As Election Day nears, polls all over map

If you’re a fan of public polls, you had a 100 percent chance of being confused Tuesday, as a bevy of surveys showing different results came out.

Public Policy Polling, which uses an auto-dialing computer to interview people, found Obama with the most comfortable margin in Nevada, 52 percent to Romney’s 43 percent. The margin of error was 4 percentage points.

The American Research Group was similar. It found a 51 percent to 44 percent spread in the race, with a 4 percentage-point margin of error.

But the Nevada Retail Association, which retains Glen Bolger’s Public Opinion Strategies to do its polling, found the race tied, at 46 percent each for Obama and Romney. The margin of error was 4.4 percentage points.

And the Review-Journal and 8NewsNow, in a poll conducted by SurveyUSA last month, found the race almost tied, with Obama at 47 percent and Romney at 45 percent.

So who’s right? Who knows? But the real answer is, who cares? No poll taken more than a week or so before Election Day is useful in predicting the outcome of a particular vote. All polls can really measure at this point is public attitudes toward candidates at a fixed point in time.

Polling is obviously an inexact science that depends a great deal on who is surveyed, with accuracy going to the pollster who mostly closely approximates actual Election Day turnout. In 2010, Mark Mellman did that well, estimating a win for Sen. Harry Reid by 6 percentage points.

In the end, Reid beat Republican Sharron Angle by 5.74 percent.

Almost all other public polls that year – including those in the Review-Journal – showed Angle ahead. And they were all wrong.

Useful pieces of information can be gleaned from polls, however. Bolger’s survey for the retail association contains an important look at a key public attitude: Is the country headed in the right direction, or is it on the wrong track?

In September 2010, two months before Republicans recaptured the House of Representatives from Democrats in the first mid-term election of President Obama’s term, 17 percent of those polled in Nevada said the country was going in the right direction, while 76 percent said it was on the wrong track.

The next thing you knew, in what was then Nevada’s only really competitive House district, Dr. Joe Heck was adding “representative” to his title while former Rep. Dina Titus was updating her resume.

But this time around, the voters are less apprehensive: 37 percent say the country is headed in the right direction, while 54 percent say it’s on the wrong track, Bolger’s survey shows.

Adman Billy Vassiliadis has watched Nevada politics for years, and he confidently predicts an Obama victory by around 6 percentage points. (The voter registration numbers in the state are slightly better for Democrats than they were two years ago when Reid won.) Those numbers augur well for Democratic candidates such as Steven Horsford in the new 4th Congressional District, Vassiliadis said.

In the presidential race, it comes down to personal style: “There’s a likability factor that’s just not there” for Romney, Vassiliadis says.

But what about the Shelley Berkley-Dean Heller Senate race, which has turned into a nasty contest as candidates trade charges of corruption? That’s a tougher call, even with an expected Obama victory. (The retail association and Public Policy Polling showed widely varying results, as do internal Democratic polls.)

One thing is certain: Nobody knows for sure right now. Not even the pollsters.

Steve Sebelius is a Review-Journal political columnist and author of the blog SlashPolitics.com. Follow him on Twitter (@SteveSebelius) or reach him at (702) 387-5276 or ssebelius@reviewjournal.com.