The second cheapest wine on a restaurant list isn't the trap it's reputed to be. Economists at the London School of Economics and the University of Sussex compared thousands of restaurant wines against their retail prices and found that the markup on the second cheapest bottle sits significantly below the markup on the four next most expensive wines, with the heaviest markup landing near the middle of the list instead. The bottle most people avoid out of embarrassment is usually one of the better-value entries on the page, and the ones they trade up to are where the restaurant makes its money.
That reverses the advice almost everyone has heard, and it has been repeated so confidently for so long that most of us order around it without ever asking where it came from. Below is what the research actually measured, where the real markup sits on a list, what a glass costs against a bottle once you do the arithmetic, and what to use instead of a price rule that was never doing the job you hoped it was.
Is the second cheapest wine a rip-off?
No, and the only study to test the question directly found close to the opposite. David de Meza of LSE and Vikram Pathania of the University of Sussex Business School published their answer in the journal Economics Letters in 2021, and their abstract puts it plainly: the markup on the second cheapest wine is significantly below that on the four next most expensive wines, so "it is therefore an urban myth that the second-cheapest wine is an especially bad buy".
The size of the data behind that is what makes it worth trusting. According to the University of Sussex account of the study, the pair worked from 470 wine menus used by 249 London restaurants, covering 6,335 wines in total, and ran each wine's name, description and vintage through a price comparison site to find the cheapest available retail price. They found an exact retail match for around two thirds of those wines. That's a real sample of real lists rather than a few sommeliers asked for an opinion, which is what almost every article on this subject rests on.
The reasoning behind the myth falls apart under its own logic too, and it did so years before anyone measured it. Mike Veseth made the point on The Wine Economist in 2016, arguing that the rule assumes restaurants have never heard of it. If a restaurant wanted to exploit the habit, it would simply raise the markup on the second cheapest bottle, at which point you'd move to the third, and the restaurant would follow you there. Play that out and you end up buying the most expensive wine on the list to stay ahead of a strategy nobody is actually running.
Where does the biggest markup on a wine list actually sit?
Around the middle of the list, which is exactly where a cautious diner tends to land after talking themselves up from the bottom. De Meza's own summary in the Sussex piece is specific about the shape of it: the percentage markup on the second cheapest wine is significantly below that on the third, fourth and fifth cheapest, and well below the peak markup, which tends to occur around the median wine on the menu. The two of them put it in a single line in the working paper they wrote for the American Association of Wine Economists, where the abstract states that percentage markups are highest on mid-range wines.
Pathania offers a reason that makes commercial sense once you hear it. Margins are kept down at the bottom of the list to encourage people to order wine at all, and they're kept down at the top to tempt anyone hovering over an expensive bottle into going one better. The middle has neither job to do, so it carries the heaviest percentage markup on the page. Nobody is being singled out for punishment at position two. The list is simply shaped by what the restaurant wants to happen at each end of it, and the middle is what's left over. The authors go further in that working paper and conclude the whole pattern is what ordinary profit-maximizing pricing of a product range predicts anyway, with no need for any theory about exploiting embarrassed diners.
This is the practical inversion worth carrying into your next dinner. If you've been using a rule to avoid getting stung, the rule has been steering you toward the stung part of the list rather than away from it, and the honest version of the advice is closer to the opposite of what you were told.
How much do restaurants mark up wine?
Roughly three times retail on average, and the two sides of the Atlantic broadly agree on the figure. The London study found the mean percentage markup of restaurant wines against retail prices was more than threefold, and the Sussex summary notes that the increase was similar for red and white. The researchers deliberately compared against retail rather than wholesale, on the grounds that the jump from supermarket shelf to restaurant table is the comparison a diner would naturally make anyway.
