answer on the web, I came across many comments on the order of "Well, everyone knows that decanting makes a difference." and "Of course it makes a difference, I can taste it. It's obvious!" Some of these people were verbally abusive toward those even suggesting that decanting might be ineffective.
I must add here that "everybody knows" wine changes over a very long time after being decanted or otherwise exposed to oxygen. Most of us have had the unpleasant experience of tasting the tag end of a bottle a few days after its initial opening only to find it had become unfit for even a spaghetti sauce additive. The answer I was seeking concerns the affect of a relatively short period of breathing time on a wine. I pondered the time period that might allow the wine a chance to 'open up' or change (if it is going to do so) without going over to the dark side. My totally random and subjective answer was: one hour.
Now, how could an objective test be devised that might provide an answer to the question? What we cannot do is taste the wine as soon as it is opened and poured, noting our impressions of taste and smell, then tasting again an hour later, noting the difference. While this might seem to serve, several problems exist, some of which I will discuss later. One obvious problem is the possibility that our olfactory and gustatory memories are short and objective descriptions of sensory experiences that might help us hold those memories are nearly impossible to qualify.
After a lot of thought, discussion and practice tasting, I devised the following procedure. The big picture: eight fairly sophisticated wine-os would taste 4 bottles of 2005 Justin Cabernet Sauvignon. This was chosen for several reasons. It is young enough (4 years old at the time) to really benefit from decanting (if there is a benefit), all 4 bottles were purchased together as part of a case and have been stored in identical conditions since purchased at the winery. Two of the wines would be opened and decanted an hour before the tasting. The other two bottles would be opened and poured just before tasting. Each taster would be served one glass of each of the four wines. The ordering of each taster’s wine would be random. Each taster would then be asked to pair the wines they thought were the same.
The procedure used was:
1. Two bottles (D1 and D2) were opened and poured into separate decanters at 7:00 PM.
2. At 8:00 PM, the other 2 bottles (B1 and B2) are opened. The four wines were poured so that each flight of wines has a two-ounce pour of each of the four bottles. The glasses in each flight were labeled 1, 2, 3, and 4. It is important to note that the wine in glass #1 for the first taster did not necessarily have the same wine as glass #1 for the second taster or the third taster etc. I won't bore you with the clever details, but when the wines were served, nobody (including me) knew which wines were in which glasses.
3. The wine was served immediately, one flight of four, 2-ounce pours per person.
4. The wines were tasted and scored. The participants were asked to pair the wines together that they thought had been decanted for an hour and the two they thought were just opened. During the scoring, I modified the instructions somewhat (in response to questions by the tasters) by asking the tasters to forget about identifying the decanted pair and just pair the wines together by taste and smell.
Some statistical info: Pairing the glasses at random should yield a 1 in 3 chance of being correct (e.g. glass #1 could be paired with glass #2, #3, or #4, leaving the other two glasses paired by default). With 8 of us, we should expect 2 or 3 correct pairings even if there is no difference at all in the wines. I'll leave it to one of you who had a statistics class less than 30 years ago to compute the odds of our particular scoring being due to chance.
Subjectively, the thing that stood out most was the lack of conversation. Usually our crew is all about sharing their thoughts and ideas. Not here. The silence was deafening! I attribute this to the tasters knowing their wines were in a different order than other tasters. I hope this also meant they were concentrating on the tasting at hand.
Each taster was given a rather simple score sheet. It read:
The two wines that have been decanted for an hour are #_______ and #_______
My confidence rating is (1=it's a guess, 5=very sure) 1 2 3 4 5
As mentioned earlier, people seemed to be having a bit of trouble with the decanted/bottle idea. To make it easier I verbally changed the scoring, telling them to pair up the wines they thought tasted the most alike. This seemed to clear up any confusion. In the end, once they had the two wines they thought tasted most alike, they then chose the pair they thought were the decanted wines. The confidence rating was an add on. I was trying to determine if people thought their selection was a guess or not.
As mentioned before, there are only three possible ways the wines can be paired. To illustrate, the wine in glass #1 can only be paired with glass #2, #3 or #4. Once that pairing is made, the other two glasses are, by default, a pair. To simplify the list below I have used only wine D1 and its possible pairings.
D1 paired with D2 1 (the only correct pairing)
D1 paired with B1 3
D1 paired with B2 4
The confidence ratings, supposedly showing how sure the tasters were of their pairings were (5 being very sure):
Confidence rating of 4 - 5 tasters
Confidence Rating of 3 - 1 taster
Confidence Rating of 2 - 2 tasters
Another way to arrange the data is by the number of times the wine appeared in the decanted group. Since each taster had 2 wines in the decanted group, there were a total of 16 wines chosen by the eight tasters as having been decanted. The breakdown is:
D1 - Appeared 5 times in the decanted group
D2 - Appeared 2 times
B1 - Appeared 5 times
B2 - Appeared 4 times
My statistics classes are a fond (I actually liked stats) but distant memory. I will not attempt a mathematical analysis nor would a confidence factor be high with a universe of eight tasters. (Here I welcome any true mathematical analysis of the data.) That having been said, I will state my conclusions. (n.b. These conclusions may be valid only for these tasters with this wine. See discussion following.)
1. The wine did not change noticeably by being opened and decanted for an hour before serving.
Only one taster grouped the wines correctly. By chance we could have expected 2 or 3. Could that mean that (for example) wine D2 was a bit different from D1 and therefore was legitimately not paired with it? Probably not, as D2 was paired with B2 three times and D1 was paired with B2 four times. Hard to figure how D1 and D2 can both taste like B2 but not like each other.
