Wednesday, December 3, 2014

Denver Post Opinion: The Price of Popularity

In a Denver Post November 30 commentary, the impact of the 2014 battle for control of the U.S. Senate is analyzed.


  • The senate election in Colorado, the quintessential battleground state, was nationalized. Obama’s popularity and partisanship defined the race. “There are no public land issues debated.”
  • The proliferation of polls marshalled into powerful forecasts weeks before Election Day created the narrative of winners and losers.
  • More than a hundred million dollars flooded into the state, mostly from sources independent from the senate campaigns and it dominated the media messages voters saw.
  • Candidates became more isolated as strategies and messages were nationalized and the main task was to avoid mistakes.
The Price of Popularity – Colorado’s New Politics

The heat of the 2014 midterm elections has barely cooled, and voters’ ears are still ringing from the harangue of political ads. But how many of those appeals to voters dealt with issues specific to Colorado? Former House Speaker Tip O’Neill coined the long-used maxim, “all politics are local.” Recent midterm elections have trended away from that dictum, but the 2014 midterms may have buried it — at least in Colorado. The most distinguishing characteristic of its two senate candidates, Cory Gardner and Mark Udall, was their red or blue identity. Their biographies were known, but not salient. The most important issues were the president’s popularity (or lack of) and the control of the U.S. Senate. Gone are the quaint days when candidates debated public land issues, building a water project or funding the interstate highways.

Voters saw the beginning of this phenomenon in the 2008 presidential race when Colorado gained its reputation as a swing state and platform for presidential politics and played host to the Democratic Convention. The success of Senator Michael Bennet despite the 2010 Republican wave, highlighted the state’s importance. Colorado’s new national political standing grew in the 2012 presidential election when the state was fully targeted and featured on the maps of CNN, NBC and Fox and others. After decades of being a flyover state, Colorado hosted a presidential debate and saw repeated visits from the candidates, especially President Obama targeting Millennial and minority voters. Millions of dollars in TV advertising were placed by the campaigns, and frequent polls showed the oscillation between Obama, Romney and, finally, Obama. The state became a key platform for the national analysts and campaigns waging war in “battlegrounds.”

This year, the attention was on party control of the U.S. Senate, and those races became marque players in the national narrative. National Journal/CNN analyst Ron Brownstein describes our federal elections as evolving into a type of parliamentary system where the candidates and local issues take a backseat to the national outcome – in this case, control of the Senate.

This nationalization of the races greatly helped the Republicans this year. Mark Udall and his equally vulnerable colleagues in states such as Arkansas, North Carolina and Louisiana wanted to run local races on their own record as incumbents. But they were rapidly swept into the national drama of the Harry Reid vs. Mitch McConnell show and concentrated on how to avoid being seen or associated with President Obama.

Throwing gas on the fire of this nationalization was the proliferation of polling and the introduction of multiple competing forecasting models. From Labor Day to Election Day, Colorado saw a record number of more than 25 published polls on the U.S. Senate race with a rush toward the end. Nate Silver, the main proselytizer of the forecasting model that gained acclaim in the 2012 presidential race (his branded website 538), helped spawn more than a half dozen models this year hosted by the New York Times, Washington Post, and liberal blogs Huffington Post, Daily Kos and others.

Although there continues to be a debate about whether polls directly affect voters, clearly they affect the reporting of a race and can alter the behavior of the campaigns — in this case, to Mark Udall’s detriment. The forecasting models, used for the first time in the 2014 “control of the senate” storyline, blend local historical data, such as who won recent presidential elections, with candidate information, such as incumbency. Before Labor Day, there were few polls, but Colorado forecasts favored Udall – an incumbent in an Obama state. After Labor Day the forecasts were dominated by poll results and suddenly Udall was behind. That was mostly a product of the polls using their likely voter models and the uptick in advertising for GOP candidates.

During the last thirty days of the campaign, the forecasts showed Cory Gardner winning by a couple of points, with probabilities of more than 70 to 80 percent. The Udall campaign was tossed on the defensive. The campaign made public its own polls that showed Udall ahead and maintained the Bannock Street Project ground game would prove the forecasts wrong.

The massive outside money pouring into the Colorado senate battle was also a product of the nationalization of the race, and it harmed the incumbent. Of the money spent, the candidates expended barely one-third with Udall ahead. But the TV time purchases of both campaigns were swamped by the dollars from independent expenditures, namely Karl Rove and the Koch brothers for the Republicans, and the Harry Reid, Planned Parenthood and Tom Steyer committees for the Democrats. In addition, Udall’s “war on women” theme was further amplified by supportive independent spending making it even more overbearing.

The nationalization of the campaigns also contributed to the isolation of the candidates. As the TV advertising became dominant and the issues limited, the candidates’ main task was not to make a mistake. The lesson learned from the 2010 and 2012 cycles, especially for Republicans, was to avoid all social issue discussions and assume you are on the record and potentially viral at every moment. Colorado U.S. Senate candidates became hard to find.

This is the new Colorado election trend. Since this election re-affirmed Colorado’s battleground status, the state can expect more of the same for the 2016 elections when Senator Michael Bennet will be up for re-election and the country will pick a new president.  What can we expect? A couple of national themes, mostly around support and opposition to President Obama; massive money, mostly from independent sources; nonstop television advertising and starting early; forecast polling with predictions on a daily basis; and candidates in controlled, mostly artificial environments. Colorado is popular and this is the price. Can’t wait.

Floyd Ciruli, founder of Ciruli Associates, is a Denver-based pollster and director of the University of Denver’s Crossley Center for Public Opinion Research.

No comments: