The winner's curse
overbidding · auction winner curse
In a competitive bid for something of uncertain value, the winner is the one who overestimated it most. Winning is therefore evidence that you were too optimistic, and the more bidders there were, the stronger the evidence.
In practice
You won the competitive pitch on price. The information in that is that you priced it lower than everyone else who looked at the same scope.
The common mistake
Treating a win as validation. In a common-value contest the win is drawn from the tail of the estimate distribution, which is exactly where the errors are.
Several parties bid for something whose true value is the same for all of them but unknown — an oil lease, a contract, a company. Each forms an estimate. The estimates scatter around the truth.
The highest bid wins. The highest bid comes from the bidder whose estimate was highest, which is to say the one who was most wrong in the optimistic direction.
Winning is evidence that you overestimated. And the more bidders there were, the further into the tail the winning estimate sits, so the more you overpaid.
The counterintuitive part
More competition makes the curse worse rather than better. With three bidders the winner is the highest of three draws; with twenty it is the highest of twenty, which is much further out.
This is why the correct response to learning that many parties are bidding is to bid less, which feels exactly backward.
Where it operates
Competitive pitching. You win the ones where you priced lowest relative to the scope you understood. Some of those you won because you understood the scope worse than the firms who bid higher — which is the curse operating on the seller's side, and it arrives as a project that overruns.
Hiring in a competitive market. The candidate you win a bidding war for is one you valued more highly than everybody else who interviewed them. Sometimes that is insight; on average it is optimism.
Acquisitions. The most documented instance. A buyer who beats several informed rivals has, by construction, the highest estimate of the asset, and the systematic overpayment in competitive acquisitions is exactly this.
Any tender where the client picks on price. The winner is whoever most underestimated the work.
Protecting against it
Adjust for how many are bidding. The more competitors, the more you should shade your estimate down, because the win itself is more informative about your error.
Bid your estimate minus the curse, not your estimate. Ask what would have to be true for you to win, then ask whether it is.
Prefer contests where the value is private rather than common. If the thing is genuinely worth more to you than to others — because of an existing relationship, a capability, adjacency to work you already do — then winning is not evidence of error. It is evidence of the advantage. This is the strongest argument for positioning: the way out of the curse is to stop competing on the same estimate.
Treat easy wins as information. A pitch won with no resistance should prompt a review of the scope rather than a celebration.
Concept web
Open the full webQuestions
What is the winner's curse?
In a competitive bid for something of uncertain but common value, the winner is the bidder who overestimated it most. Winning therefore carries information that you were too optimistic.
Why does more competition make the winner's curse worse?
Because the winning bid is the highest of however many estimates were made. With more bidders the winning estimate sits further into the optimistic tail, so the overpayment is larger.
How do you avoid the winner's curse?
Shade estimates down as the number of bidders rises, ask what would have to be true for you to win and whether it is, and prefer contests where the thing is genuinely worth more to you than to others.