Wine Colour Chart: Read a Wine's Age From the Glass
Updated
A wine colour chart has five rungs, not fifty, and the two bodies that train professional tasters both use theirs to reach an age range rather than a year.
Red wine runs through purple, ruby, garnet, tawny, brown and white wine through lemon-green, lemon, gold, amber, brown, according to WSET's Level 3 Systematic Approach to Tasting. The Court of Master Sommeliers Europe grid is shorter. It lists "White: Straw, Yellow, Gold, Amber" and "Red: Purple, Ruby, Garnet", then keeps a separate hue line: "Silver, Green, Orange, Purple, Ruby, Garnet, Brown". Reds lose colour as they age and travel towards orange and brick. Whites gain it and travel towards gold and amber. Both WSET scales end at brown. The direction is dependable. The speed is not, which is why the Court's grid asks for nothing narrower than "1-3 yrs, 3-5 yrs, 5-10 yrs, 10 yrs+" and leaves the top band open. Grape variety, pH, storage temperature and rot in the vineyard each move the colour clock independently of the calendar.
What is a wine colour chart, and which one do professionals use?
It is a named ladder of hues, one for whites and one for reds, used to place a wine before you smell it. Three published versions matter, and they disagree in ways that tell you what each band is for.
WSET Level 3 assesses appearance in three fields: clarity, "clear – hazy (faulty?)"; intensity, "pale – medium – deep"; and colour. It then asks for "other observations", listed as "legs/tears, deposit, pétillance, bubbles".
At Diploma level the same organisation shortens the lists. White becomes "lemon – gold – amber – brown" and red becomes "purple – ruby – garnet – tawny". Lemon-green disappears from the white scale, and brown disappears from the red one.
That second edit is the interesting one. A Diploma taster has no box for a brown red wine, though brown survives on the Diploma white scale. Read it as WSET's judgement about which observations are still worth a colour term at that level.
The Court of Master Sommeliers splits the job differently again. Colour is the core reading. Hue is a second, separate line, and it is the line that carries silver, green, orange and brown. So a taster can record a garnet core with an orange hue at the edge. That combination is the whole point of the exercise: it says the wine is further along than its centre admits.
The red wine colour chart, band by band
Purple, ruby, garnet, tawny, brown, in that order. Read each rung as a chemical state rather than a paint swatch, because every one of them is a stage in one long reaction between pigment and tannin.
| Band | Where it sits on the published scales | What is happening in the glass |
|---|---|---|
| Purple | First on WSET's red list, and on both the colour and hue lines of the Court's grid | Free monomeric anthocyanins dominate. Copigmentation alone "can contribute about 30–50% of the color in young red wines", according to He and colleagues, writing in Molecules in 2012 |
| Ruby | Second on both scales | Polymerisation is under way. In their companion paper the same authors put it like this: about 25% anthocyanins may have polymerised with flavonoid compounds by the end of alcohol fermentation, and this level will rise to more 40% after one year's aging |
| Garnet | Third on both. The last colour term the Court's grid offers for red | Monomeric pigment keeps falling. Anthocyanin concentration in Kotsifali and Mandilari wines dropped "threefold over the nine-month period of maturation", according to Basalekou and colleagues |
| Tawny | Fourth on WSET. Absent from the Court's colour line, which hands the job to the hue field | Pyranoanthocyanins take over. Most pyranoanthocyanins possess yellow to orange colour and contribute to the tawny colour shift associated with red wine aging (He et al.) |
| Brown | Last on WSET Level 3. Dropped from the Diploma red list | Oxidation has run to its end. The Court's grid keeps brown only on the hue line and never offers it as a colour |
The laboratory version of this ladder is a single ratio. Basalekou's team defines hue as absorbance at 420 nm divided by absorbance at 520 nm, so as yellow-brown climbs and red falls, one number rises. They describe the outcome in the same words a taster would use: "the purple hue of young red wines is replaced by brickish red to tawny red hues". Zhang and colleagues, tracking Marselan and Merlot through a year in bottle in Foods in 2024, found the same thing in two single-variety wines: colour chroma "decreases continuously during aging, and the hue shifts from red to orange".
