Voters vs. Zogby: a once-golden pollster is tarnished.

NEVER mind the concession from John Kerry. The GOP ought to be demanding a concession from pollster John Zogby.

After his striking successes in 1996, when he accurately projected Bill Clinton's margin of victory as much narrower than other pollsters were expecting, and 2000, when he picked up on the late Gore surge, Zogby established himself as a brand name in polling. After not buying the Clinton hype in 1996, he became a favorite of the Right--touted by the New York Post, Fox News, and Rush Limbaugh. Conservatives may not have liked his assessment of the last-minute swing to Gore in 2000, but it's tough to argue with accuracy.

But somewhere along the line, whatever "secret sauce" Zogby was using in his weighting of demographics and calculations of likely voters went sour. In 2002, his final polls were pretty lousy. In Minnesota, Zogby predicted Democrat Walter Mondale over Republican Norm Coleman by 6 points; Coleman won by 3. In Colorado, Zogby picked Democrat Ted Strickland over GOP incumbent Wayne Allard by 5; Allard won by 5. In Georgia, Zogby picked Democrat Max Cleland over Republican Saxby Chambliss by 2; Chambliss won by 7. In Texas, Zogby's final poll had Republican John Cornyn over Democrat Ron Kirk by 4 points; Cornyn won by 12. Zogby's final poll in the Florida gubernatorial race had Jeb Bush winning by 15, but only three weeks earlier he had Bush winning by only 3. Bush won by 13 points.

In 2002, Zogby got the winner wrong more often than any other pollster--but those results were quickly forgotten by the media, and Zogby was able to keep dining out on his reputation as the Polling Hero of 1996 and 2000.

In 2004, Zogby became the pollster of choice for Democrats, as he always had some morsel of good news for them. In May, he boldly declared the race was Kerry's to lose, touting the myth that Kerry is a "good closer." Month after month, Zogby kept finding reassuring news for Kerry that somehow eluded other pollsters. He conducted dubious Internet-based polls of self-selected web users, and found all kinds of good news for Democrats: leads for Kerry in Arkansas, Tennessee, Missouri, Nevada, and Florida outside the margin of error. "Don't expect any real bounce after the Republicans convene in New York later this month," he wrote in early August.

The public should have been cautious about Zogby's polling results, because even this year there was a big red dashboard light flashing a warning about Zogby's sampling: His May 19-20 telephone poll before a South Dakota special election for the House of Representatives had Democrat Stephanie Herseth leading Republican Larry Diedrich 52 percent to 41 percent. Actual results: Herseth 51, Diedrich 49.

But Tina Brown reported that the Manhattan chardonnay-and-brie crowd was abuzz over Kerry's standing in Zogby, and by fall, she was gleefully passing on any tidbit: "In October 2000, pollster John Zogby asked real voters this question, set in the Land of Oz: For president, would you vote for The Tin Man or The Scarecrow? The result: an exact tie, 46 percent to 46 percent. But this October, The Tin Man has 40 percent to The Scarecrow's 10 percent." Call off the election! Bush is losing the metaphor!

Not only did the Left insist that Zogby's good-news-for-Kerry numbers were right; all contrary results had to be wrong. This election season saw the loudest and angriest voices on the left accusing mainstream pollsters like Gallup of engaging in a vast conspiracy to understate Kerry's support. MoveOn.org, the liberal advocacy group, purchased a full-page ad in the New York Times denouncing Gallup's methodology as over-sampling Republicans and warning that George Gallup Jr., the son of the poll's founder, is a "devout evangelical Christian."

Late afternoon on Election Day--awfully late for a final call--Zogby predicted that Kerry would win Florida, Ohio, Iowa, and New Mexico (0 for 4!) and get at least 311 votes in the Electoral College, while Bush was assured of only 213. (The remaining 14 electoral votes were too close to call.)

There's no other way to say it: The Big Z's final polls were garbage. His final poll had Colorado too close to call; Bush won by 7 points. He had Florida by a tenth of a percentage point for Kerry and "trending Kerry"; Bush won by 5 points. Zogby had Bush winning North Carolina by 3; the president won John Edwards's home state by 13. Zogby had Bush leading Tennessee by 4; the president won by 14. Zogby called Virginia a "slight edge" for the GOP; Bush won by 8. In West Virginia, Zogby predicted a Bush win by 4; the president won by 13. And in the vital swing state of Wisconsin, Zogby had Kerry up by 6; the final margin was 1 point.

Those Zogby numbers aren't a with-in-the-margin-of-error measurement of the voting public. They are a DNC wish list.

Audaciously, Zogby bragged on his website after Election Day that a New York Post headline had said Zogby "Got It Right" and that the Boston Globe had called him "Largely Accurate." Zogby said his tallies reflected what happened in most states quite accurately, though he was off in Ohio and Florida.

"I do think we need to have realistic expectations for pollsters," he told Gannett. "We're there to say it's trending this way or that way. For anyone to expect we're going to get it right to one-tenth of a percent every time is unrealistic."

Larry Sabato, director of the Center for Politics at the University of Virginia, said he thinks highly of Zogby and has taught some of his younger family members. But he thinks Zogby may never recover his status as the pollster of record. "The Zogby mystique is dead," Sabato said. "For a private client, a pollster has lots of functions, but for the public a pollster only has to answer two questions: Who's ahead and who's gonna win. It's hard for me to believe that even partisans just want to hear good news. As Democrats look to 2006 and 2008, they're going to want the real story, even if it includes bad news."

This marks the third year in a row that many pollsters, not just Zogby, have dramatically underestimated GOP turnout. (Recall that in 2003, the final L. A. Times poll showed Arnold Schwarzenegger with 40 percent, Cruz Bustamante with 32 percent. Final result? Ah-nuld 49, Bustamante 32.) This year may have yielded the first evidence of a theory long espoused in GOP circles: that polls underestimate the strength of Republican candidates because conservative voters are more likely to tell a phoning pollster to buzz off. Joe Lenski and Warren Mitofsky, who managed the horrifically wrong Election Day exit polls, theorized that their results vastly overstated support for Kerry because Democrats were more willing to talk to pollsters after voting.

There is no joy in Leftville, as Mighty Zogby has struck out.