Heat experiment no. 2: Shocking!

A couple of months ago I yanked a nice bottle of Chablis from its companions in my treehouse and placed it outside in the torturous conditions of an Arizona summer. Here are the temperature records for the two places:

Treehouse interior.png

Inside the treehouse

Treehouse exterior.png

Outside the treehouse

(If you are wondering about that dip inside the treehouse around May 27, that’s when we turned the treehouse into a temporary storage for all the produce and meat for Abby and Brendan’s wedding party.)


I was going to leave it out all summer but after noticing the bulging cork I decided to have mercy. So Amy and I had a blind tasting experiment, using the same protocol as for my breathing experiments. We both easily identified the odd wine out. But I misidentified mine as the abused bottle. Here are my notes:

The middle one has a slightly dry acrid note, less fruit on the nose, more acid on the palate. Generally a lost of fruit and thinner, less interesting.

So I figured the middle one was the abused one. I was wrong. Of course, after the reveal, I started second guessing, and wrote the following notes (warning, this is no longer Science):

After experiment tried to find a flaw with the abused one. Maybe a slightly toasty funky note on the nose. Maybe a little flabby on the palate. Maybe a slightly sour milk note. And as I taste the wines more, I think the unabused wine has a little more spine and structure.

The abused wine is a little more viscous. More like yummy syrup than a complex structured juice.

This should all be taken with a grain of salt (oh no, yuk) given my prior expectation that heat would wreck a white Burgundy. I was thinking there would be some clear oxidation and discoloration. Maybe I should have left it all summer as originally planned. Tasting the abused wine again tonight (June 12), a day after the experiment, I agree with myself that it is o.k. but kind of nasty. I’m going to pour it down the drain just as soon as I have confirmation on that. Really I am.

The big takeaway is that this exposure to heat did not make the wine undrinkable. Although in this case the difference was much clearer than it was for my previous heat experiment.

Breathing experiment no. 6

You know the procedure by now. The wine was an aged cru bourgeois Bordeaux, which I chose in response to this tasting note:

Initially, the wonderful black fruit is completely dominated by oak, tannin, and acidity. It takes at least an hour in the decanter for the black fruit to start showing.

The decanted wine was aged for 3 hours. Everyone did very well:

Subject Order Notes
Jim ddb 3 decanter more alcohol
Housten bdd 1 bottle had not opened as much
Bill dbb 1 bottle richer aroma
Sally bdb 2 decanter decanted one has less aroma
Amy bdb 2 bottle less bright than the other two

Everybody correctly identified the odd wine out. Housten and Sally also correctly identified the source, although for opposite reasons. Sally chose 2 as the decanter because it had less aroma, Housten choice 1 as the bottle because it had not opened up as much.

I, it seems, agree with the tasting note that the decanted wine is fruitier. This doesn’t support my hypothesis that decanting the wine depletes the aroma, unless you invoke the masking theory proposed in this post.

[Added after posting] Amy’s note, however, agrees with the hypothesis, although she misidentified the source.

Breathing experiment no. 5


I love this label

For this experiment last night (May 29) I chose an older wine, a 2004 Rioja.  I was interested in whether we would witness the effect described in the 1999 article I mentioned in my previous post:

All four wines [1945 first growth Bordeaux] were dead. Letting them breathe had not improved them; it had killed them.

We followed the usual protocol, with the decanted wine receiving 6 hours of air. Here are our notes on the experiment:

Very difficult to tell. Middle aroma seems a bit richer, so I’m guessing dbd based on aroma.

On the palate very similar.

Amy is guessing 1 is the odd one out.

This time we were both right (Amy’s configuration was bdd), and we both used the aroma alone to make the choice. I’m intrigued by the following tasting note that appeared on CellarTracker this morning:

Classic Rioja with floral tinged red fruit and gorgeous cured tobacco. Fragrant, yet quite tightly wound. This needs hours and hours in the decanter to open things up and let the acidity calm down; it becomes soft and supple when it does. Lovely.

I didn’t notice any difference in the acids, but I have another bottle of this and might try again. Although I should admit up front that I do find it hard to believe that the acid-tannin composition changes much over a few hours. My tasting note on the wine is here. I agree the wine is “tightly wound,” and maybe did not give it enough attention to see past that.

Breathing experiment no. 4

As a reminder, I am working on the hypothesis that the main effect of allowing wine to breathe is a slight loss of complexity in the nose. I’ll relieve the suspense and say right now that this experiment confirmed the hypothesis (as did experiments nos. 2 and 3, at least as far as my own perceptions go).

