Polls, Lies and ‘The Prediction Business’
These are the days of uncertainty. The pollsters tell us so.
You wouldn’t know it from the pundits, however. They rend their garments or sing hosannas as if each little survey statistic was handed down from on high, chiseled into tablets.
“Is Mitt Romney panicking?” asked a Washington Post headline in mid-September after polls indicated President Obama had received a post-convention bounce. “Did Obama just throw the entire election away?” an alarmed Andrew Sullivan asked after the first presidential debate, reeling off a flurry of the president’s poor polling numbers.
The fact is, every day is a day of uncertainty — and we hate that.
Psychologically, it gives us comfort to determine our fates, says Dr. David Reiss, a San Diego-based psychiatrist and expert on personality dynamics.
“We’re sort of hardwired to predict what is going to happen to protect ourselves,” he says. After all, for much of mankind’s recorded history, a prediction could literally mean the difference between life and death — whether it was determining a village’s vulnerability to attack or attempting to figure out the size of the coming harvest.
But, he adds, there’s also a sense of power and ego in knowing the future: “If I can predict things, I have some control and it releases feelings of helplessness or fear,” he says. “The more I can predict, the smarter I am and the better I look to others.”
Throw in the chattering, nattering, opinion-splattering classes — which traffic in confidence and authority — and you create a stew of statements that seldom veers from bet-the-house conviction.
In his new book, “The Signal and the Noise: Why So Many Predictions Fail — But Some Don’t,” statistician and fivethirtyeight.com creator Nate Silver looked into almost 1,000 predictions of the D.C.-based “McLaughlin Group” pundits. They were about half right — no better than a coin flip. These Beltway insiders “displayed about as much political acumen as a barbershop quartet,” Silver writes.