I 'm going to show you data in a minute that's slightly terrifying," warned Mark Brown, the audacious president and CEO of the Sazerac Company, in a cozy conference room at the Buffalo Trace Distillery in Frankfort, Kentucky. He had summoned a few of us whiskey writers for the grand reveal of five consumer-selected finalists in the Single Oak Project, an ambitious four year program designed to help determine the best bourbon in the world. No biggie. The next day we would sample and rate those five favourites to establish the one that would ultimately be bottled as Single Oak Project, Kentucky Straight Bourbon Whiskey, but this first day was devoted to that terrifying data.
Four years earlier, I had been a part of a group of mostly the same writers when we gathered at the distillery for the launch of the SOP, an epic undertaking best explained as the Human Genome Project of whiskey. Its aim is to identify, map and isolate various factors in the bourbon making/aging process to determine what variables have which effect on the final product, all in the name of finding the "Holy Grail," as Brown puts it.
The SOP started in 1999 when Brown, distiller Harlen Wheatley and the late Ronnie Eddins, Buffalo Trace's longtime, influential warehouse manager, hand-selected 96 trees from Missouri's Ozarks. They had each tree split into top and bottom segments. Each half was used only to make one barrel, yielding 192 unique casks. That top/bottom factor then went into the mix with six other variables: recipe (wheat or rye); entry proof (105 proof or 125 proof); stave seasoning (six months or 12 months); grain size (tight, average, or coarse grains); warehouse (concrete floor or wooden rick floor); and char level (char #3 or #4).
For four years, the 192 barrels were released twelve at a time every three months. When compared against each other, those 192 bourbons would give you 1,396 flavor combinations. Buyers registered on the SOP website to rate the ones they sampled. The variables were unknown. A total of 5,645 people joined the site and ultimately logged 5,086 reviews. Each whiskey was reviewed 26.5 times, on average.
And that input could hold the key - well, one of the keys - to decoding the genetic makeup of the ultimate bourbon. (Here's where the slightly terrifying data comes in.) According to the dizzying array of figures culled from the online assessments, whiskeys aged in barrels made with wood from the bottom half of trees with high char levels were rated the highest when evaluated by color. The top aroma scores were awarded to whiskeys made with the wheat recipe and aged in barrels from the bottom half of the tree. Where flavor is concerned, ratings were all over the map, but the most consistent variable was whiskies from bottom-half barrels. There was no noticeable differences based on warehouse style.
The whiskey that came out on top with our judging panel, which included legendary bartender Gary Regan, Spirit Journal's acclaimed publisher/editor Paul Pacult, and bourbon authority Chuck Cowdery, was a rye recipe bourbon aged in a bottom-half barrel with average wood grain size and level 4 charring. Staves were seasoned 12 months and entry proof was 125. By science or by fluke, that yet-unheard-of combination resulted in a complex, balanced spirit with pronounced notes of fruity spice, a hefty mouthfeel and a clean finish. It will be released in eight years.
But the SOP is just one of a multitude of experiments being carried out in pursuit of the Holy Grail. There have been 35 released to date, including the Experimental Collection series; many more are in the works. A number come thanks to the company's long-running research that's led to the identification of 300 chemicals present in the wood and distillate. With that information, Wheatley explained, they're able to create a chart that illustrates what each whiskey looks like to a chemist. That data, believe it or not, actually translates into quantifiable flavor profiles and can actually be plotted on a grid. A number of other experiments are possible because of the new Warehouse X, a cutting-edge building designed with five independently operating chambers that allow for the controlled manipulation of natural and UV light, temperature, humidity and airflow.
This elaborate research is, of course, huge in scale. Add to that the ever-increasing demand for the distillery's familiar brands and inevitably space will become a problem. At present, there are 59 warehouses (seven years ago there were 12) with 346 aging floors (once there were 60). Wheatley estimates there are about 1000 points of data in each warehouse. Today there's the capacity to store 1.5 million barrels. At the going rate of production (they can fill 800 barrels in eight hours), the distillery will be out of storage space by 2017. Based on a 2047 forecast, they'd need a new barrel warehouse every five months for ten years to accommodate the mass volumes, beginning in 2017.
To prepare for that escalation, Buffalo Trace purchased 250-plus acres adjacent to the distillery. They're already growing their own grains there—rye, barley, Boone County White Corn. The second day, Brown and Wheatley drove us through the property where grass grows waist-high. Brown pointed into the expansive distance and asked us to envision the 50 barrel warehouses that would be built over the next 21 years. What he said next suddenly justified all his research.
"I don't think anyone understands warehouse design. They started being built in the 1800s and everyone just kept going that way. Why not push the envelope?" he posed. "We're beginning to understand warehouse profiles better. If, for instance, you want a higher proof, one warehouse is better than another. Now we're about to spend $200 million building new warehouses and we don't know why things like this are occurring."
While Brown was quick to note that the Holy Grail is still a long way off, the SOP does offer some "tantalizing clues." But it also poses thorny questions, like: if taste and smell are subjective, can there ever be a universally acceptable supreme whiskey?
According to Monty Python legend, the Holy Grail possesses miraculous powers. Maybe it's worth considering: is that miracle a quantifiable measure of greatness? Or an exciting, dramatic chase that generates conversation and debate and pushes the industry forward? Maybe we need to ask: Is the Holy Grail a product? Or a process?