The white wine colour chart, band by band
Lemon-green, lemon, gold, amber, brown. Whites run the ladder in the opposite direction to reds: a red wine spends its life giving colour up, a white wine spends its life acquiring it.
| Band | WSET Level 3 | WSET Diploma | Court of Master Sommeliers |
|---|---|---|---|
| Lemon-green | lemon-green | dropped | green appears on the hue line, not the colour line |
| Lemon / straw / yellow | lemon | lemon | straw, yellow |
| Gold | gold | gold | gold |
| Amber | amber | amber | amber |
| Brown | brown | brown | brown appears on the hue line |
The instrument reading here is absorbance at 420 nm on its own. Ricci, Parpinello and Versari followed six commercial white wines for 10 months after bottling at 20 ± 1 °C. They found that optical density at 420 nm, their index of browning, showed a progressive increase that fitted the zero-order kinetic rate: steady, not sudden. Across those six wines they put apparent activation energy for browning between 46.8 and 88.3 kJ per mole, which is the formal way of saying the browning rate climbs sharply with storage temperature.
That is why gold on a white wine is an ambiguous signal until you know how the bottle was kept. It can be ten years of correct cellaring or one bad summer in a courier's van.
It is also why the published ladders will not settle a sweet wine for you. Neither WSET nor the Court prints a separate colour scale for sweet whites, so a gold reading on a Sauternes 2001 lands in the same box as a gold reading on a dry white of the same year, and the two do not mean the same thing.
How do you tell wine age by colour in practice?
Tilt the glass away from you over something white, and read the edge before the middle.
The Court lists a field for precisely this, Rim Variation, defined as "Colour change from centre to edge". A young red is one colour from centre to rim. An older one holds garnet in the core and goes orange, then watery, in the last few millimetres.
Three more fields on the same grid do work that colour alone cannot:
- Concentration, recorded as "Pale/Translucent, Medium, Deep, Opaque". This is depth, not hue, and it is mostly a variety and vintage reading rather than an age one.
- Brightness, recorded as "Dull, Bright, Brilliant". Dullness on an otherwise correct colour is a warning that arrives before the hue does.
- Extract/Stain, recorded as "None, Light, Medium, Heavy". How much pigment the glass keeps when the wine runs back down.
Light matters more than most people allow. The traditional tool for the job was the tastevin, "a shallow cup allowing one to see the color of the liquid in the dim light of a cellar". The shallow dish and the dimpled silver existed because a cellar candle is a poor place to judge a hue. A phone torch and a white napkin solve the same problem.
And then you stop. The Court's grid never asks for a vintage from sight. Age Range, "1-3 yrs, 3-5 yrs, 5-10 yrs, 10 yrs+", sits in the initial conclusion, after nose and palate. A vintage is only asked for at the final conclusion.
Why does the colour change at all?
Because the pigment that makes a young red wine red is unstable, and most of it is invisible even when it is there.
Start with the pH. He and colleagues report that "at red wine pH (3.3–3.5), the equilibrium is largely towards the hemiketal state, which is colorless", and put numbers on it: "at pH of 3.4–3.6, 20–25% of anthocyanins are in the colored flavylium forms, whereas at pH of 4.0, only 10% of anthocyanins are in such ionized state". A fifth of the pigment does the visible work. The rest is sitting in a colourless form waiting.
Then the reactions. Anthocyanins condense with tannins, directly or through acetaldehyde, and react with pyruvic acid and other partners to form new, more stable pigments. Joana Oliveira, writing for IVES Open Science in 2022, describes the result plainly: "a color change is observed from red/violet to a more brick hue due to the formation of several anthocyanin-derivatives such as A and B-type vitisins and other pyranoanthocyanins".
The new pigments are more stable and less red. That is the whole story of a red wine's appearance over thirty years, and it explains why colour loss and colour change are the same event rather than two.
