Bidding Strategy and the Psychology of the Saleroom
What Is the Winner's Curse, and Why Winning Can Mean Overpaying
What is the winner's curse? In any sale where several bidders estimate the same uncertain value, the buyer who wins is the buyer whose estimate ran highest, which makes the final bid the most optimistic valuation in the room.
What is the winner's curse? In any sale where several bidders estimate the same uncertain value, the buyer who wins is the buyer whose estimate ran highest, which makes the final bid the most optimistic valuation in the room. Three petroleum engineers named the effect in 1971, from offshore oil leases. Auction glossaries have carried the term ever since without connecting it to the case where it bites hardest: a unique contemporary work with no comparable sale behind it. This article is for collectors who already bid and want the mechanism, the arithmetic and the rules that contain it.
Key points:
- What is the winner's curse, in one line: the systematic failure to account for the fact that winning is itself evidence your estimate was too high. Kagel and Levin summarise it as "you win, you lose money, and you curse".
- Christie's publishes that bidding "advances in increments of 10%", so the step between the last underbid and the hammer is set by the house.
- Phillips charges a standard buyer's premium of 29% plus VAT on the first £1,500,000 of hammer price, or 34.8% including VAT in London.
What Is the Winner's Curse, and Where the Term Came From
The winner's curse is the systematic failure to account for an adverse selection effect built into the act of winning. In a common value auction, where the object is worth essentially the same to every bidder, each participant holds a private estimate of that value, and, as John H. Kagel and Dan Levin of Ohio State University write, "the winner's estimate of value is (one of) the highest estimates". Bidders who ignore that fact earn "below normal or even negative profits".
E. C. Capen, R. V. Clapp and W. M. Campbell reached the same conclusion from field data. The three petroleum engineers found that oil companies bidding for Outer Continental Shelf leases suffered unexpectedly low returns "year after year", and published the finding in the Journal of Petroleum Technology in 1971. Richard H. Thaler carried the idea to a general economics audience in his 1988 Journal of Economic Perspectives column.
Auction glossaries answer what is the winner's curse in a single line and move on. Kagel and Levin report that experienced industry executives tested in the laboratory "suffer to the same extent from the winner's curse as do student subjects".
Common Value: Why a Unique Work Sharpens the Curse
In a pure private value sale, each buyer knows what an object is worth to them and the winner wants it most. Contemporary art at auction is a hybrid: part of a work's value is personal, and part of it is a common value, the resale level other collectors would also pay. That common component is what every bidder is estimating, and it is where the curse operates.
The width of the estimate spread decides how badly the winner overshoots. A screenprint from an edition of 100 has comparable sales behind it, so every bidder works from a narrow range and the highest estimate sits near the centre. A unique painting with no comparable sale gives a wide range, and the highest draw from a wide distribution sits far above its middle. Kagel and Levin document a second effect: bidders who learn to avoid the curse against four rivals fail again when the field grows to seven.
| Setting | What every bidder is estimating | Why the top bid overshoots |
|---|---|---|
| Offshore oil lease | The same barrels under the same seabed | Identical object, so the highest draw wins |
| Print from an edition of 100 | A resale level visible in comparable sales | Narrow spread, small overshoot |
| Unique painting, no comparable sale | A price nobody has observed | Wide spread, so the winning estimate sits furthest above the centre |
Resale does not reliably repair an overpayment. Arthur Korteweg, Roman Kräussl and Patrick Verwijmeren examined 32,928 paintings that sold repeatedly between 1960 and 2013 in The Review of Financial Studies, and found that correcting for the tendency of appreciating works to come back to market cuts average annual index returns from 8.7% to 6.3%.
How Optimism Alone Sets the Final Price
An estimate is an opinion with a house name on it. Christie's auction glossary defines it as "a monetary range that reflects the expectation of our specialists of price at auction, excluding premium and taxes". Phillips writes that a bid inside the range "should, in our opinion, offer a chance of success", while adding that "many lots achieve prices below or above the pre-sale estimates".
The reserve cannot exceed the low estimate at either Christie's or Phillips, and Sotheby's confirms that the auctioneer "may open the bidding on any lot by placing a bid on behalf of the seller" up to that reserve. Above the reserve, only live competition moves the price. The last underbidder is the single market check on the winner, and once that underbidder stops, the buyer pays one full increment above the second-highest opinion in the sale.
Scale sets how many opinions there are. Public auction sales reached USD 20.7 billion in 2025, up 9%, according to the Art Basel and UBS Art Market Report 2026 by Arts Economics, with works above USD 10 million growing 30% while the segment below USD 50,000 fell 2%.
