Today, let's go for it on fourth and short. The
University of Houston's College of Engineering
presents this series about the machines that make
our civilization run, and the people whose
ingenuity created them.
It's Super-Bowl Sunday. I'm
in my normally-quiet neighborhood, only eight
thousand feet from the stadium where it's all about
to happen. Helicopters circle. We hear horns and
sirens. A family of four could live all year on the
cost of four good seats. The only person I know at
the game is a reporter from my radio station, KUHF.
He drew the long straw and gets to cover it.
It's all pretty strange. I never got the hang of
watching football without being committed to one of
the teams. And both the Panther and Patriot players
strike me as equally nice young men.
But my interest rose dramatically when I read
today's New York Times. It told about a
paper written last year by Berkeley economist David
Romer. The title is: It's Fourth Down and What
Does the Bellman Equation Say? Romer comes to
some surprising conclusions.
Take, for example, the title question: "It's fourth
and two; should we go for it or should we punt?"
How many times has your team reached the other
team's forty-yard line and, faced with fourth and
two, opted to punt? How often has the announcer
knitted his brow and intoned the words, "They made
the right choice."?
Well, Romer has collected historical data, strung
together sequential probabilities, and discovered
that punting on fourth and short is very often a
bad call. He's also reached the more general
conclusion that coaching decisions consistently
tend to err on the conservative side.
There's more. The New York Times has also learned
that Patriots coach Bill Belichick not only read
Romer's paper, but has also been putting it to use.
So, now, as the game begins, I turn on my own TV
with renewed interest and curiosity. (Football
noise)
On their first drive, the Patriots reach the
shadows of the Carolina goalposts on fourth and
short. They ignore Romer's advice. They opt for a
field goal, miss, and go away empty. Later in the
half, they begin a drive and come up on fourth and
inches. This time they heed Romer, they go for it,
gain a first down, march to a touchdown, and go on
to win the game.
Now, I'd like to say that Super Bowl
XXXVIII was won in a University of California
office. I know that'd be going too far. And Romer
is no fool; he knows that his input data are not
perfect.
However, there is that tendency today, toward
conservative decisions -- and not only in football.
Maybe it's because we impose a greater penalty for
losing when a person gambles, than we do
when he's played it safe -- even when gambling is
actually more prudent.
So Romer has sounded a message for our times. We've
reached a point in football -- and business and
technology -- where risk-aversion becomes a virtue
beyond its usefulness. That's why it was nice to
watch the Patriot's gamble pay off -- even though
it left my wife owing our New England grandson
twenty-five cents. (Her risky gamble, it seems, did
not pay off.)
I'm John Lienhard, at the University of Houston,
where we're interested in the way inventive minds
work.
(Theme music)
D. Leonhardt, Incremental Analysis, With Two Yards to
go. The New York Times, Week in Review,
Sunday, February 1, 2004, pg. 12.
For David Romer's article, see:
http://emlab.Berkeley.edu/users/dromer
The 2004 Super Bowl XXXVIII was played at
5:30 PM, Sunday, February 1, 2004, in Houston's
Reliant Stadium, between the New England Patriots
and the South Carolina Panthers. New England won by
a field goal during the last nine seconds of the
game. The final score was 32 to 29. The KUHF
reporter covering the game was Jack Williams.
