Does a More Expensive Red Wine Actually Taste Better?

In 2008 six economists ran seventeen blind tastings across the United States. Five hundred and six people tasted 523 wines — from a $1.65 bottle to a $150 one — without seeing a single label or price. Six thousand one hundred and seventy-five ratings came back. When the researchers checked whether the expensive wines had been rated higher, they found the correlation ran the wrong way.

Not “weakly positive.” Negative.

That result has been quoted ever since as proof that wine pricing is theater. It isn’t quite — and the reason why is more interesting than either camp usually admits.

The belief

“You get what you pay for.” Spend more on a bottle, get a better bottle. To test it, this analysis asks three falsifiable questions. If the belief holds: blind tasters should rate expensive wine higher; professional scores should rise steeply and reliably with price; and a cheap wine should rarely beat an expensive one. If all three fail, the belief is dead. If they split, the belief is conditional — and the condition is the finding.

Thesis: the price tag is telling you something

Start with the strongest case for paying more, because it is stronger than the cynics allow.

Across the entire reviewed wine market, price and score move together — hard.

That is not a rounding error. Sorted into price bands, the pattern is a clean, monotonic staircase: every step up in price buys a higher average score and a much better shot at the 90-point mark that drives shelf placement and sales.

The meta-analytic literature agrees. Pooling more than 180 hedonic price models built over twenty years and across many countries, Oczkowski and Doucouliagos found a partial correlation of +0.30 between what a wine costs and how it is rated — moderate, consistent, and with no evidence of publication bias.

And at the top of the market, the premium is real in the other direction too: a study of 266,301 bottles sold in the US found that the price premium attached to quality is statistically significant only for wines scoring above 90 points. Below that line, a 50-point wine and an 89-point wine are not reliably priced differently at all.

So the believer’s case is not superstition. In the market as it actually operates, price carries information.

Antithesis: take away the label and it collapses

Now remove the price tag.

In the blind data, ordinary drinkers rated the more expensive wine slightly worse. A tenfold increase in price bought roughly four points less on a 100-point-equivalent scale. The effect is small but statistically significant (p = 0.038), and it does not go away under scrutiny: with individual fixed effects it strengthens, and when the authors threw out the cheapest and most expensive deciles to look only at the $6–$15 range where most wine is actually bought, the negative coefficient grew roughly fourfold.

Trained tasters did reverse the sign — about seven points up for the same tenfold jump. But the authors were careful, and honest: the net expert coefficient sat at p ≈ 0.10, and they wrote that it “remains an open question whether this coefficient is positive.” The published abstract in the Journal of Wine Economics hedges even further than the working paper did, downgrading “positive” to “non-negative.”

Experts and non-experts predicted the same rating at exactly one price point: $25.70 a bottle. Below it, the experts liked the wine less than everyone else did.

Then there is the question of whether the experts are measuring anything stable. Robert Hodgson persuaded the California State Fair to let him slip triplicate pours — same wine, same bottle — into judges’ flights across four years.

Another tenth of judges scored the same wine anywhere from gold-medal to no-medal. The median spread across replicates was about four points — one whole medal category. Judges were perfectly consistent about 18% of the time, and even that mostly happened on wines they rejected: the panel agrees about what is bad, not what is good. And consistency did not persist — a judge who was reliable one year was no more likely than chance to be reliable the next.

Widen the lens from one competition to thirteen, across 3,000-plus wines, and the same picture: of roughly 1,500 wines that won a gold medal somewhere, more than 70% received no award at all in at least one other competition. Hodgson’s conclusion was that a wine’s chance of gold at one competition is statistically independent of its results elsewhere.

Finally, the mechanism. Plassmann and colleagues put twenty people in an fMRI scanner and told them they were tasting five Cabernets identified by price. There were really only three. One $5 wine was served twice — once labeled $5, once labeled $45; a $90 wine was served as $90 and as $10. Subjects reliably reported the same liquid as more pleasant when it carried the higher price (p < 0.001), and their medial orbitofrontal cortex — the brain’s hedonic accountant — lit up more brightly to match. Eight weeks later the same people tasted the same wines with no prices attached and reported no differences at all.

