Numbers may not lie, but they don't always tell the whole story, either.
When former City Commissioner Charlie Hales finished first in May's mayoral primary, he seemed to be sitting in a strong position.
He finished with 37 percent of the vote, ahead of his closest competitor, state Rep. Jefferson Smith (D-East Portland), who finished with just shy of 33 percent.
But WW's analysis of the primary vote shows Hales' first-place finish leaves him without any meaningful electoral advantage for his fall runoff with Smith.
And it all has to do with those who didn't bother to vote in May but who will probably find a stamp for their ballot by November.
The precincts Hales won had higher-than-average turnouts. Smith's strongest precincts had lower turnouts—but tend to vote more heavily in fall elections.
In other words, voters who will decide this race didn't cast their ballots in May. And the parts of the city most likely to surge in turnout in November are full of Smith voters.
The analysis suggests Smith will have a hard time continuing to credibly play the underdog—a role that seems to have served him well so far.
The primary electoral terrain shows a clear divide. Hales won every precinct west of the Willamette River, and the higher-income neighborhoods of Laurelhurst, Irvington and Alameda. Smith won a large swath of his eastside turf.
"If you look at the upper-income areas, Hales carried those," says Portland pollster Tim Hibbitts, a partner in DHM Research. "It had something more to do with socioeconomic status and less to do with east-west."
Hibbitts says upper-income voters tend to turn out more reliably in primaries. That would explain the higher turnout in precincts Hales won (45 percent on average) compared to Smith's (38 percent). Citywide, turnout was 40 percent.
But WW's analysis shows that Smith's precincts will come close to matching the turnout of precincts Hales won in the fall, if 2004 and 2008 are any guide.
Precincts Hales won have 10,000 fewer total potential votes than those won by Smith, whose support came from areas with more young, middle- and lower-income voters.
The 135,000 ballots cast in the mayoral primary—if recent history is reliable—will double in the general election, with a bigger proportion of new votes coming from precincts where Smith's performance was strongest.
All of which means, Hibbitts says, "If everything holds equally—yes, it will be a benefit to Smith."
WW's analysis doesn't dare predict the outcome. A lot of things can and will happen between now and Nov. 6: unflattering news coverage, flip-flops on issues, more staff shake-ups or even natural disasters.
At a bare minimum, though, WW's analysis shows that Smith's chances of winning increase along with overall turnout. The analysis shows Smith strongest when looking at 2008 as a model, compared to 2004.
Four years ago, Barack Obama's candidacy drew an unprecedented number of young voters and drove Multnomah County turnout to 86 percent. Hibbitts predicts 80 percent for 2012—more typical for a presidential election.
"I want to at least win one precinct on the west side, but not to aim at a particular place," Smith says. "The objective we set for our campaign is to build...the strongest grassroots campaign in the history of the city."
Both Hales' and Smith's campaigns hit the streets in the primary, but there is evidence Smith's ground game made a meaningful difference.
On primary election night, the early returns showed Smith 9 percentage points behind Hales. As the night wore on, the gap narrowed to less than 5 points. These later ballots are ones that came in during the campaign's final days—suggesting Smith's last-minute efforts to get out the vote had an impact.
For his part, Hales says he doesn't know why the vote appears to have broken between precincts with higher household incomes and those with lower.
"I won precincts in every part of the city," Hales says. "It happens that I won all of them on the west side—including in [Smith's] legislative district."
One question remains: Where will voters who supported Eileen Brady go in the general election?
Brady, who finished a distant third, won about 29,000 votes in the primary, or about 21 percent. If, as WW's analysis assumes, those voters divide proportionally between Smith and Hales, the effect could be a wash.
"I don't anticipate any candidate getting that block vote," Hibbitts says. "Even if Brady endorses, I donât think theyâre going to go en masse.â
FACT: In May, Smith won his home precinct in Hazelwood with 39 percent. He finished behind Brady in Eastmoreland, Hales' turf, where Hales won with 52 percent.