For whites the driver is oxygen rather than condensation, and the visible product is the same brown that ends the red scale.
What throws the colour reading off?
Four things, and three of them have nothing to do with time.
Grape variety. He and colleagues note that not all the V. vinifera red varieties, such as 'Pinot noir' and red-colored mutants of the white grape varieties, contain acylated anthocyanins, and acylated forms are the more stable ones. They also record malvidin-3-O-glucoside and its derivatives varying "from more than 90% in Grenache to just less than 50% in Sangiovese". Pale is a fact about the grape before it is a fact about the year, which is why Romanée-Conti can look a decade older than a Merlot-dominated Pétrus of the same vintage and be nothing of the kind. Put a Bordeaux 2015 next to a Burgundy 2015 and the gap is obvious in the glass and meaningless as evidence.
Storage temperature. The Australian Wine Research Institute reports a recommendation to keep bottled wine in a cool 15 to 20 °C location in its storage guidance, and warns that "any storage place where the temperature exceeds 25°C for long periods and 40°C for short periods can affect wine quality". The same page cites Robert Parker's buyer's guide estimate that "somewhere between 10 and 25% of wines sold in the USA have been damaged due to exposure to extreme heat". That figure is Parker's from 2008, so treat it as an order of magnitude rather than a current market reading. A heat-damaged bottle reads old on the colour chart because, chemically, it is.
Rot in the vineyard. Laccase, the enzyme released by Botrytis cinerea, oxidises pigment directly. Its presence implies "the deterioration of the red colour" and, at worst, "the precipitation of the coloring matter (oxidasic haze)", according to Giménez and colleagues. They found the three anthocyanins carrying three substituents on the B-ring, petunidin, delphinidin and malvidin, "were degraded much faster" than the two-substituent pair, cyanidin and peonidin, and reported peonidin "is even not degraded by laccase". Red fruit picked with botrytis in it can therefore reach bottle already short of the pigment its age would suggest.
Bottle condition. A low fill and a compromised cork accelerate everything above. If the colour and the fill disagree with the label, read the ullage scale before you read the hue.
What do wine legs mean?
Alcohol, and only alcohol. Nothing about quality, sweetness or age.
The physics has been settled since 1855, when the physicist James Thomson gave the first correct explanation of what is now called the Marangoni effect. Alcohol evaporates faster than water and lowers surface tension, so the thin film climbing the side of the glass loses alcohol, gains surface tension, and pulls more liquid up behind it until droplets form and fall. The guidance is blunt about the folklore: It is sometimes claimed incorrectly that wine with 'lots of legs' is sweeter or of a better quality. In fact the intensity of this phenomenon depends only on alcohol content.
The demonstration is free, and the same entry gives you the method: the effect "can be eliminated completely by covering the wine glass (which stops the evaporation of the alcohol)".
Both professional grids record legs, and neither draws an age conclusion from them. WSET files "legs/tears" under other observations. The Court lists Viscosity/Tears on a "Low, Med-, Medium, Med+, High" scale, then asks its initial conclusion for climate, grape variety, possible countries and age range. Alcohol is fixed at bottling and does not change on the rack, so tears can inform the climate guess. They cannot inform the age one.
From a colour band to a drink date
Colour gets you to a band. What you want is a date, and a band is a decade wide at the top end.
That is the gap the maturity data closes. We hold colour bands against per-vintage maturity for Bordeaux, Burgundy and Sauternes, so a wine that looks a rung ahead of its calendar age can be checked against where its cohort actually sits rather than against a generic ladder.
Get the maturity charts before you pull the cork
The useful version of this skill is knowing which bottle in your cellar is about to peak, and which one you have twelve years to think about. Our vintage maturity charts put every wine on that timeline, so the colour in the glass becomes a check on the data rather than the only evidence you have.
Join the newsletter and we will send the maturity charts as they update, vintage by vintage, with the wines whose windows are about to open.
Then use drink-now for the bottles already in your cellar, or the Burgundy producer atlas if the wine in the glass is pale for reasons that start in the vineyard rather than in the bottle.