How the Primaries Rewrote the Script: Lessons for the Campaign Ahead

It was going to be short and simple: Hillary Clinton vs. Rudy Giuliani.

Those were the long-ago and far-away days of initial preferences, when the two best-known candidates held commanding leads for their parties' presidential nominations. That it didn't end that way underscores an eternal truth of American politics: Campaigns matter. Now, with the primaries at last over, deciphering how they played out can provide invaluable intelligence on what comes next.

Exit poll results are clear: Sen. Hillary Clinton, D-N.Y., lost the nomination because her party's voters wanted change. And former New York Mayor Rudy Giuliani, R-N.Y., lost it because his party didn't.

Beyond those realities lie a wealth of data that both rebut conventional wisdom and give the presumptive nominees, Sen. Barack Obama and Sen. John McCain, a blueprint of bridges to build, strengths to promote and shortcomings to address.

Obama cornered the market on change: Across all Democratic primaries, 50 percent of voters called it the single most important candidate attribute, more than twice as many as picked No. 2, experience. Change voters favored Obama by a vast margin, 68-29 percent.

Change, though, is a double-edged sword -- it can be good or bad. In national ABC/Post polling, while Democrats clearly prefer "new ideas and a new direction" over strength and experience, the public overall divides more evenly -- with a slight tilt toward the stability of experience rather than the chance of change.

Indeed on the Republican side it was experience that worked for McCain -- he won voters who cared most about it by a 61-27 percent margin, his best vote-winning attribute.

Yet McCain faces challenges in the core of his party. "Experience" wasn't the most attractive candidate quality to GOP voters; cited by 24 percent, it finished a distant second to shared values, cited by 44 percent. And among values voters McCain finished only third, behind former Govs. Mike Huckabee, R-Ark., and Mitt Romney, R-Mass., alike. Similarly, McCain was weak among "very" conservative voters, evangelical Christians and opponents of legal abortion -- overlapping groups at the heart of the Republican constituency. While large numbers of them are unlikely to vote Democratic, McCain needs to motivate them to vote at all.

If appealing to conservatism is a task for McCain, it was an insurmountable one for Giuliani. In all 29 Republican primaries in which exit polls were conducted, a remarkable 65 percent of voters on average described themselves as conservatives, more than in exit polls since 1976. When it came time to vote, a pro-choice, pro-gun-control, sometime pro-gay-civil-unions Republican was not their cup of tea. Triangulated by McCain (more appealing to moderate and the few liberal Republicans), Romney (conservatives) and Huckabee (evangelicals), Giuliani was gone before January was out.

Having prevailed in his primaries -- not least because Huckabee and Romney continued to split their core constituencies -- McCain's task ahead involves some intricacies. He needs to reassure and energize the GOP base (which still broadly supports George W. Bush) without alienating the vital and somewhat grouchy center (which clearly does not). He needs to address concerns about his age. And he likely needs to enunciate a rationale for his candidacy beyond the "steady hand" that has so far served him well.

Obama's task if anything is trickier. His well-documented weakness in the primaries among older and less well-off voters underscores the potential hazard of a campaign built more on aspirational themes than on an image of nuts-and-bolts problem-solving. He lost working-class white voters (those without college degrees) to Clinton by a 31-point margin, 62-31 percent, and white seniors by 63-29 percent. Seniors, in particular, are much less open to change than their younger counterparts.

There also was substantial affinity-group voting in the Democratic primaries. White women went to Clinton by a 60-34 percent margin; African-Americans to Obama by 82-15 percent. White men in many ways were swing voters; overall they went to Clinton by a narrow 48-45 percent, but with enormous state-to-state differences. In Kentucky, Obama lost white men by 70-26 percent, underscoring his particular problem with Southern white men. On the same day, in Oregon, he won them by a 2-1 margin, 66-33 percent.

