Perhaps no campaign figures more prominently in the pantheon of modern political mythology than Senator Michael Bennet’s election in 2010. Mr. Bennet, a Democrat, won a come-from-behind victory in Colorado after trailing in just about every poll of the final weeks of the race.
It is hard to overstate the lore surrounding the Bennet campaign. The story leads Sasha Issenberg’s book, “The Victory Lab.” The legend holds that the Bennet campaign employed cutting-edge field methods, presaging President Obama’s re-election, to turn out enough irregular voters to mount a comeback. The Democrats named their huge 2014 turnout effort the Bannock Street Project after the road where the Bennet campaign resided for four years. In part because of this history, many assumed that Mark Udall, a Democratic incumbent, was a clean favorite to win re-election.
Instead, Colorado has emerged as a state where Democrats desperately need to do a better job of turning out voters than they did in 2010. Mr. Udall trails in the polls by an even wider margin than Mr. Bennet did four years ago, and he’ll need an even more favorable electorate to defy the odds again.
Mr. Bennet’s victory is so impressive because he won despite facing a hostile electorate — one in which most voters probably supported Mitt Romney in 2012. Mr. Bennet pulled it off in no small part because he faced Ken Buck, a candidate who alienated many moderate women with his views on rape and abortion. Mr. Bennet attacked Mr. Buck as an extremist for his views on the state’s “personhood” amendment.
How well did Mr. Bennet do among unaffiliated and Republican voters? Think about it like this: Almost 40 percent of registered voters in Colorado were Republican in 2010, but Mr. Buck managed to win only 46 percent of the vote. He suffered big defections among registered Republican voters, and was clobbered among the state’s already Democratic-leaning registered unaffiliated voters.
Mr. Udall tried and apparently failed to make similar gains with similar attacks against Cory Gardner’s past support for the state’s “personhood” amendment. By most accounts, Mr. Gardner has run a smart race. As a result, Mr. Udall does not seem poised to run so far ahead of Mr. Obama; he could easily even run behind Mr. Obama. He now needs to make up for it with a more Democratic electorate than the one that narrowly elected Mr. Bennet four years ago.
How much more Democratic would the electorate need to get for Mr. Udall to win? A very rough guess is that he needs an electorate about halfway between 2010, when Republicans outnumbered Democrats by more than six points, and 2012, when Republicans outnumbered Democrats by two points and when unaffiliated, young voters represented a much larger share of the electorate.
That rough guess is illustrated by data provided by SurveyUSA and Monmouth University, which both conducted public polls from lists of registered voters. On average, the polls found Mr. Udall ahead by five percentage points among registered unaffiliated voters, with Mr. Udall enjoying a smidgen more support among registered Democrats than Mr. Gardner did among registered Republicans.
If we assume those averages are right — and, obviously, you should not take the averages of two sets of subsamples to the bank — Mr. Gardner would win by about two percentage points if the partisan makeup of the electorate resembled 2010, but he would lose by two points if party affiliation matched 2012. Since many of the unaffiliated voters joining in a 2012-type electorate would be younger and more diverse than the ones deemed likely to vote in recent polls, the tally might underestimate how Mr. Udall would actually do with the 2012 electorate.
Halving the gap between the 2012 and 2010 electorates would not be easy. If the Udall campaign wanted to do it strictly by mobilizing more voters, it could take getting at least an additional 50,000 Democratic-leaning voters to the polls — or perhaps one-fifth of the registered Democrats voters who voted in 2012 but not in 2010, when Democrats had their celebrated get-out-the-vote effort. If the Gardner campaign could mobilize additional voters of its own, then Mr. Udall’s requirement would jump further.
Mr. Udall, however, is not just counting on brute mobilization. The Democrats have one big thing working in their favor: the advent of universal mail voting. For the first time in Colorado, all registered voters will receive a ballot in the mail — whether they request one or not.
The question is how many of the voters who didn’t participate in 2010 will choose to do so once a ballot arrives at their door. No one knows exactly what to expect, but both campaigns will surely spend the next week chasing down the voters who haven’t submitted their ballots.
Democrats are hoping that the combination of universal mail voting and their mobilization efforts will substantially enlarge the electorate, drawing from the pool of young, Democratic-leaning voters who did not participate in 2010 but did in 2012. Data from the voter registration file indicates that these voters are indeed substantially more favorable to Democrats than the 2010 voters, and a new Democratic-sponsored poll found that these voters support Mr. Udall by a 14-point margin.
If the new voters break for Mr. Udall by that margin, Democrats will need more than 250,000 voters to close a two-point deficit among voters returning from 2010. It could easily take more if Mr. Udall’s deficit is larger than two points, or if new voters are still somewhat more Republican, older, or whiter than the overall pool of voters who did not participate in 2010 but did in 2012.
Given the degree of uncertainty about eventual turnout with universal mail voting, you can start to imagine why Democrats think they still have a way to defy the odds in Colorado.
The good news for observers is that the composition of the Colorado electorate won’t stay in the realm of speculation for long. Universal mail voting means that about 85 percent of ballots will probably be tabulated by Election Day. Thanks to The Upshot’s subscription to the Colorado voter file, we’ll have a good sense of the partisan and demographic makeup of Colorado voters well before the polls close.
So far, the early vote looks pretty good for Mr. Gardner. More than 650,000 ballots had been tabulated by Sunday night, and registered Republicans outnumbered registered Democrats by 10.5 points, 42.8 to 32.3. If it holds, that will be even better for Republicans than 2010.
But it is clear that the earliest mail-in ballots are unrepresentative: Forty-three percent of early voters are 65 or over, and disproportionately from Republican jurisdictions. And buried in the data is a number that might augur well for Democrats: Twenty-one percent did not participate in 2010, and those voters are far more Democratic than those who participated in 2010. If the pace continues and everyone from 2010 returns to the polls, turnout will grow by 450,000 voters.
There’s not much question that Democrats will gain as the remaining young voters from Democratic jurisdictions mail in their ballots. The question is how much, and whether it will be enough.