What Is the Winner's Curse Worth in Cash: a Worked Example
Phillips publishes the arithmetic. On a hammer price of £2,000,000, its standard buyer's premium is 29% on the first £1,500,000 and 22% on the balance, giving £545,000, and £654,000 once 20% VAT is added. The buyer pays £2,654,000, or 32.7% above the hammer.
In the €800 to €50,000 band where most contemporary lots trade, a €20,000 hammer with LLB Auction's published 20% buyer's premium invoices at €24,000. One 10% increment past the underbidder raises the hammer to €22,000 and the invoice to €26,400, so a single step of optimism costs €2,400.
What Is the Winner's Curse in a Timed Online Sale?
A timed online auction runs without an auctioneer, over a window of seven to fourteen days at LLB Auction, and that changes which defences are available. Christie's defines a max bid as "the highest bid you are willing to place on an item in an online auction", executed automatically in increments until it is exceeded. Sotheby's describes the equivalent absentee bid as a maximum executed "at the lowest price possible" and "never for more than the maximum bid amount you indicate".
Kagel and Levin found that open outcry English auctions attenuate the winner's curse, because watching lower bidders drop out reveals their estimates, and that the same bidders gain nothing from that experience in sealed-bid formats. A maximum bid returns no such information, so any shading for the number of rivals has to be applied before the bid is lodged, which is why how to set a maximum bid in an online auction deserves more preparation than the live equivalent. Sotheby's is blunt: a leading maximum bid "does not guarantee that another client will not outbid you". Proxy bidding or bidding live decides how much of that risk you carry.
Seven Rules That Defend a Collector Against Overpaying
Knowing what is the winner's curse changes nothing until it becomes a procedure you follow on sale day.
- Set the ceiling before you read the catalogue entry. Fix the number from comparable sales and condition first.
- Count the premium inside the ceiling. A 20% buyer's premium turns a €20,000 ceiling into a €24,000 commitment before shipping or resale royalty.
- Shade the bid for the size of the field. Kagel and Levin showed that seven rivals produce more severe adverse selection than four, so the same estimate justifies a lower bid.
- Treat the estimate as the house's expectation. Christie's builds it from condition, rarity, quality, provenance and comparable sales, and excludes premium and taxes.
- Buy the condition report before the work. Sotheby's sells every lot "AS IS", so an unexamined restoration is a discount you failed to negotiate.
- Lodge a maximum bid and leave it untouched. Raising a ceiling in the closing hours is the behaviour the curse feeds on, a pattern examined in the psychology that makes smart collectors overpay.
- Accept the lot you lose. A bought-in lot, in Christie's definition one that failed to meet its reserve, costs you nothing.
How LLB Auction Clears the Threshold Before You Bid
LLB Auction works as a buyer's advocate, narrowing the estimate spread that makes what is the winner's curse expensive.
Curated intake. Roughly 40% of submitted works are declined, so the catalogue narrows what a bidder has to guess at.
Documented due diligence. Every lot carries certificate verification, an ownership history and a three-page condition report, which compresses the private variance that widens estimate spreads.
Published costs. The buyer's premium is 20% and the seller's commission 10%, disclosed before bidding. The Contemporary Art Spring 2026 sale of 26 May 2026 offered 25 lots and sold 23, for roughly €39,480 in hammer.
Register at llb-auction.com to receive the catalogue and condition reports for the next timed sale before bidding opens.
Conclusion
The bidder who wins a unique work has, by construction, produced the highest estimate of a value nobody can verify. That is a structural feature of common value bidding, which is why the defences are procedural: a ceiling fixed in advance, the premium counted inside it, a shade for the number of rivals, and a maximum bid left alone once lodged. What is the winner's curse, in the end, is the price of treating a winning bid as a valuation, and the collectors who escape it decide what a work is worth before the sale opens.
Also worth reading:
- Bidding with a cool head in the saleroom
- The psychology that makes smart collectors overpay
- Setting your ceiling: what a work is worth to you
- Enchérir sans se laisser emporter : méthode et sang-froid
Sources:
- Auction glossary : Christie's, 2026
- Guide for Buyers : Sotheby's, 2026
- UK Auction Buyer's Guide : Phillips, 12 April 2026
- Winner's Curse : Kagel and Levin, Ohio State University
- Common Value Auctions and the Winner's Curse : Kagel and Levin, Princeton University Press, 2002
- Anomalies: The Winner's Curse : Richard H. Thaler, Journal of Economic Perspectives, 1988
- Does It Pay to Invest in Art? : Korteweg, Kräussl and Verwijmeren, Review of Financial Studies, 2016
- Art Market Report 2026, Auctions : Art Basel and UBS, Arts Economics, 2026