2. The wine did not 'open up' or get smoother as a result of an hour of decanting.
I say this because the decanted wines were guessed 7 times as being decanted while the wines that had not been decanted were chosen 9 times. This slight bias toward non-decanted wines almost certainly has no statistical significance.
Further and More
If you are like many (most?) wine lovers, these conclusions probably grate, irritate and are just plain wrong. At best they apply only to this wine. If you've read this far, hang with me a bit longer as I try to unwind (justify?) the results.
First off, why do we think wine should change after being decanted? Several reasons come to mind. First, wine certainly tastes different after 20 or 30 minutes in the glass than it did when first poured. I think this is the reason that so many may dismiss these conclusions out of hand. Secondly there are many volatile compounds that surely must evaporate from the opened wine at differing rates. This surely must affect the flavor and smell of the wine as it airs. Thirdly, as oxygen is introduced into the wine, oxidation of some of the compounds occurs changing the smell and taste of the wine. There are other reasons, but they seem to boil down to appeal to authority (yikes!) or some combination of the above. Let's look at these in reverse order.
Yes, decanting introduces oxygen and to a certain extent it will react with the wine and change both flavor and aroma. This actually happens continuously, though very slowly, as oxygen infuses through the cork in an unopened bottle. The question is, does this happen in an hour? My contention is that this is more an explanation for why wine seems to taste different after airing (see below). Put this on hold for a bit...
There are certainly volatile compounds in wine. If there were not, wine would not smell and 'legs' would not form on the glass. I would be very surprised if the volatiles don't escape wine at differing rates. The question here is: Does this happen quickly enough and do the rates differ enough to change the flavor and taste of wine over a relatively short time in the decanter? I'm of the opinion that this again is more an explanation for why wine seems to taste different after having been opened for a time.
OK, I can hear it now. "Damn-it! Wine does taste different after it sets in the glass for 30 minutes! Are you trying to tell me it doesn't? You can't possibly believe it doesn't! If you believe that, you haven't had wine!" OK. I hear you. And I want to first say that I agree. Wine tastes different after it has had some time in the glass. I also believe that wine does not change noticeably by having had some time in the glass. How is that possible?
The problem is that we are talking about two things here, taste and wine composition. One, composition, is an objective, measurable, quantity. Though difficult and maybe impossible right now, it is possible to reduce our favorite Napa Cab to a list of chemicals with associated quantities. This list, accurate as it may be, would not tell us how the wine tastes. The taste and smell cannot be quantified. Your taster and mine are different. As much as we might not like it (or maybe we do), it is subjective. It changes. All of our senses do to a certain extent. When we want to be sure of our senses we employ measuring tools that do not change. What looks to be about five feet long to us today might look shorter tomorrow. If we use a tape measure, we can be assured our senses won't be deceived.
So how do we knit the findings (no significant changes in the decanted wine) with our experience? A very good analogy can be made with two experiences I remember vividly as a kid. The first was an experiment I did with my sister. We filled three buckets with water: one hot, one room temperature and one cold with added ice. We first put both hands in the room temperature bucket-it felt, well, room temperature. We then put one of our hands in the hot water and the other in the ice water and left them there for a minute or two. We then put both hands simultaneously back into the room temperature water. It is amazing how different the water felt to the two hands. Room temperature water now feels both very warm (to the hand that was in ice water) and very cold at the same time! I have since noticed this change also with sight. Clover one eye and look at a brightly lit colored sheet of paper with the other for a minute. If you can get a pair of the old 3-D glasses, wear them for a few minutes instead of looking at the colored paper. Remove the paper (or glasses), and look at your surroundings one eye at a time. The effect is amazing! Colors that one eye sees brightly are muted or changed as seen with the other eye!
Now think about wine with its complex taste and smells. As the taste buds and olfactory sensors are overwhelmed by the high concentration of volatile chemicals not normally found in the air about us, isn’t it even more probable that the sensors get “used” to these smells and tastes (some more than others) causing the taste/smell experience of a wine to change? Further, think of the great interpreter of these and all other senses, your brain. I would think it is not only possible but extremely likely that our brain would change our perception of a wine after having time to process.
Since doing this experiment, I have had considerable feedback. Probably the strongest conclusion I have reached is that people have beliefs about how the world works and are very reluctant to change them. Those that participated in the experiment still talk about how wine ‘opens up’ or that a particular wine ‘needs to be decanted’.
One person expressed skepticism about the wine used. He thinks that an older wine would show the difference that a younger one would not. I am open to this and am waiting for anyone to supply me with four bottles of a nice 15-year-old Bordeaux to repeat the experiment. Anybody...?
Though the 4 bottles used were as identical as I could make them, there exists the possibility that subtle differences between them could affect the result. My thought is that this experiment could be done (and applied to the 15-year-old Bordeaux?) on a smaller scale with a single bottle and a nitrogen wine preservation system. Wine could be taken from the bottle and allowed to ‘breathe’ for an hour or so, more could be taken just before tasting for comparison (all double blind, of course). Again, anybody volunteering his or her nitrogen system?
I welcome your thoughts and comments not only on this experiment but on other ways we can objectively explore wine lore. Be advised that we will have to drink wine to do the exploring. Anybody...?