Why you should ignore Carolyn Goodman's poll

A bit of controversy erupted last week in the mayor's race (I know, shit breaks out every other day) over some dueling polls. Chris Giunchigliani's campaign released a poll conducted by Beneson Strategy Group showing the race statistically tied. Carolyn Goodman's campaign was forced to release their own Magellan Strategies conducted poll--after spending the weekend telling reporters they can't release the numbers because they're so good no one would vote--that showed Carolyn up 53-31.

Freelancer Steve Friess was skeptical of Chris G's poll, airing his frustration on Twitter that Goodman's numbers weren't being reported as breathlessly as Giunchigliani's. But anyone familiar with Magellen Strategies knows they have a poor reputation. And that could explain why no one is giving Carolyn's polling numbers any credibility.

Take for example:

“Tom Tancredo will become Colorado's next governor,” predicted David Flaherty, president and CEO of Magellan Data and Mapping Strategies, a Louisville, CO, pollster that works for Republicans around the country.

That statement was made Friday before the 2010 gubernatorial election in Colorado. Democrat John Hickenlooper went on to beat Tancredo 51-36 four days later. Magellan also predicted Meg Whitman would win the governor's race in California. She lost to Democrat Jerry Brown.

Another "oopsie" in 2010 by Magellan, they had to issue an apology for their flawed polling in the Kentucky senate race. Their polls showed Republican candidate Rand Paul beating his closest competitor by 20-30 points. What they forgot to mention: the poll they conducted did not mention any candidate but Rand Paul. Rand Paul went on to win the election 55-44.

In 2010 Magellan released a flurry of poorly conducted polls, all benefiting Republican candidates. Their website at the time had the tag line: Magellan Strategies Helping Republicans Win Elections. Magellan is on the payroll of the Republican National Committee and Koch-funded astroturf conservative organization, Americans for Prosperity.

I've been told that Magellan's bad results are due to bad methodology. One person was polled by Magellan at an office number. They pollster did not ask for a specific person nor did they ask any screening questions ("Are you registered to vote in Las Vegas?") either. Magellan is simply a disaster.

I should start my own polling firm. Get paid tens of thousands of dollars to produce crap? Light Skinned Polling is open for business.

But hey, if you want to trust Magellan, be my guest.

Everyone else who lives in a place called reality, keep knocking those doors and making those calls. It's making a difference.

--
Laura