Say it both ways

The same choice, put as what you keep or as what you lose, gets different answers. Restate it the other way before you decide.

Amos Tversky and Daniel Kahneman (1981) asked people to choose between two plans against a disease expected to kill 600 people. Put as lives saved, plan A saves 200 for sure, and plan B has a 1 in 3 chance of saving all 600 and a 2 in 3 chance of saving none: 72% chose the sure plan. Put as deaths, plan A means 400 die for sure, and plan B has a 1 in 3 chance that nobody dies and a 2 in 3 chance that all 600 do: now 78% chose the gamble. The plans were the same; only the description changed.

Their earlier prospect theory (Kahneman and Tversky 1979) explains why. People judge outcomes as gains or losses from a reference point, not as end states. Each extra gain pleases less than the one before, so saving 600 does not feel three times as good as saving 200, and the sure 200 wins. Each extra loss hurts less than the one before, so losing 600 does not feel 1.5 times as bad as losing 400, and the gamble wins. Framing moves the reference point, and the choice moves with it. Separately, a loss hurts more than an equal gain pleases, by roughly two to one in later estimates (Tversky and Kahneman 1992).

It is not only students in a laboratory. Barbara McNeil and colleagues (1982) asked patients, students and doctors to imagine choosing a treatment for lung cancer. All three groups were more likely to choose surgery when its results were given as the chance of surviving than as the chance of dying, though the figures were the same.

Neither plan is wrong in itself: in the examples above, both keep the same amount on average. What is hard to defend is choosing differently because of the words. The simple check is to say the choice both ways, or as plain end states (“we end with 200 homes saved for sure, or a 1 in 3 chance of all 600”), and see if your answer survives. It will not remove the pull of a loss, but it shows you when the wording is doing the deciding.

Why it matters

Each extra gain pleases a little less than the last, and each extra loss hurts a little less than the last. So with a sure gain people play safe, and facing a sure loss they gamble to avoid it. The facts have not changed; only the words have.

Away from the game

When an offer, a plan or a price is put to you as a gain ("saves 200") or as a loss ("400 lost"), say it the other way before you answer. If your answer changes, the wording was deciding, not you. The same holds after a setback: a loss you want to escape pulls you towards long shots.