OPINION — To watch the World Series or the baseball playoffs is an invitation to enter the realm of the irrelevant statistic.
When the Red Sox pushed across their first run in the first inning Monday night, a Fox announcer burbled that Boston also scored in the first inning of the opening game when they won the Series in 2004, 2007 and 2013. Of course, no players from 2004 and 2007 are on the active Red Sox roster — and scoring in the second or fifth inning can also be an effective way of winning a ballgame.
Highlighted on the screen every night are the postseason stats of the players, even though the Dodgers, for example, played just 11 games to advance to the World Series. Virtually invisible during the telecasts are what the players accomplished during the full 162-game season, even though these numbers have infinitely more predictive value.
Those of us who closely follow politics can easily feel superior to the statistical naïfs in the World Series broadcast booth. But often we fall into a similar trap — the arrogant belief that the future is knowable if you crunch enough numbers.
The 2018 midterms have brought a new level of sophistication to political analysis.
The New York Times in conjunction with Siena College has completed more than 60 congressional polls with the real-time results flashed on the computer screen with each call. Polling analysts like Nate Silver at FiveThirtyEight have honed their predictive models. And traditional race-by-race handicappers like Nathan Gonzales, Charlie Cook and Larry Sabato’s Crystal Ball do an exemplary job on the granular level.
Yet, even after hearing cautionary words about polling and polling analysis, we tend to overreact to every statistical gyration.
A Rutgers-Eagleton Senate poll from New Jersey, released Wednesday, shows scandal-scarred Democratic incumbent Robert Menendez up by just 5 points in a state that Hillary Clinton carried by 14 points. Is the race tightening? Could Democrats blow a seemingly safe seat? Or is the poll just a margin-of-error quirk?
Multiply this reaction by nearly 500 contested races for Congress and governor, and you have a rough sense of the stress levels on both sides less than 300 hours before the polls open on Election Day.
The truth is that all the polling in the world cannot prepare you for a midterm election with this broad a terrain. And there are many hard-fought House races, like in Maine’s 2nd District, where there have only been a handful of reputable polls.
Sure, broad questions like the generic ballot (“Do you prefer a Congress controlled by the Democrats or the Republicans?”) can offer national clues. But is the Democrats’ current 8-point average edge on this question enough to comfortably overcome the GOP-leaning effects of incumbency and gerrymandering in the House?
A recent Washington Post-Schar School poll of 69 hotly contested House districts found the two parties effectively tied. But is this good news for the Republicans because they are holding their own in battleground races? Or is this a formula for a Democratic sweep since more than 90 percent of the districts surveyed are currently GOP held?
Knotty questions like these can make political pundits sound like medieval theologians debating how many angels can dance on the head of a pin. At a certain level there are no hard truths about the 2018 elections yet — only personal tendencies towards optimism or pessimism.
After enduring Election Night 2016, many Democrats are like battered children brooding that once again victory (this time, winning the House) will be snatched from their grasp. Similarly, there are Republicans who believe that Donald Trump will help them defy the polls, while other GOP strategists privately worry that a president with no impulse control is leading them to political doom.
For all the twitchiness of the political class, it is also worth remembering that the prevalence of early voting in many states cuts into the ability of a single late-breaking event to reshape the midterms. Already, in early-voting states, we are approaching the point when it is too late to effectively drop opposition research.
Watch: Initial Early Voting Data Appears to Favor GOP in Several Key States
Sometimes politics is simply perverse. This is not a reference to the Trump triumph in 2016, but rather to an earlier midterm election that should serve as a cautionary tale for all those eager to make glib predictions.
In 1986, the Democrats won seven Senate seats by a collective margin of 98,000 votes. Those 98,000 votes were the difference between a 55-45 Democratic Senate and one that was controlled by the GOP. As House Speaker Tip O’Neill said triumphantly after the election, “If there was a Reagan revolution, it’s over.”
In the 1960s, an ad campaign for Schlitz, a century-old Milwaukee beer, featured the tagline, “You only go around once in life — go for all the gusto you can.”
In elections, as well, you only go around once.
So it doesn’t matter that in some alternate universe Hillary Clinton is president and the Republicans held the Senate for Ronald Reagan in 1986. Whatever happens on Nov. 6 will seem foreordained by the gods, but in truth, it will be just one of dozens of equally plausible outcomes.
What four decades covering politics has taught me more than anything is humility. Neither I nor anyone practicing punditry on cable TV has the gift of prophesy.
But, if I had to guess, I would say that the election comes down to how voters sort out the three dominant fundamentals: A booming economy, albeit with a suddenly shaky stock market. A president who violates every rule of decency with his incessant fearmongering and vicious attacks. And a Democratic Party able to aggressively compete across a broad swatch of GOP-held House districts while grappling with a Senate map wildly tilted toward the Republicans.
Beyond that, I will be a traitor to the pundit class as I dispense with glib predictions. Instead, I will bow toward the unknowable as I breathlessly wait until the returns come in on Election Night.
Walter Shapiro, a Roll Call columnist since 2015, has covered the last 10 presidential campaigns. He is also a fellow at the Brennan Center for Justice at NYU and a lecturer in political science at Yale. Follow him on Twitter @MrWalterShapiro.