This experiment was conducted 20 May 2017. I used a wine that is often said to benefit from air, a red burgundy.  The methodology was the same as experiment number 3. The wine breathed for 3.5 hours, whereas in previous experiments it has been more like 6. (I realize I should be recording these numbers.)

My three glasses had wine from the decanter, decanter, and bottle, respectively. Here are my notes (written before the reveal):

ddb at first because the right one seemed bit funky on the nose

second thoughts the middle seems slightly spicier, so maybe dbd

right one seems to have a fruitier palate
no difference in tannins or acids

left two more floral on the nose

going with first instinct

This result is consistent with my theory that breathing causes the nose to lose complexity, although in this case some might find the unmasked floral character of the decanted wine more appealing. It is also consistent with the observation that the effect is quite small; note the uncertainty and wavering.

By the way, my friend and colleague Phil Daro sent me a link to this blog post, which in turn has a link to this 1999 article from the New York Times. Reading them almost made me think I don’t need to bother with any more breathing experiments. They confirm my preconceptions. Except Science! Let’s go from preconceptions to statistical evidence. It will take a lot of breathing experiments to do that. The things I do for Science.


Interesting addendum: my daughter wandered in about an hour after the bottle had been opened, and wanted to try the experiment. She identified the configuration correctly (dbd). Her reasons were entirely from the nose: she thought the middle wine had an aroma that was more mushroomy and gamey, whereas the other two she said were more metallic. Possibly my “funky” had evolved into her “mushroomy and gamey” in one hour. Maybe I need to do an experiment with multiple decanting periods.

Breathing experiment no. 3

IMG_0309This was my first breathing experiment using the Coravin, on May 4, 2017. The methodology is the same as before, except that it uses one bottle instead of two. I poured half the bottle into a decanter 6 hours in advance using the Coravin, and left the other half corked under its blanket of argon. I used a Barbaresco because of this tasting note on CellarTracker, which says that “Over an hour or so, pepper and astringent tannins on the palate overcame the fruit.” This suggests a different hypothesis from mine, that breathing causes the nose to lose complexity. Here the author describes a change in the palate.

The only experimental subjects were me and my wife. I rolled the die for her and she for me. Here are the results:

Person Configuration Choice Identification
A dbb 1 b
B ddb 3 b

We both chose the odd wine out correctly, and I correctly identified it as the wine from the bottle. I tried using both the nose and the palate: the third wine had a spicier more complex nose, so according to my original hypothesis it was the wine from the bottle. On the other hand, I fancied at first that the second wine had less fruit on the palate, which would have suggested a configuration of dbd. After a few sips I lost that difference so went with my first instinct. More evidence for my hypothesis.

But! My wife described the difference between her wines so:

No. 1 is different, sharper and more distinctive, with more acid, whereas the other two seem a little more soft.

This seems somewhat consistent with the tasting note on CellarTracker, in that it can be seen as describing a loss of fruit with breathing. Note that I described the same thing in my very first breathing experiment using this methodology. There, however, I also described the nose of the decanted wine as spicier and more aromatic. Which contradicts my hypothesis.

Obviously, more experimentation is necessary. I think I need to tighten up the descriptors a bit and apply them more consistently. For future experiments I will limit the descriptors to more complex/less complex for the nose and more (fruit, tannin, acid)/less (fruit, tannin, acid) for the palate.

Setting up some long term experiments


First, on April 15, 2017 I put a bottle of Chablis outside my treehouse. It is in the shade but will experience the full summer heat otherwise. I will compare it with a bottle of the same wine from inside the treehouse some time in the fall. (The treehouse is my temperature-controlled storage facility.)

Second, on April 23, 2017 I moved 9 bottles from the treehouse to a cupboard in my wine room inside the house. I have more than one of all these wines, so each has a companion in the treehouse. The wine room can get close to 80º F in the summer, compared with 60º F in the treehouse.  I will monitor the temperature and humidity in both places and do a comparison every so often over the next 10 years or so.


After taking this photograph I put all the wines into a horizontal position, the same as the wines in the treehouse. The complete list is:

2013 Arnot-Roberts Cabernet Sauvignon Fellom Ranch
2012 Domaine Lucien Boillot et Fils Gevrey-Chambertin 1er Cru Cherbaudes
2010 Château Branaire-Ducru
2012 Clusel-Roch Côte-Rôtie
2012 Domaine William Fèvre Chablis Grand Cru Bougros
2010 Lisini Brunello di Montalcino
2015 Joh. Jos. Prüm Graacher Himmelreich Riesling Kabinett
2014 Rhys Pinot Noir Alpine Vineyard
2010 Stony Hill Chardonnay

