Off the Record

Election afternoon was no sweat, from where Jimmy Breslin was sitting. “Kerry’s winning,” Mr. Breslin said on the phone. “He’s

Election afternoon was no sweat, from where Jimmy Breslin was sitting. “Kerry’s winning,” Mr. Breslin said on the phone. “He’s gonna win good. How could I be wrong?”

Sign Up For Our Daily Newsletter

By clicking submit, you agree to our <a rel="noreferrer" href="">terms of service</a> and acknowledge we may use your information to send you emails, product samples, and promotions on this website and other properties. You can opt out anytime.

See all of our newsletters

The columnist hadn’t seen the first purported exit-poll results, which had Senator John Kerry surging to victory by unimagined margins (60-40 in Pennsylvania?). Nor would he be watching when that bulletin was knocked down within half an hour.

Instead, Mr. Breslin was listening to Jimmy Breslin. In his morning Newsday column, he had declared that his prediction of a sweeping Kerry victory-first offered in May-still stood. “I am so sure,” Mr. Breslin wrote, “that I am not even going to bother to watch the results tonight.”

It was also, Mr. Breslin wrote, his last regular column. The proudly old-line newspaperman, who’d gotten his start at the Herald-Tribune , was walking away from Newsday . His latest contract extension had expired, he said on the phone, and he wants to work on longer projects.

That put Mr. Breslin two up on the rest of the press. Nobody else dared to name a winner, and nobody was free to walk away. All the months of coverage, the column-miles of analysis, the days logged on Southwest Airlines to Cleveland and Fort Lauderdale …. By Tuesday morning, it had yielded nothing.

“Down to the Wire, Deeply Divided,” The Wall Street Journal proclaimed, over a vote-proportioned electoral map split roughly 2:2:1 among Bush red, Kerry blue and we-have-no-idea white. On Slate , whose poll-tracking project refused to count any state as undecided, a Monday-morning surge for Kerry had receded to leave the final predicted count at 269-269.

Throwing the deliberations open to outsiders didn’t make things any decipherable. The New York Post gave a full page to readers’ letters: 10 for Bush, 7 for Kerry. The New York Times brought in a panel of bloggers to discuss the election on its Op-Ed page: Three were for Kerry, three were for Bush and two were against the media.

“If I weren’t busy writing about it, I’d probably just have a drink and forget about it,” The Journal ‘s Web watcher, James Taranto, said Monday night.

Mr. Breslin had opted to do just that, minus the drink. He took a morning swim, he said, then voted. “I went in there radiating,” he said. “I looked great, I guess.”

Now, he was going to think about writing a book about Brooklyn Dodgers general manager Branch Rickey, for Viking, and a play about a man who survived the World Trade Center attacks. And he was going to stay in. “I should be out celebrating my insight, my genius with a ton of whiskey tonight,” Mr. Breslin said, ruefully. “But I can’t handle the hangovers.”

The rest of the press, though, did have to keep working-even if its Tuesday stories had nothing to say, and its Wednesday ones would be pre-empted by TV. The press needed to know.

So enter the bloggers again, and the Web publications. ERROR 404: CANNOT FIND SERVER. Hang on …. There: the bloggers. The Web publications.

This was the equalizer. You could give the online folks convention passes. You could write 5,000-word profiles of them. But what they really needed was Election Day.

There were hours to fill before anyone would know the answer to the Big Question: Mr. Kerry, Mr. Bush or (please … no!) another grinding recount. But that left dozens of Little Big Questions out there-a whole list of them, scribbled in pencil in the margins of that 269-269 preliminary map.

“Can Bush get 45 to 50 percent of the vote in Dubuque County?” ABC News’ morning online tip sheet, The Note, asked. Other critically tiny questions: “How well Bush does among women in the suburbs of Philadelphia and Pittsburgh …. Kerry’s [margins] in Palm Beach County (high 50s or low 50s?).”

And The Note had a round of tiny answers, too. Maine would be “[m]ostly cloudy with a chance of rain this afternoon. Highs in the upper 40s. North winds 5 to 10 mph.” And Pensacola? Pensacola is “Cloudy. Showers and thunderstorms likely. Lows around 70. Southeast winds 10 to 15 mph.”

Yes, it would all be worthless in 12 hours. But how much more reporting would be worthless in 24 hours, or 36? How much of it would be worthless in a week? Six months? A year?

