They do a lot of experimenting at Buffalo Trace. So much so that they installed a separate micro-distillery about six years ago and recently built warehouse X, which allows them to vary environmental factors during the aging process. They’ve been releasing the results of some of their test runs under the Experimental Collection label since 2006, with many of those bottlings having origins dating back to the mid 1990’s. But their boldest such undertaking began in 1999, when they started the Single Oak Project.
This is a series of 192 single barrel bourbon bottlings which are differentiated by seven variables. Six of the variables have two options each, and the seventh has three, giving 192 possible combinations. First, 96 White Oak trees were selected from the Missouri Ozarks. The selected trees were categorized by their grain structure and split into three groups; 1/3 of them having course grain, 1/3 of them having average grain and 1/3 of them having tight grain. Next, each tree was cut in half, giving a top and a bottom. As these tree sections were milled into staves, they were tagged and tracked so that each barrel could be made entirely from an individual tree section.
With the first two variables down, they had six groups of 32 barrels to which they could apply the other five variables. First, the staves were air seasoned for either 6 months or 12 months. Then the barrels were given either a #3 char or a #4 char. They were then filled with either rye recipe bourbon or wheated bourbon, with a barrel entry proof of either 105 or 125. Finally, they were either aged in Warehouse K (with wooden floors) or Warehouse L (with concrete floors).
The significance of some of these factors is less obvious than others. Apparently a tree’s chemistry varies along its height. The lower half has a higher concentration of lignins and the top half has a higher concentration of tannins. This is a really interesting variable because it is not normally an option offered by the cooperages that supply the barrels, whereas char level, grain structure and stave seasoning time are. As for warehouse style, both K and L are constructed with brick walls. But the wooden floors of warehouse K allow for more airflow, resulting in bigger temperature differences through its nine floors and from season to season. Conversely, the concrete floors of warehouse L act to moderate the temperatures, keeping them more consistent through its five floors and from season to season.
Once filled, the barrels were all aged for 8 years, then they were bottled at 90 proof. They are offered only in the 375ml format, with a suggested retail price of $47. Rather than putting them all out at once, Buffalo Trace is releasing them over the course of four years, 12 barrels at a time, every three months. Those releases started in the spring of 2011 and now just two remain to be seen, with the last one scheduled for winter 2015.
The whole purpose of this endeavor is feedback. While samples of every bottling have been sent to a select number of whiskey journalists (Drinkhacker and Scotch & Ice Cream consistently review each release), they are primarily relying on reviews posted by consumers to the Single Oak Project website.
There does seem to be a segment of the bourbon community that loves to hate Buffalo Trace, which I believe to be a vocal minority. While I’m certainly not part of that group, I do have some criticisms of this project. Don’t get me wrong though; I love the concept of this grand experiment, I just have some issues with manner in which it was implemented.
First is the conveyance of information, which is somewhat haphazard. I recently had a similar complaint about the E.H. Taylor Jr. collection. I understand that they are looking for unbiased reviews of these bottlings, so it makes sense that you are only able to access the “DNA” profile for a barrel number after you complete a review of it. At the same time it is a little annoying that I can’t view any of the information about the other barrels when I have no intention of buying more Single Oak Project bottles.
Also, the info put out on the website and in press releases isn’t as thorough as it ought to be. I had to do quite a bit of digging to find blogs with official statements describing the difference between the two warehouses and the top and bottom halves of the trees. Some sources report the grain structure for each barrel as tight, average or course, and others simply list the number of rings per inch. It took me quite a while to determine that they were calling 11 to 14 rings/inch average, 10 or less course and 15 or more tight. As far as I can tell the range goes from 7 to 21 rings per inch.
It looks like all of the barrels from warehouse K were on its seventh floor, and all of the barrels from warehouse L were on its third floor, but that information was not clearly stated on the website, as it should have been. There are also a few additional variables that are being tracked, but they are not discussed anywhere. The barrels are listed as coming from trees harvested from location A or location B, but no further information is provided. Also, the number of staves in each barrel is listed, and as far as I can tell they range from 30 to 65, but this is another topic that is not expanded upon.
The other issue I have is format and price. There was a lot of buzz when the series was first announced, but that has faded quite a bit over the last three years. It’s an experiment; there are some great barrels and some duds, but most of them are just reasonably good. Of course, reviews vary; everyone has their personal preferences. That being said, I think a lot of consumers are wary of taking a gamble of a ½ bottle at nearly $50, especially when it will most likely be no better than “good”. Even if someone was enthusiastic enough about this series and gave it their full backing, I can’t imagine anyone spending nearly $10,000 to try every bottle.
The bottle that I have was from the second release, so it’s been out for about three years. Admittedly, I’m a little late to the game, but I would think all of the reviews for this barrel should be in by now. I only count 37 of them for barrel number 63. Assuming a yield of 400 bottles (375ml) from a barrel, that’s a participation rate of about 10%. I think they would have been much better off with 200ml bottles at $20 apiece. Or they could have split each barrel into 375ml bottles and 100ml bottles. With the larger bottles priced at $40 and the smaller bottles packaged in groups of 6 for $60 to $65, I think they would have gotten a much better response and a lot more feedback.
As for the reviews, I’m sure it wasn’t easy for the folks at Buffalo Trace to come up with a universal format that most people would be comfortable using. They had six categories with multiple choice selections and each was followed by a 1 – 10 rating choice. Most professional reviews (Whisky Advocate, etc) use a 100 point scale, but only 40 points or so are actually used – the worst of the worst never really score below 60. I tried to give scores (something I don’t usually do in general) that considered the full range of the scale. Whether or not others did that could be a bit of a flaw in the system. I also found some oddities in the questions. Color is something I observe and am aware of, but I’ve never considered assigning a quality rating to it. “Dry” and “thin” were options for mouthfeel, but I usually associate those terms with flavor (or lack thereof). The “overall” category presented an odd mix of flavors and sensations as options, it took me a while to wrap my head around that one.
As for the bottle that I have, I’ll list its “DNA” first, followed by the review I provided to the Single Oak Project website (underscored words were additional to the multiple choice options), and finally an overall assessment.
Age – 8 years
Entry proof – 105
Recipe – wheated
Barrel char - #4
Stave curing time – 12 months
Tree section – top half
Grain structure – tight, 17 rings/inch
Number of staves – 42
Harvest Location – A
Warehouse location – L / 3 / 27 (warehouse / floor / rick)
Color – Copper
Color rating – 7
Aromas – Butter, Cinnamon, Nutmeg, Bread
Aroma rating – 7
Mouthfeel – Creamy, Hot
Mouthfeel rating – 6
Flavors – Caramel, Cinnamon, Leather, Mint, Nutmeg, Molasses
Flavor rating – 8
Finish – Dry, Earthy, Spicy, Long, Hot, Astringent
Finish rating – 7
Overall – Chewy, Dry, Complex, Spicy, Tannic
Overall rating – 7
Full bodied. Starts off a little thin (flavor-wise) up front, but picks up quickly. Good complexity but not all that well integrated. It goes back and forth between its pros and cons (big and spicy vs. astringent and acidic) as it moves from the mid-palate through the finish.
This project is already an ambitious undertaking. I realize that adding one more variable would double the size of it, but I think they could have learned so much more if age was a factor as well. It’s generally accepted that some warehouse locations are better for long term aging, while others produce better quality young whiskey. But what about factors like barrel char level, entry proof and grain structure? Can those be varied to suit whiskey that is destined to be bottled at a particular age? If they had doubled up and made two barrels from each tree section, they could have aged them out to 6 years and 10 years. That would have made this the ultimate bourbon experiment in my opinion.