There is an old saying in the Medoc that the best vines are the ones that can see the river. It sounds like folklore invented for tourists. It is a description of two physical mechanisms, and once you understand them, a great deal of Bordeaux pricing starts to look less arbitrary.
The first is gravel. The Gironde spent a very long time depositing stones along its left bank, and those deposits sit in low mounds the locals call croupes. Gravel drains, so a vine planted on it is never sitting in wet soil, and the roots go down rather than sideways looking for water. It also absorbs heat all day and releases it overnight, which is the only reason Cabernet Sauvignon can finish ripening at this latitude. The same grape on the clay a few hundred metres further inland makes a thin, green, unpleasant wine. Nothing about the winemaking changes. The ground does.
The second is the water itself. A large body of water is slow to heat and slow to cool, so a vineyard beside the estuary swings less between day and night and is less likely to be hit by spring frost. Being able to see the river is a proxy for both: high enough on a gravel bank to drain, close enough to the water to be protected. It is not a poetic claim. It is a soil report.
What classification measured, and what it did not
Bordeaux has two famous lists. The 1855 classification ranked sixty-odd Medoc properties by the prices merchants had been paying, and it has been revised once in a hundred and seventy years. Below it sits the cru bourgeois system, a broader and more recent list that has been rewritten, abolished, litigated and reinstated more than once, and which works by application: an estate submits itself, samples and paperwork included, and a panel decides.
That word, application, is the part people skip. An unclassified property is not necessarily a property that failed. It may be a property that never entered. And a serious estate that declines to enter is making a specific argument: that its price in the market is a more honest measure of its wine than a committee's verdict, and that submitting samples for judgment invites being ranked below chateaux it outperforms.
Sociando-Mallet
Sociando-Mallet sits at Saint-Seurin-de-Cadourne, at the northern end of the Haut-Medoc, on a gravel rise directly above the estuary. It is, geologically, exactly the kind of site the saying is about, and its neighbours across the boundary are classified Saint-Estephe properties selling for considerably more. Jean Gautreau bought it in 1969 when it was close to derelict and spent the following decades rebuilding it, and he kept the estate out of the cru bourgeois classification on principle. The property has been famous ever since for the same thing: making wine at a level its label does not claim.
The style matches the owner. This is a firm, dark, structured wine with a lot of Cabernet in it, built to age rather than to charm. It is not a soft, forward Bordeaux and it does not try to be. Young vintages are tight and can seem austere on first pour. Give them air, or give them years.
What we have
There is essentially one wine here, which makes the Sociando-Mallet collection unusually easy to read. Sorted by price from the bottom, the list is really a list of ages. The lower rungs are vintages from the last fifteen years, where our stock is deepest, including Sociando-Mallet 2012 at around HK$298. The upper rungs are bottles from the nineties, which are a different proposition and priced on how few of them are left.
What we would pick, and why
Sociando-Mallet 2014 at around HK$248 is the bottle we would buy by the case. 2014 was a decent Left Bank year, the wine has had a decade to settle, and at this price it is doing something most Bordeaux at the same money cannot: tasting like a serious wine rather than a cheerful one. It needs half an hour in a decanter. Give it that and it does not taste like a HK$248 bottle.
Sociando-Mallet 2011 at around HK$288 is the one to put on a banquet table where several courses are coming and the wine has to hold its own against strong food. It is fourteen years old, savoury rather than fruity, and the tannin has turned from grip into frame. Firm Cabernet with age on it handles soy, braises and roast meats better than a soft, fruity red does, which is a genuinely practical point in Hong Kong and not a theoretical one.
If you want the deepest and most structured of the recent vintages, Sociando-Mallet 2013 at around HK$298 is worth a look for the opposite reason to the usual one. 2013 has a poor reputation across Bordeaux, and estates with good drainage suffered least in a wet year. This is one of them. It is the value bet on the list, though it is not the wine to open in a hurry.
Where paying more stops buying flavour
At the top sit bottles like Sociando-Mallet 1993 at around HK$748 and a 1997 near it. Both are thirty-odd years old, and both cost roughly three times the 2014. They are not three times the wine. What they are is a different experience: less fruit, more earth and dried leaf, a shorter window once the cork is out, and a real chance that a given bottle has faded. That risk is the purchase. If the year on the label means something to you, or you want to show someone what old Bordeaux actually tastes like, it is money well spent. If you want the best glass of wine on this page for the money, it is not.
Who should not buy Sociando-Mallet: anyone who wants to open a bottle and drink it immediately without thinking about it. This wine is tight on first pour and rewards a decanter, and if that is a chore rather than a pleasure, buy something rounder. And if you are keeping bottles in a Hong Kong flat without a wine cabinet, this is exactly the producer where that hurts most, because its whole argument is that the wine improves with time. Buy the vintages that are already there, or buy storage first.