Matteo, that's both hilarious and dead-on accurate, re: Violette.
Periodically one reads the buy the back label article. The one aimed at newbies that says essentially you may know nothing about an area or subject but these are importers you cam trust. If you see one of these back labels on a bottle, well, chances are that it is a good bottle.
That's not entirely right. The newbie article, which of course I've written more than once, doesn't claim "a good bottle," but rather "a bottle with an above-average stylistic predictability," which you do address in your following paragraph. It's really not any different than suggesting to the consumer that buying by producer rather than appellation in Burgundy will increase one's chances of replicating previous successes and failures. I don't see that as particularly controversial, frankly. The articles aren't operating at the "yes, but Dugat made a suprisingly successful XYZ in 2004" level, they're addressing an "I don't know any of these wines, but I want a way to improve on random selection" audience.
I also think the stylistic predictability of importers varies. Sometimes, it's more of a philosophical predictability than it is an organoleptic one. For example, I can see certain importers eschewing a producer whose wines taste very much like other wines in their portfolio because they're made with inoculated yeast. And I also think that geographical differences are difficult to account for; how do you compare Dressner's Loire with Solomon's sunbelt? It's not like either intrudes much on the other's territory, and yet both are importers than I think the articles you're referencing would suggest have more identifiable house styles than others.
But on to the actual question...
Your blind tasting is an interesting idea and I'm definitely going to try it at some point. But the test you're proposing doesn't really say as much that's useful about the back-label assessment as you might hope. First, you're still tasting as much or more for producer as you are for importer...which is just replicating the function of the importer. There's not really a way to get around the fact that the Occhipinti and the Donnafugata are made differently in both intent and practice from moment one, and the points of difference are due to the producers, not the importers. And if you line up a dozen Barolos from a half-dozen importers, at (say) two wines from each producer, noting that the Brovias are different from the Rivettis is probably not all that difficult for an experienced taster of Barolo.
I think that to get a better read on this, you'd have to come close to a portfolio-wide tasting, recognizing that you're still going to have to figure out a way to usefully compare a bunch of Muscadets with a bunch of Australian chardonnays when contrasting importers that don't cross appellation paths. (And yes, I think you have to try if you're trying to prove or disprove the notion that Lynch's wines have different stylistic boundaries than the Grateful Palate's wines.) What would be
more useful is to find multiply-sourced wines that are made differently. Gaillard's Rose Poupre with the Rozet, for example, though in the past the former wasn't actually imported by anyone, and that might still be true. (Even then, you're really just tasting for winemaking differences without a really broad sample.) It might be easier with some of the Lynch wines, wherever he shares coverage with anyone. Or, at the least, not just a dozen Barolos, but paired Barolos from the same sites but different producer/importer duos. You're still tasting both producer and importer, but at least you're increasing the amount of controlled data, which I think is the major hurdle here...the other being, as previously noted, that Bourgueil vs. Priorat vs. Barossa is not the simplest comparison in the world.