American trade guidance lands in the same place from the other direction. Provi, which sells ordering software to bars and restaurants, tells its operators that 200 to 300 percent over retail is the accepted markup standard, and works the example through: a wine that retails at $20 should be listed somewhere around $60 to $80. The same guidance notes that rare, vintage and specialty bottles often run higher again, into the 400 to 500 percent range, which is worth knowing before you read a steep price on an unusual wine as evidence of quality.
Knowing the multiple changes how a list reads. A $60 bottle in front of you was probably a $20 bottle in a shop, and that isn't a scandal, because it's paying for the room, the glassware, the staff who carry it and the risk of an open bottle going unsold. It does mean the price on the page is telling you about the restaurant's economics rather than about the wine, which is the whole difficulty with using price position as a proxy for anything at all. Our piece on what you're actually paying for in a cheap bottle against an expensive one goes further into where the money goes.
Is it cheaper to order wine by the glass or by the bottle?
By the glass costs a little more, and much less than the folklore claims. The same researchers checked it, and the Sussex account reports that a large 250ml glass, which is about 8.5 US fluid ounces, works out at less than 7 percent more per unit volume than buying the bottle. Choosing to drink by the glass does not put you significantly out of pocket, which quietly retires another rule people order by.
The American convention for setting that price is blunter and worth knowing. Provi's guidance describes the most basic approach as pricing a glass at the wholesale cost of the whole bottle, so a bottle bought in at $15 gets listed at $15 a glass, and it notes that most restaurants pour four to six glasses from a 750ml bottle. The alternative method it describes is to divide the bottle's list price by the number of pours, which usually arrives at a similar number and handles the case where a bottle doesn't sell out before it fades.
For a table of two having a glass each, the glass is usually the sensible order and you're paying a small premium for the freedom to have two different wines. Once a table is drinking three or more glasses of the same thing, the bottle starts winning on both price and quality, since a bottle opened for you is fresher than one that's been standing behind the bar since lunchtime. If you'd rather bring something of your own, the corkage math tells you when that beats both options.
Why does everyone order the second cheapest bottle anyway?
Because it feels like a socially safe answer to a question the list makes uncomfortably public. Food and Wine polled readers and colleagues on exactly this in 2025 and found the habit is close to universal outside the trade, with 100 of 119 respondents saying they pick the second cheapest bottle. The reasons people gave were about how the order would look rather than about the wine, running along the lines of not wanting to seem cheap while not wanting to overspend either.
That piece also shows how firmly the myth still holds among people who work in wine. It quotes wine expert and content strategy consultant Heather Muriello arguing that the second cheapest slot "often carries one of the highest markups" precisely because restaurants know it's the go-to choice, and that operators put a wine there with a lower wholesale cost or a higher margin to take advantage of it. The same magazine's reporting carries no measurement alongside the claim, and the London data across 249 restaurants points the other way, so the fairest reading is that a plausible pricing strategy has been widely assumed rather than widely practiced.
There's a more cheerful account of the same slot from people who build lists. In the Food and Wine piece, Liz Martinez, a general manager and sommelier in Washington DC, describes using the lower end of a list to slip in unusual bottles that guests might not otherwise try, on the grounds that the price makes people adventurous and she'd rather see guests drinking wine than admiring it. A cheap wine on a list someone cared about is often there because somebody wanted you to find it.
So what should you order instead?
Set your budget first, then choose inside it on grounds that actually predict taste. Price position on a list tells you about the restaurant's margin strategy, as the research above makes clear, and it can't tell you whether you'll enjoy what arrives. Those are two separate questions, and the price rule fails because it answers the first one while you're asking the second.
The information that does predict taste is on the same page, in the grape, the region and the style words next to each entry. That's where a little reading pays off, and our guide to which wine terms genuinely predict how a bottle will taste sorts the useful words from the decorative ones. It's the same skill that turns a whole list into something readable rather than a wall of names, and it doesn't require knowing any of the producers.