Overall, Obama won 10 of the 13 states where he won white men, and lost 13 of the 19 states where he lost them (the size of the black population was a factor). And white men are another group in which age and socioeconomic status also played a role, with better-educated and younger white men better for Obama, less-educated and older white men more favorably inclined toward Clinton.

Obama also displayed weakness in other groups, notably Hispanics, losing them by 61-35 percent, and Catholics, by an identical 61-35 percent. White Catholics, along with political independents, are a classic swing voter group in presidential general elections -- a group whose allegiance swings between parties, and one that's big enough to make a difference. And Obama, while well short among white Catholics, won independents by 12 points, 52-40 percent. Both are worth watching closely in the months ahead.

Other groups bear watching as well. Obama did notably well among change-attuned young voters, winning Democrats under age 30 by a 20-point margin, 58-38 percent. But young adults tend to be the least reliable voters; their turnout is always a question. This year they averaged 14 percent of Democratic primary voters, a larger share than in 2000 or 2004 but roughly matching, not exceeding, their 1984-92 levels.

McCain, in his primaries, did notably well among seniors, a more reliable voter group in terms of turnout, winning them by 47-26 percent over Romney. (While McCain's own age is a net negative, its flipside, experience, helps him; watch for him to try to turn the tables by painting Obama as callow.)

Like young adults in the Democratic primaries, African-Americans, though a natural affinity group for Obama, did not increase their turnout overall. They accounted for 19 percent of Democratic primary voters on average, about the same as in years past. (In a notable difference between the parties, blacks accounted for an average of just 2 percent of Republican voters.)

Women are crucial as well; a more Democratic group than men, they accounted for an average 57 percent of Democratic primary voters, vs. just 46 percent of Republican primary voters. (And white men, 47 percent of Republican voters, made up just 28 percent of Democratic voters.)

A key question, one that's been heavily debated, is whether Clinton's supporters would rally to Obama as the nominee, defect to McCain or, perhaps more likely, sit it out. The question's a fair one, but it's too soon to answer. As we've reported previously, a look at polling data since 1988 shows that, in the heat of their primary battles, supporters of a losing nominee routinely have been loathe to say they'd vote for the winner in November. Whether those wounds heal relies on time, and the nature of the endgame.

The potential effect of racism is another wildcard. In the last national ABC/Post poll, 12 percent of Americans said they were uncomfortable with the idea of an African-American president, 6 percent, "very uncomfortable." (Far more, 39 percent, were uncomfortable with the idea of a 72-year-old first-time president.) In another measure, in the eight Democratic primaries for which we have data, 12 percent of voters were whites who described race as an important factor in their vote, and half of them said they would not support Obama as the nominee in November. Whether and where that holds, and whether he compensates with other voters, also need monitoring.

What should matter most in the general election is the red meat of issues. They count for less in the primaries because candidates in the same party tend not to be deeply differentiated on basic issues (Giuliani was an exception). With the primaries done, watch for issues -- the economy and the war in Iraq chief among them -- to come to the fore.

With the war (which McCain supports) unpopular, and the economy and broader outlook deeply negative, some analysts have been tempted to cast the general election as a slam-dunk for the Democrats. That's not the case; depending on the poll, Obama leads McCain narrowly or not at all, and the two divide public preferences on issues and attributes alike.

There are two fundamental reasons. One is that this is more a center-right than a center-left country. On average this year 34 percent of Americans have described themselves as conservatives, vs. 23 liberals. (What closes the gap is that moderates are more closely aligned with the Democratic Party.) The other is a generation-long trend in political party affiliation, which, despite ups and downs, has been in the Republicans' favor. On average in 1981, when ABC News started polling, Americans were 13 points more apt to call themselves Democrats than Republicans. So far this year, it's been 9 points.

The Republicans did even better -- absolute parity with the Democrats -- in 2003, before public views of the war and then the economy went sour, prompting an exodus from the president and his party. That leaves us with the last wildcard of 2008, George W. Bush. If the Democrats make the election a referendum on his presidency, they gain a vast advantage. If, instead, it's a post-Bush election, all bets are off.