1. No upward movement for BHO in polls
  2. Scant improvement in registration
  3. Volunteer cadre growth slow
  4. New electorate growth difficult

Jim, with Tampa behind us, let me add emotion to your metrics. You draw the conclusions.

Keating's polls encapsulate the problem. No Obama movement here in 90 days. Still below the critical 50% threshold. At 51% among women, we're 5 points below goal with them. Contact reports are in the same range.

Compared to past Colorado August polls, we match Obama's 2008 and Bennet's 2010 numbers. On the ground, it's different.

In 2012 Republicans are holding their own. From January to August in '08, by contrast, we out-registered Republicans 5 to 1. This difference creates a problem. Here's why.


Jim, last year you laid out the registration goal for Colorado. 2010 had cost us registration-wise. We dropped 180,000, Republicans 100,000. Since January, we advanced 29,000; GOP numbers are up 26,000. Their initial advantage in actives means the January gap (113,000) isn't closing. We are pounding Republicans in contact numbers, but not earning results to match 2008. Greatly lower enthusiasm and voters who've vanished in the economic turmoil. And there are few Colorado Democrats we didn't find before 1/1/12. “Why bother?” potential registrants ask. This denies us the ammunition of victory: votes.

We won't close the gap, unlike the 65,000 surge of fall 2008 (field director reports). There is no new majority for us. Instead of a 1% overall registration gap (a la 2008) or 4% gap (a la 2010), we face a probable 6% registration gap. Factor in the GOP advantage in turnout by party. For Colorado, it was an additional 1% in 2008 and an additional 7% in 2010. Somewhere in between and we can't win Colorado.

The wild card is the results of the 600,000 multi-state registration mailing coordinated by PEW and Gessler. (That's down from the original 900,000 estimate.) That may help, despite lower interest.


Volunteers linger below '08. We do best in liberal, upscale areas (no recession for them) but are far behind in blue collar segments. In blue collar Adams County, we have not hit our stride. Tea Party chatter helps the GOP in Adams, and many voters link county corruption to Democrats.

Collectively, districts with 20% Hispanic or African-American populations are 30,000 below 2008's voting levels. Minority areas are more than five points behind merely matching 2008. We have the African-American votes, but – despite efforts over the spring and summer – we have more to do. When Mejia lost the mayoralty AND endorsed Romer, that signaled Hispanic problems. Those with ties to Mexico and those who see themselves as native Americans have different issue constellations. (If Republicans ever got smart….)

Women are our best segment, given media taking women's issues front page. My concern is that down-scale single women won't adequately turn out even if they talk to door knockers. Our lululemon women just don't win the attention of K-Mart shoppers. While women's concerns can be emphasized, they rapidly lose importance given the economy. (Fortunately, Romney hasn't played the “men and women” card – explicitly emphasizing issues where both feel short-changed.)

Despite strong Obama support, Hispanics won't have 12% of turnout, our goal. Colorado youth will be tougher to recruit and turn out. Some small counties with college outposts, like Gunnison, likely won't perform at '08 levels. The bright spot is Boulder, but we don't have Northern Colorado in the tank as we did in '08. (13,000 saw Obama this week; four years ago, it was 45,000.) Larimer GOP isn't handicapped by its leader using party funds for personal problems.

I ghosted this memo for Jen Cheyne, the Obama campaign's institutional memory in Colorado.

Jim Messina, known as a salty-tongued enforcer/fixer, is the Obama campaign head in Chicago. Last November he told Colorado reporters, “We will register more voters and turn out more voters than we did last time.”

Keating is a Colorado Democratic pollster with ties to Udall, Hickenlooper and Hancock.