This is the moment Current Venue was built for. Sommy recognizes the restaurant you're sitting in, loads that venue's own wine list, and lets you filter it the way you'd actually decide: a price range you're comfortable with, by the glass or by the bottle, a color, a grape if you have one in mind. What's left gets read against your Taste Profile, the living picture of your palate that Sommy builds from what you rate, so the shortlist is drawn from the wines on that specific list rather than from wine in general. The price filter does the job the second cheapest rule was badly approximating, which is to keep you inside a budget, and then something other than price picks the actual bottle.
It's worth being straight about the limits. Sommy can only work from the list it can see, so an obscure grower with nothing written about it anywhere is still a leap of faith, and a wine list that has changed since it was last captured will have entries that don't match. The feature narrows a page of unfamiliar names down to a few worth considering. It doesn't promise you'll love the one you pick, because nobody can promise that about a bottle you haven't tasted.
Can you work out a wine list before you arrive?
Yes, and it takes the decision out of the moment when everyone at the table is watching you hold the list. Planned Visit does the same work as Current Venue ahead of time, so you can look at where you're eating on Friday and know on Tuesday which two or three bottles are worth your attention.
That matters more than it sounds when you're the one ordering for other people. Most of the discomfort in this whole ritual comes from having to make a public decision quickly, with an audience and a member of staff waiting, which is exactly the situation in which a simple rule like the second cheapest one becomes appealing. Deciding in advance removes the time pressure without removing the choice. If you'd rather work the room properly on the night, our walkthrough on ordering wine at a restaurant covers asking for help in a way that gets you a good answer, which every sommelier quoted above agrees is the strongest move available to you.
Frequently asked questions
Is the second cheapest wine a rip-off?
No, and the research says so directly. Economists at LSE and the University of Sussex, publishing in Economics Letters in 2021 and drawn from 6,335 wines across 249 London restaurants, found the markup on the second cheapest wine is significantly below the markup on the four next most expensive bottles. They describe the rip-off belief as an urban myth.
Which wine on a restaurant list has the highest markup?
The wine around the middle of the list carries the heaviest percentage markup, according to the same study. Margins are held down at the cheap end to encourage people to order wine and at the expensive end to tempt them to trade up, which leaves the median wine doing the earning. The authors state it plainly in their working paper for the American Association of Wine Economists: percentage markups are highest on mid-range wines.
How much do restaurants mark up wine?
More than three times retail on average in the London data. American trade guidance points at a similar band, with Provi advising operators that 200 to 300 percent over retail is the standard, so a $20 retail bottle lists at roughly $60 to $80, and specialty bottles often run higher.
Is the house wine worse than the second cheapest wine?
Not automatically, and in a restaurant that cares about wine the house pour is often chosen with real attention because it represents the place. The cheapest wine on a good list is frequently the one the buyer put the most thought into, since it has to be both inexpensive and good enough to serve constantly.
Should you order wine by the glass or by the bottle?
By the glass costs under 7 percent more per unit volume than the bottle in the London study, so it's a fair deal for one or two glasses and buys you the freedom to drink different things. Once a table wants three or more glasses of the same wine, the bottle is usually better value and reliably fresher.
Is it rude to order the cheapest wine on the list?
Not to anyone worth eating in front of. Several sommeliers and wine directors quoted in the trade press say they take particular care over their least expensive listings, and the discomfort people feel about ordering them is about self-image rather than about anything the restaurant thinks.
Order the bottle, not the position
The second cheapest wine has spent a decade carrying a reputation it didn't earn, and the only people who ever measured it found it to be one of the fairer prices on the page. If you like the look of it, order it without the small apology most of us have learned to attach.
What's worth dropping is the idea that a position on a price ladder can stand in for knowing anything about the wine. Decide what you want to spend, which is a question only you can answer, and then choose within that on grape, region and style, which are questions the list will actually answer if you read it. Do that a few times and the ritual stops being an exam you might fail in front of your friends, and turns back into the far better thing it was supposed to be, which is picking something you're looking forward to drinking.
