From jancisrobinson.com |Enologix - same as it ever was in Napa
Written by Alder Yarrow for jancisrobinson.com
Fifteen years ago, at the height of wine critic Robert Parker's influence and fame, WIRED magazine wrote an article about Enologix, an obscure company in northern California started by a man named Leo McCloskey who audaciously claimed to be able to use statistical analysis to predict wine scores based purely on the chemistry of the finished wine. Enologix had already been in business for six years prior to the article's publication, but the coverage in WIRED rocketed Enologix to controversial prominence in California wine.
As the landscape of wine criticism has broadened and diversified in the past 15 years, with Robert Parker seemingly headed for retirement and other critical outlets gaining in number and strength, it seemed like a good time to find out what Enologix was up to, and whether it was registering any of the changes sweeping through American wine culture.
Some time during the year 2001, an industry acquaintance handed the WIRED article to winemaker Sam Spencer. Spencer, who had his own label at the time, went on to serve as winemaker for California négociant Cameron Hughes and most recently worked for Head High wines, a brand in private-equity mogul Bill Price's portfolio of wine holdings.
'I read the Xeroxed copy of this article, and thought the whole concept was so interesting', says Spencer. 'I was intrigued. And honestly, back in 2000 I had delivered wines to Parker that got panned and I didn't understand why. I knew if I wanted to succeed I needed someone to help me.'
'When I started talking to Leo, all I wanted was a solution', recalls Spencer. 'I didn't really grasp the depth of what he offered, and he didn't really sell himself that well. He offered a different set of metrics that were related to balance and perception of wine in a way that winemakers weren't talking about, and I realised this shit was for real. I saw something I knew I needed and went after it.'
Over the subsequent five years, Spencer-Roloson wines would receive several scores over 90 points from Robert Parker's Wine Advocate.
'What Leo initially brought to my winemaking was a discussion about tannins and complex anthocyanins', says Spencer. 'Leo and his team walked me through the ratios and the non-linear interactions between flavonoids and I learned a different model for winemaking.'
Most people are far more comfortable using the phrase 'the art of winemaking' rather than discussing 'the statistical correlations between bound and unbound flavonoids and perceived pleasure', but if you ask McCloskey, wine is as easily described by numbers as baseball. In fact, one of his favourite metaphors for what Enologix does invokes Moneyball, the book (and subsequent movie) about how the Oakland A's applied rigorous statistical methods to assemble the highest-performing team in baseball at the lowest relative cost.
McCloskey, along with his business partner and wife Dr Susanne Arrhenius, studied chemical ecology at the University of California where they both received doctoral degrees.
'My training is in large systematic models', says McCloskey. 'As a chemical ecologist you're aware of the largest factors influencing any ecological model — and grapes are just an ecological model.' McCloskey and Arrhenius, along with mathematician Dr Marshall Sylvan, established what they called the 'first mathematical model for California luxury wine market performance.' This model was built in part, says McCloskey, on an analysis of the Bordeaux first growths, which, McCloskey likes to point out, consistently produce the largest volumes of top-rated wines in the world.
According to Enologix's recent newsletter, 'Our models can deduce the formula that estimates the performance of luxury wine quality, prices, volume and revenues. The formulas led to a boom in scores and legitimised colour and flavour testing.'
That final claim seems hard to dispute. 'Leo brought something to the market that really mattered', says Spencer. 'Net-net, this is the way wines are being made these days. Everyone is talking about phenolics now, but when Leo was getting started, he was really the only one.' Bruce Cakebread, president of Cakebread Cellars in Napa, agrees that what was once radical is now commonplace. 'Twenty years ago we didn't have this kind of information to aid in winemaking', he says. 'Now pretty much everyone around the world is measuring tannins and complex anthocyanins.'
Cakebread says Enologix's ability to correlate wine characteristics to critics' scores plays little or no role in Cakebread's winemaking regimen. 'In the early days of Enologix, talking about scores was a way to get their foot in the door and explain their product,' says Cakebread. 'Everyone has moved on from that. Now it's really about understanding vintage-wide trends and figuring out how to adjust our fermentations based on what we're seeing come in from the vineyards.'
Spencer agrees. 'If you become an intuitive user of this data, it's not about dialing in a specific score, it's about being able to push and pull your fermentation the way that you would use bracketing in photography', he says, referencing the process of taking multiple pictures at slightly different exposures to achieve the optimal image.
A quick poll of a number of small producers around California, however, found few winemakers measuring phenolics during fermentation. Larry Schaffer of Tercero Wines in Santa Barbara County notes, 'the only things I look at are Brix, pH, total acidity and temperature'. He goes on to say many of his friends also evaluate nitrogen levels and volatile acidity, among other things, but phenolics still aren't a common point of analysis for many wineries.
McCloskey treats his methods as closely held intellectual property, offering very little to outsiders interested in validating his approach. He claims to have analysis data on more than 150,000 wines at this point, and his approach to wine quality involves correlating several dozen chemical compounds such as tannins, terpenes, phenols, anthocyanins and other flavonoids with key variables in the winemaking process such as temperature, extraction and colour density. By crunching the numbers for dozens of wineries every vintage, including some extremely large players such as Joseph Phelps, Caymus, Plumpjack and Cakebread, Enologix gathers a wide representation of the fruit quality from around Napa, allowing clear benchmarks to be established to guide winemakers in their harvest and fermentation decisions.
'We have gone from being a controversial company to being apple pie', chuckles McCloskey, who proudly boasts of a 98% client retention rate, before moving on to explain that the company's new focus involves extending his statistics into the realm of farming, correlating elements such as row direction and water availability with top scores from the major publications.
While McCloskey's methods include criteria and parameters for white wines and lightly coloured reds, it's clear that his focus has always been Cabernet Sauvignon and the Bordeaux varieties, and that his methods are tuned to producing big, velvety reds. What's more, it appears that these methods are most powerful at a certain level of scale, where winemaking decisions involve multiple fruit sources and many different fermenters.
'You need scale for things to have an upside', says McCloskey. 'Two thousand cases is the minimum.' That isn't to say that McCloskey's analysis doesn't work on smaller quantities of wine, but may suggest the more business-focused mindset that McCloskey shares with his customers, namely that the winemaking game is about making the largest quantity of high quality wine you can.
'It's about price, volume, and quality, and managing risk in that equation', says McCloskey.
'Look at [Château] Lafite. They bet the farm – literally all 225 acres [91 ha] – on a single wine, or a single wine and a second wine. And yet they're outselling every other wine in the world. That's the model. It's much more powerful. I think it's all headed in that direction. One batch of wine is going to drive the profits.'
One look at the statistics quoted in the most recent Enologix newsletter is enough to place McCloskey's perspective clearly in context. According to this newsletter, Enologix customers comprised 10% of all 2012 Napa Cabernet Sauvignons rated 95 points and above in the Wine Advocate, but accounted for 25% of the volume. The average batch size for these high-scoring wines was approximately 4,000 cases. According to McCloskey, his client base over the years includes roughly 220 different wineries, and he has about 40 active customers each year.
By the time I finished my conversations with McCloskey and some of his customers, I found the Talking Heads lyric, 'Same as it ever was, same as it ever was', rolling around in my head. When it comes to Napa Cabernet, things look much as they have for more than a decade, with prices and scores headed upwards each year. And as far as Leo McCloskey is concerned, that's been the plan all along.