The Winston Group is a strategy and research firm dedicated to making ideas matter.

Too Many Variables Still in Play to Predict Iowa, N.H. Results

by David Winston

The Iowa caucuses are now just 23 days away, and the New Hampshire primary is one month from today. A fair number of reporters, talking heads and partisans of every stripe have focused on the latest poll numbers to try to predict the outcome of these first-in-the-nation presidential contests. This column is a cautionary tale for the political oddsmakers who may put more stock in early-state polls, especially Iowa polls, than they should.

As things stand now, if Sen. Hillary Rodham Clinton (N.Y.) wins the Iowa caucuses, the race for the Democratic nod is likely over. If she doesn’t, where the candidates win, place and show will have a huge impact on both the Democratic and GOP primaries in New Hampshire.

Until Mike Huckabee began his surprising surge over the past three weeks, the Republican race in Iowa hasn’t been much of a contest. Mitt Romney’s strategy has been to invest millions of dollars and much of his time putting together an impressive organization — he has led in the polls for months.

Rudy Giuliani and Sen. John McCain (Ariz.) have made only token efforts there while ex-Sen. Fred Thompson (Tenn.) hasn’t caught fire with social conservatives. That left an opening for Huckabee, who finally caught the eye of evangelical voters. Now the question is, can he beat Romney, and if he does or comes close, what impact will it have in New Hampshire? If Romney loses Iowa, his campaign will take a serious hit.

New Hampshire then becomes a must-win for him. McCain, who has decided not to contest Iowa, says he will finish first in the Granite State. That would give him a huge boost, but he could probably survive second place.

Giuliani, like McCain, doesn’t expect to win Iowa. His strategy is to do well enough in New Hampshire and South Carolina to be viable in the bigger, more moderate states. Without a respectable finish in one of the two states, it’s not clear how Thompson remains viable.

That’s the state of play on both sides, but picking winners in either place at this stage is risky business because so many variables can still impact the outcome of both contests. The most important of these variables is turnout.

Iowa, for both parties, is far more a test of a candidate’s organization than his or her popularity in polls. The ability to get tens of thousands of supporters to show up and spend hours at a precinct caucus on a cold winter’s night in the middle of New Year’s week isn’t easy. It requires time, money and committed volunteers, factors that polls simply can’t measure.

Pollsters attempt to “screen” out non- caucusgoers, but a little history shows how difficult that
can be. In 2000, there were about 582,510 registered Republicans in Iowa, but only approximately 87,230 attended the caucuses — five out of six stayed home that night.

Even more important, 41 percent of those who did participate were first-time caucusgoers. What that means is that many caucus attendees are candidate- and campaign-driven rather than the kind of consistent participants that characterize a primary state — only 15 percent of Republicans voters were participating for the first time in the 2000 New Hampshire primary.

That year, President Bush won the caucuses with only 35,787 votes (6.1 percent of registered Republicans), but that reflected 41 percent of the turnout. Today, there are 575,000 registered Republicans in Iowa. Theoretically, any candidate with 10 percent in the polls could win the caucuses. But identifying and turning out those 57,500 voters is easier said than done without a solid organization in place.

It’s hard to underestimate the differences between a caucus state like Iowa and a primary state like New Hampshire. While organization is always important, momentum, media coverage and ad money all play a bigger role in New Hampshire than Iowa. Equally important, while the two states may be demographically similar, Iowa and New Hampshire Republicans differ when it comes to issues.

In Iowa, one of the variables will be the turnout of evangelical Christians, who made up about 37 percent of the attendees in 2000. In New Hampshire, however, that number was only 16 percent. Social issues that matter in rural Iowa may not have the same impact in fiscally conservative New Hampshire.

There also is a certain independent streak in New Hampshire voters that shouldn’t be taken lightly. Remember McCain got only 5 percent in the 2000 Iowa caucuses. He went on to win the New Hampshire primary with 44 percent.

Another variable in New Hampshire is the role of independents who can vote in either the Republican or Democratic primary. Exit polls in 2000 found that 63 percent of those who voted in the Republican primary identified themselves as registered Republicans while 32 percent identified themselves as independents. McCain was the beneficiary, winning 61 percent of the independents’ votes.

In 2008, the independent vote is up for grabs and the outcome of the Democratic caucuses may have a major impact on the GOP primary. If Clinton wins Iowa, chances are a larger share of independents may opt to vote in the Republican primary, which would be good news for Giuliani and McCain and bad news for Romney. If Sen. Barack Obama (D-Ill.) or ex-Sen. John Edwards (N.C.) win or come close, independents may play in the Democratic primary, thus changing the possible outcome of both races.

With so many variables defining who votes and in which primary still in play, take the present polls with a grain of salt. They can give you sense of direction, but they are not written in stone.

PDF Version