Perhaps this was the reason the old media had been glaring so intently at the blogs all along: In the long run, we’re all ephemeral. In the meantime, who could get the hottest ephemera?

Through the morning and past lunchtime, the leader in blogland was not some pajama-clad revolutionary but The New Republic . Recognizing that the day’s facts would be turning over fast, the weekly commentary magazine rounded up 30-odd stringers, in 16 key states, to create a minute-by-minute breaking-news operation.

The New Republic auxiliaries were paid $50 each, plus a free New Republic Online subscription, for the day’s activities. Their regular employers ranged in industry standing from the Miami Herald to various local weeklies. Their dispatches included weather reports (New Mexico: “the sky is clear”), background on local races (South Carolina: “conventional wisdom is that [Republican Jim] DeMint has not quite managed to grab defeat from the jaws of victory”), and, as the day went on, a steady stream of polling-place anecdotes.

At one Cincinnati polling place, a stringer wrote, workers are telling voters, “Check your chads.” In Philadelphia, a MoveOn staffer offered to fetch some “bling” for TNR ‘s stringer. “[S]he returned with a faux gold-plated necklace whose cursive letters spelled ‘Kerry.’ Apparently it had been an act of spontaneous endorsement by an ordinary American who had dropped off dozens of the necklaces at MoveOn’s office.”

Editor Peter Beinart said that the magazine had figured readers would be “desperate for any sliver of information” on Election Day. By late afternoon, as the glazed-eyed public kept hammering on their “refresh” buttons, the slivers added up to more than 10,000 words and were still accumulating. The idea, Mr. Beinart said, was to provide “great value for a short period of time.”

“Every hour from now on,” Mr. Beinart said in mid-afternoon, “the utility of the information starts to decrease.”

Mainly, that was because the exit polls were starting to leak out. At 2:50 p.m., Slate ‘s Jack Shafer posted a teaser story explaining the online magazine’s plan to flout journalistic tradition-as it had in the past-and publish the exit poll results when it could get them. That alone caused the magazine’s server to stall out immediately.

Soon thereafter, a string of numbers appeared at the top of the Drudge Report, showing Mr. Kerry ahead in at least seven battleground states, including that unexpected 20 percent bulge in Pennsylvania. Those numbers then disappeared, but not before they’d showed up elsewhere on the Web.

At the National Review Online , where three blogs had been posting turnout bulletins and mulling over campaign rumors-“[T]here’s no question that inside the Kerry camp at least some people think their man has already lost”-the episode triggered a spasm of anxiety. “Do not , again, do not take any exit-poll reports too seriously,” one poster pleaded. “JUST GET OUT THE VOTE. Exit polls not always reliable, ESPECIALLY early ones.” Liberal blogs then picked up on the National Review ‘s moment of panic, which led to a fit of debate at the National Review about whether the response had been too panicky.

Meanwhile, the information kept coming. At 3:15, Mr. Shafer brought out his first batch of exit poll readings-less dramatic than Matt Drudge’s batch, but still showing Mr. Kerry ahead. “KERRY LEADS” Slate proclaimed, before dialing the headline back to “Early Exit: A Squeaker.”

Mr. Drudge, for his part, was interpreting the result as “Kerry in striking distance-with small 1% lead-in Florida and Ohio.” That claim, however, picked up caveat after caveat as the minutes went by: Al Gore had led Florida by 3 points in 2000 exit polls …. Exit polls had showed the 2000 race tied in Colorado …. The Drudge Report did not, however, stop posting exit-poll updates.

It was the 2000 exit polls, especially Florida’s, that had led to the downfall of the entire election-reporting system. Four years later, there isn’t even any settled theory of why the polls went wrong-whether they betrayed some underlying pro-Democratic assumptions, or simply reflected the fact that a lot more people had tried to vote for Al Gore in Florida.

As the polls started to close, and the hands of the nation lifted off computer mice and reached for the TV remote, Mr. Shafer’s last update was headed “Mucho flattering to Kerry,” while Mr. Drudge’s page declared that things were “TIGHTENING.”

Then, as television took over, the ever-ingenuous Mr. Drudge adopted a new banner. “ENOUGH OF THE MEDIA EXITS,” his report declared in red capital letters. “LETS COUNT THE PEOPLE’S VOTES!”

Off the Record