A bottle arrives with a label that has clearly had a hard life. Stained brown at the bottom, one corner lifted, the printing faded on one side. Most buyers read that as damage and mentally knock the price down, which is reasonable, and then read it as a warning about the wine inside, which is not.
The two things are almost unrelated, and in one respect the ruined label is good news.
Reading a bottle before you open it
Paper labels rot in humidity. A cellar that is properly damp — the kind of stone-floored, unheated space that keeps a wine alive for fifty years — will destroy the paper on the outside of a bottle over a few decades. Mould spots, lifted corners, illegible print. Meanwhile a bottle stored in a dry, warm apartment will keep a crisp, clean label right up until the moment you discover the wine is dead. Cosmetic label damage on an old bottle tells you it was in a real cellar. It is closer to evidence than to a defect.
What actually matters is on the other side of the glass, and there are three things worth learning to look at.
The first is fill level, sometimes called ullage. Wine evaporates through the cork very slowly, so the level drops with age; that is normal. What you are judging is whether it has dropped further than the bottle's age explains. For a fifty-year-old Bordeaux, a level in the shoulder is expected. A level down at mid-shoulder or lower on a twenty-year-old bottle says the seal failed, oxygen got in, and the wine has probably gone brown and flat.
The second is the capsule. Look for seepage — a sticky, dark residue under the foil or crusted around the lip. That is wine that has been pushed past the cork, and it is usually a sign the bottle spent time somewhere hot, because heat expands the liquid and forces it out. Seepage is a genuine warning where a torn label is not.
The third is provenance, and it is the one that cannot be seen at all. Where has this bottle actually been for thirty years? That question is worth more than every visual cue combined, and it is why a bottle with a scruffy label out of a known cellar is a better bet than a pristine one with no history.
All of which matters most for estates that are actually worth keeping for thirty years. There are not that many of them, and one of the most reliable sits in the south of Saint-Julien.
The estate built for the long version
Chateau Gruaud-Larose farms one large, unbroken block of around eighty hectares, which is unusual in a region where most properties are a patchwork of scattered parcels. It sits further inland than most of its neighbours, and the important detail is underneath: gravel on top, but clay not far below it.
Clay does two things. It holds water, so the vines do not shut down in a dry August. And it produces a wine with more tannin and more density than the free-draining gravel closer to the estuary. As we have set out in writing about the Bartons, Saint-Julien usually sits between Pauillac's firmness and Margaux's give. Gruaud-Larose leans hard towards the firm end and then keeps going. It is dense, tannic, sometimes a little rustic when young, and it takes a very long time.
The mistake people make is opening it at six or eight years, finding it hard and unyielding, and concluding it is austere. It is not austere. It is unfinished. This is a wine whose whole design assumes somebody will wait, which is exactly why so many old bottles of it still exist and still work.
What we have
The Gruaud-Larose range spans a wider stretch of time than almost anything else we carry, and it splits into three quite different purchases.
The drink-now tier is Sarget de Gruaud-Larose, the second wine, from younger vines and unselected lots. The 2018 is around HK$270 and the 2017 around HK$300.
Recent grand vin is the middle tier and it is the one to be careful with. The 2017, at around HK$570, is a perfectly good bottle that is nowhere near ready.
Then the mature end, which is where this estate earns its reputation. The 1994 is around HK$680, the 1997 around HK$650, and the 1996 around HK$980. Further back still there are pre-war and nineteenth-century bottles, some carrying exactly the cosmetic label damage discussed above.
What we would pick, and why
The 1996 is the bottle we would put on the table. It is the most expensive of the three mature wines and it is worth the gap: 1996 was a cool, late season that suited Cabernet Sauvignon on the northern Left Bank particularly well, so the wines came out firm, high in tannin and structured for the long haul. That is Gruaud-Larose's exact temperament, and at twenty-eight years old it has finally done the work.
The 1994 is the pick if you want the same experience for less, and you should know what you are trading. 1994 was interrupted by September rain, so the wines are leaner and less complete than 1996. What that means in practice is that it arrived at maturity sooner and has less left in reserve. Drink it this year rather than next.
The Sarget 2017 is the pick for volume, and there is nothing apologetic about that. It is a Saint-Julien second wine at around three hundred dollars, ready now, and it will do more for a dinner of ten people than one bottle of anything above it.
Who should not buy this, and where the money stops working
Do not buy young Gruaud-Larose unless you have somewhere cool to put it. This is the clearest case in our Bordeaux range of a wine that is a promise rather than a product. The 2017 grand vin needs another eight to ten years in decent conditions, and in an unairconditioned Hong Kong flat those ten years will not deliver what they would in Bordeaux. If you have no wine fridge, buy the mature bottles — somebody else has already done the hard part, and the price of that service is built into the difference.
Do not buy the very old bottles as a safe purchase. A wine from before the war is a known risk taken on purpose. Most of the time it is remarkable. Some of the time it is tired, and no amount of good provenance guarantees otherwise. That is the deal, and anybody who tells you otherwise is selling.
And past a certain point on this list you are buying rarity rather than flavour. A hundred-year-old bottle is a historical object that happens to be drinkable. If the goal is the best glass of Saint-Julien for the money, it is somewhere in the 1990s, not in the nineteenth century.