Price did not change the taste. It changed the pleasure.

The case study: a $19 Malbec

Which brings us to El Libertador, a Mendoza Malbec from Revolution Wine Company — 100% Malbec, 13.5% alcohol, fruit drawn from Agrelo, Ugarteche, Cruz de Piedra and the Uco Valley, and an average price of about $19 a bottle.

The crowd likes it. Vivino gives it 3.9 out of 5 across 287 ratings; Delectable, 8.8 out of 10 across 43. Its 2013 vintage rates 4.1. A public-radio wine program recommended its Cabernet sibling as a wine that “over-delivers” for under $20.

The professionals are cooler. Wine-Searcher’s aggregate is 87 points from just two critic scores — both in the 85–89 band, neither above 90. James Suckling tasted the 2022 and called it “simple and direct, but flavorful… pristine and quite drying.” CellarTracker has no community score for it at all. And across all 80,458 reviews in the Wine Enthusiast corpus, El Libertador does not appear once.

That gap — crowd-liked, critic-unbothered — is the whole analysis in miniature.

Within Argentine Malbec, the price-quality correlation is actually stronger than across wine generally — r = 0.692 against 0.606. Malbec is a category where paying more genuinely does tend to get you a better-reviewed bottle. Which makes the cheap-and-excellent Malbec rare rather than typical:

So El Libertador is not a giant-killer. At $19 it sits exactly at the median price for its category, earns an unremarkable 87 from the two critics who bothered, and wins a warm reception from ordinary drinkers. It is, precisely, an average-priced wine that ordinary people enjoy more than professionals do — which is what the blind-tasting literature predicts should happen at that price point. Below $25.70, experts and civilians diverge, and the civilians are happier.

Where the two worlds meet

The apparent contradiction dissolves once you notice the two bodies of evidence are not measuring the same thing.

Critic scores are sighted judgments made by people who know the price, the region, and the producer’s reputation. The meta-analysis is explicit that a wine’s reputation predicts its price better than its measured sensory quality does. So the r = 0.61 in the corpus is partly a measure of quality and partly a measure of how thoroughly price and prestige travel together before a glass is ever poured.

The blind data strips reputation out. What is left is a small negative effect for untrained palates and a fragile positive one for trained ones.

But “the cheap wine sometimes wins” is not the same as “price is meaningless.” In the reviewed market, the distributions barely overlap:

And the sub-$20 shelf is not uniform either. At the same money, some origins consistently review better than others:

Synthesis

The belief is mixed, and it fails and succeeds on predictable terms.

Where it fails. Blind, the premium buys nothing for an ordinary drinker — the measured effect is slightly negative, and it gets worse in the price range most people shop in. The professionals meant to certify quality cannot reproduce their own scores on the same wine, and gold medals do not survive being entered in a second competition. A price tag demonstrably changes reported pleasure while leaving the sensory experience untouched. Anyone who believes their $60 bottle would out-taste a $20 one in a covered glass is, on the evidence, probably wrong.

Where it holds. Price is nonetheless a real signal in the market as it exists. It correlates with expert scores at r = 0.61, it does so consistently across 180-plus studies at +0.30, and the score distributions of cheap and expensive wine overlap by only a few percent. Trained tasters do shift positive, above about $25.70 a bottle. Money buys something — it just mostly buys reputation, consistency, and the expectation of quality rather than a sensation your untrained tongue can pick out of a lineup.

What that means for a $19 bottle. El Libertador is not evidence that cheap wine beats expensive wine; its own critic score is a modest 87 and its category punishes bargain-hunting harder than most. It is evidence of something narrower and more useful: below roughly $25, expert and amateur judgment come apart, and the amateur is the one enjoying themselves more. A wine that ordinary drinkers rate 3.9 out of 5 and critics shrug at is not a failure of the wine. It is a demonstration that above a fairly low threshold, the thing you are buying with the next dollar is increasingly not flavor.

The honest version of “you get what you pay for” is this: you get what you pay for, but a good deal of what you are paying for is the knowing.