Today we did a blind tasting of ten current releases of Champagne with the winners of this year's GuildSomm online wine competition along with ten other bay area sommeliers. We chose the wines to be equally divided between widely known negociants and buzzworthy growers. Mid-price and high-end price points were equally represented for both.
We asked everyone to rank their preferences from 1-10. The results are below. There was lots of variation within the scores and many wines were one person's favorite and another person's least favorite. I personally preferred the Dom a bit more than the final tally and the Chiquet and Clicquot a touch less. Otherwise I am in total agreement with the results. (My top three were Comtes de Champagne, Heidsieck and Dom) The Bouchard and the Collin were the only two I personally disliked—not a surprise if you know me—but I spoke to one person who had them as their favorites. Taste is obviously subjective, but I do find it interesting that four of the top five were large houses and four of the five in the lower half were growers—three being well over $100. This has been the case every time I have done this exercise. Also the top two wines were $50 retail!
1. Charles Heidsieck, Brut, NV ($50)2. Gaston Chiquet Reserve, ‘Cuvée de Réserve,’ Brut, NV ($53)3/4 (tie). Taittinger, ‘Comtes de Champagne,’ Brut Blanc de Blancs, 2006 ($130)3/4 (tie). Veuve Clicquot, ‘La Grande Dame,’ Brut, 2006 ($160)5. Louis Roederer, Brut, NV ($45)6/7 (tie). Dom Perignon, Brut, 2004 ($155)6/7 (tie). Pierre Peters, ‘Cuvée de Réserve,’ Brut Blanc de Blancs, NV ($60)8. Ulysse Collin ‘Les Roises,’ Extra Brut Blanc de Blancs, 2012 ($120)9. Vilmart & Cie, ‘Coeur de Cuvée, Brut, 2007 ($150)10. Cedric Bouchard, ‘La Boloree,’ Pinot Blanc Roses de Jeanne, Brut Blanc de Blancs 2011 ($145)
It is certainly fair to point out the dosage is not the end all be all, and yes, there are wines that are so plush by nature that they don't need much at all. But I do think it is worth pointing out, not only Geoff's initial and undeniable recognition that big houses carried the day, but that, so did higher dosage wines. In fact, the one non-grower house that didn't make the top 5 had the lowest dosage of its peers, and the one grower that did make the top 5 had, if I recall, the highest dosage of its peers. That seems a far too significant data point to blow off.
Of course I would love to taste those 357 wines. Just for science, of course. And I don't mean to belabor this, but I do think you're being a bit obtuse. Sure "most" of the wines at the top were NV, but just 3/5. That's one wine away from being a minority. But the dosage levels are important for two reasons. 1) It wasn't "most" of the wines at the top with higher dosage, it was "all" of them. Here are the dosage levels in order of finish 11,8,9,8,11,6,6,1-2,6,0. On average, the wines at the top were 9.4 and those at the bottom were 3.9. The one with zero dosage finished last and the one with almost zero finished 8th. That is a rather clear statement. In fact, seemingly the clearest of all data points thus far brought up and looked into. 2) Low dosage is a polarizing factor. Anecdotally speaking, I can recall a number of geeky FB polls among wine types asking people's opinions on low dosage champagne. And when I look in the comments section, there are some charged opinions on both sides. I don't see this debate rage on about single grape vs blend or NV vs vintage. Sure, big house vs grower, but that's also been covered here and represents another significant data point. Not saying its the only factor. Like you said, perhaps the Bouchard was down there because it was flawed. But you, yourself dogged the Colin for being too "lean and mean", which is typically the complaint about low dosage bubbles. And it was one of only 2 wines that had nearly zero or zero dosage. The Dom finished last among its NM peers and, guess what? It had the lowest dosage of all of them. But, like you said, it was only 10 wines...
Charlie Deal I'm not being obtuse. First, it's not clear that the higher dosage wines won out absolutely. Vilmart CdC typically is 8-9 g/l dosage (according to Tyson Stelzer's Champagne Guide). Depending on the dosage of the Vilmart, it could have had a higher dosage than the Chiquet or the Veuve. Second, I believe that one needs to be careful from making conclusions from a single, very limited set of data. What you've got is a good hypothesis (and it's interesting), but that means you need to go out and do more testing, not try and extrapolate a conclusion. If you really wanted to make your case, we could do a trial with a bunch of somms, repeat the exercise on a larger sale and also slip in a few brut zero wines that we have dosed to different RS levels, and then run the results. Additionally, we didn't assign any value to scores, we just used rank order, which might not take into account how much one person preferred their wine 1 to their wine 10 when comparing to other people in the group. With that information you might have been able to tease out some relationships, but right now it just isn't enough. And for the analysis, the Bouchard should be discarded from the data set since it was regarded by the tasters as faulted and wouldn't be included. Through my own research, I have slightly different dosage levels (11,8,9,8,10-11,5-6,6-7,1.7,8-9) and if I averaged out the top five and the bottom four (assuming you'd count a 5-6 as 5.5, etc) you'd get 9.3 and 5.55 respectively for the dosage of said wines. That comes out to 3.75 g/L difference, which I personally feel isn't much and I'd wonder if we altered some wines by that amount whether we'd get a major change in the results. In the end, we had a room full of 20 geeky somms (not just a facebook poll open to all) that produced an interesting, but limited, set of data. It's something that I'd love to explore more (I'm actually now trying to figure out how to build a staff training for my servers based on this experience). If you're ever in Seattle, I'd love to pour you some bubbles and chat your ear off about it.