In an uncertain situation where people face the prospect of losing, they would prefer to lose
through omission - choosing not to act - than commission - acting in a way that ends in failure.
That is one of the conclusions of new research on economic decision-making by Professors Rafael
Tenorio and Timothy Cason. Their work, which appears in the latest issue of the Economic
Journal, analyses the decisions of contestants in the long-running US game show The Price is
Right and shows that on the whole people pursue the game strategies that are most likely to bring
them the best outcome.
The Price is Right is the longest-running game show in television history, and is now in its
thirtieth season on the US television network CBS. On the show, contestants play various games
for high stakes such as cars and exotic trips. These games provide economists with valuable
'natural experiments' in decision-making under uncertainty.
Tenorio and Cason analyse a simple game called 'The Wheel', played twice on each broadcast of
The Price is Right. Three contestants spin a wheel divided into twenty segments, each labelled
with the numbers 5 through 100 in intervals of five. Each may spin up to two times, with the
second spin being optional. The winner is the contestant whose one or two spins sum to the
highest number that does not exceed 100. Each player completes her spins before the next one
begins.
The critical decision in this game is whether or not to take a second spin. A second spin can
potentially increase the chance of winning because it can bring the sum total closer to 100. But by
taking a second spin, the contestant also risks going over 100 and losing immediately. So the
question is what is a contestant's best or 'optimal' strategy?
The second spin decision obviously depends on the outcome of the first spin. The authors show
that the critical threshold that determines if a second spin is optimal also depends on the
contestants' order of play: for example, the first player must be more aggressive and take chances
with a second spin because the two subsequent players could exceed her spin total. In addition, the
optimal threshold depends on the 'bonus' cash that is awarded to players whose spins total exactly
100.
Comparing actual spin decisions for 282 plays on the game show and for 775 plays in an
analogous laboratory experiment to the optimal spin strategies derived using game theory, the
researchers find that:
l Players usually decide correctly whether or not to use their second spin, even though to
derive the optimal spin theoretically, a computer must evaluate sixty-four million potential
spin combinations!
l Players do make errors, but only near the first-spin threshold where the expected earnings
from taking a second spin are very similar to the earnings from forgoing a second spin.
l These errors exhibit a systematic bias. Players rarely spin a second time when they should
not; instead, the overwhelming majority of mistakes occur when players fail to take a
chance and spin a second time when it is optimal to spin.
The researchers interpret this result as follows: if players face the prospect of losing, they prefer
that this occur through their own omission (they chose not to spin again, but a later player beats
their spin total) than commission (they spin again and immediately lose by totalling more than
100).
This 'omission bias' has also been documented in psychology for a variety of contexts, ranging
from hypothetical moral decisions (such as how to keep another individual from harm) to the
casino game Blackjack (also known as '21').
Finally, the authors show that play on the game show does not differ significantly from their
analogous laboratory experiment with monetary stakes that are only a small proportion of those on
television. This is an important finding for the emerging field of laboratory research in economics,
since it shows that behavioural conclusions are applicable to a wide range of payoff sizes. This
suggests that laboratory research in economics may be highly relevant for understanding real
economic decision-making.
Notes for Editors: 'To Spin or Not to Spin? Natural and Laboratory Experiments from The Price
is Right' by Rafael Tenorio and Timothy Cason is published in the January 2002 issue of the
Economic Journal. Tenorio is at DePaul University, Chicago, Illinois; Cason is at Purdue
University, West Lafayette, Indiana.
For Further Information: contact Rafael Tenorio on +1-312-362-8309 (email:
rtenorio@condor.depaul.edu; website: http://economics.depaul.edu/rtenorio/); Timothy Cason on
+1-765-494-1737 (email: cason@mgmt.purdue.edu; website:
http://www.mgmt.purdue.edu/faculty/cason/); or RES Media Consultant Romesh Vaitilingam on
0117-983-9770 or 07768-661095 (email: romesh@compuserve.com).