I will soon join the ranks of the prognosticators, attempting to predict for you the course of the presidential nominating process in both parties. Before leaping into that void, however, I want to remind everyone, including myself, just how difficult it can be to read a whole novel into the few tealeaves available this year.
Among the leaves at the bottom of the cup are national polls. At this early stage, they reveal precious little. Four years ago this month, the ABC/Washington Post poll identified Joe Lieberman as the frontrunner, with 27 percent of the Democratic primary vote. Dick Gephardt held second place with 14 percent, John Edwards was in third and John KerryJohn KerryOvernight Tech: Meet the key players for Trump on tech | Patent chief staying on | Kerry aide goes to Snapchat | Uber's M settlement Tillerson met with top State official: report The Hill's 12:30 Report MORE occupied fourth, commanding just 10 percent. Those results were no anomaly. Over the next eight months, 35 national polls had Lieberman either in the lead or tied for first.
As campaigns intensified in the fall, the national polls painted a different, but equally inaccurate picture of the eventual result. In early September, the CNN/Gallup/USA Today poll put Dick Gephardt in first with 16 percent and Howard Dean in second with 14 percent. Kerry and Lieberman were tied for third. By November, this same poll showed Dean moving into first place, 2 points ahead of Lieberman, while Kerry languished in fifth. By December, every poll had Dean in first place and all but two had either Lieberman or Wesley Clark in second.
Why do they keep taking these polls? More to the point, why does anyone else pay attention to them?
Sophisticated prognosticators glance only briefly at the early national polls, focusing instead on the early states. First out of the box in Iowa four years ago was Zogby, who had Gephardt leading with Lieberman close behind. Through June, all but one public poll of purported Iowa caucus-goers put Dick Gephardt in first place.
By December, things had changed, though the results were no more predictive of the outcome. At year-end, Zogby had Dean at 26 percent in Iowa, Gephardt at 22 percent and everyone else in single digits.
Polls aren’t the only indicator. Foraging among the tealeaves inevitably leads prognosticators to fundraising. In not one quarter of 2003 did John Kerry lead in fundraising. John Edwards took that title for the first quarter, Howard Dean in each of the subsequent three.
Well, if national polls, state polls and fundraising don’t tell the tale of the tape, surely savvy Washington insiders grasp the underlying dynamics. National Journal surveyed 50 Democratic insiders in January 2004, just weeks before the caucuses. Forty-three of them picked Howard Dean as the most likely nominee, three said Gephardt and two each predicted Gephardt and Clark. Every single insider proved to be wrong.
With more ink spilled over Iowa and New Hampshire than any other story in politics, one could hope that expert reporters might skillfully sift through the signs to forecast the outcome. Just days before the 2004 caucuses the New York Times reported, “Not even the biggest boosters of Mr. Edwards or Mr. Kerry are saying the two have much of a chance of winning the caucuses here, given the overwhelming organizational strength of both Dr. Dean and Representative Richard A. Gephardt.”
A few days later Kerry won with 38 percent; Edwards garnered 32 percent. “Frontrunners” Dean (18 percent) and Gephardt (11 percent) were left in the dust.
By focusing on 2004, I’ve made the test more difficult than it usually is. In truth, most of the time, the frontrunner during the “invisible primary” wins the nomination. Unfortunately, the tealeaves cannot always tell the prognosticators which will be the exceptions; sometimes you have to read the whole story to find out how it ends.
Mellman is president of The Mellman Group and has worked for Democratic candidates and causes since 1982, including Sen. John Kerry (D-Mass.) in 2004.