Breathing experiment no. 2

It was to be over a year before my second breathing experiment. This time I used an inexpensive (but absolutely delicious) Spanish garnacha, and conducted the experiment with my wife, my daughter, and her boyfriend. The methodology was the same as described in the previous post. I asked people not only to choose the odd wine out, but also to identify whether it was from the bottle or the decanter. Here were the results:

Person Configuration Choice Identification
A bbd 1 b
H bdd 3 b
B bbd 3 d
S dbd 3 b

Once again I was the only person to choose correctly, and, unlike the last time, I correctly identified the source as well. It would be strange if this trend continued; random guessing would lead to one third the people getting it right in the long run. But it is too early to make a big deal about it. Here are my notes:

Aroma on 3 seemed slightly faded. Palate indistinguishable.

This experiment confirms my hypothesis that one effect of breathing is decreased complexity in the nose.

One thing I worry about is that bottle variation can confound the results of these experiments. So after this experiment I purchased a Coravin™ (for Science!) and will use it next time to decant half of one bottle. Stay tuned.

Does wine breathe?

It is commonly believed that you should let some wines breathe by opening them a few hours before drinking them, possibly decanting them as well. My first experiment to test this hypothesis was on Christmas Day 2014, with two bottles of 2001 Château le Meynieu.  I decanted one bottle for 3 hours and then opened a second bottle just before a blind comparison of the two, along with some of our dinner guests. There was very little discernible difference. I imagined one had a little more fruit on the palate, and that turned out to be the one that had just been opened. But I wasn’t sure of any real difference between them.

IMG_6056I wasn’t very satisfied with the experiment for a couple of reasons. For one thing, the wine was pretty ordinary, and for another, everybody knew they were different and was looking for it. So the following Thanksgiving Day, 2015, I chose a Barolo, a wine which people often say needs to breathe, and I used the Sesame Street method: each person was given three glasses and had to pick the one that was not like the others. One bottle was decanted 6 hours in advance, the other opened immediately beforehand. One guest served as a lab assistant, and, in another room, randomly chose and poured one of 6 possible configurations for each guest: ddb, dbd, bdd, bbd, bdb, dbb (where b = bottle and d = decanter). Nobody knew which position the odd glass out was in or whether it came from the bottle or the decanter, and each person had an independently and randomly chosen configuration. Here were the results

Person Configuration Choice
B dbb 1
D dbd 3
R bbd 2
A dbb 3
R&P dbb 2
J ddb 2

Only one person correctly picked the odd glass out, and, as it turns out, that person was me. Here are my notes from that occasion:

I had a feeling one of the three was more aromatic and spicier, more tannic on the palate, whereas the other two had a richer flavor on the palate. I was not at all sure it wasn’t my imagination, however. As it turns out, I correctly picked the glass that was not like the other two, but mis-identified it coming from freshly opened bottle rather than the decanter.

The reason I identified the odd wine out as coming from the freshly opened bottle was a hypothesis that the mostly like effect of breathing was to dispel some of the more volatile components in the aroma. Clearly more research was needed. I didn’t do my next breathing experiment until a few weeks ago, which I will describe in next week’s post.

My first wine experiment

I started collecting wine seriously in 2012. Apart from enjoying the wine, I am interested in finding out for myself whether the things people say about wine—that it ages, that you should let it breathe, that you must keep it at a certain temperature—are true. Over the last few years I have conducted various experiments to find out. Here I will describe my very first experiment, conducted in the summer of 2014.

In 2014, I left a bottle of 2012 Cascina Chicco Barbera d’Alba Granera Alta out in the Arizona heat all summer (100º–110º F). I had another bottle of the same wine in my temperature controlled storage facility (about 60º F). In November I did a blind tasting with my wife and another couple. The abused bottle had a wrinkled label, so I wrapped the bottles in pink tissue paper; once the corks were removed you couldn’t tell the difference. (In the photo the abused bottle is on the right.) Since I was the one who removed the corks I asked my wife to go into another room and, on the toss of a coin, either swap the bottles or not. Each guest had two glasses and was served from the two bottles in the same order.

Amazingly, the wine left out in the sun was drinkable. One of my guests could not tell the difference between it and the properly stored wine, and all agreed that the difference was subtle. I found that the abused wine was fruitier and lacking in acid, with harsher tannins, and a spicier aroma. The properly stored wine had more structure. It also had a slightly funky mulchy element to the nose. It was a much better food wine because of the acid; the abused wine was insipid during the meal. But it was quite drinkable, and preferred by two of my guests who like a softer style of wine.


This summer I am conducting the same experiment with a white burgundy, which I will write up in the fall.