Pollster Brad Coker was all smiles after the election. He had just "nailed" 30 Senate and gubernatorial races throughout the country.
But what delighted the 31-year-old Columbia resident most, perhaps, is that two days earlier, he had predicted that Republican Charles I. Ecker would upset County Executive M. Elizabeth Bobo by 300 votes.
Election night, Ecker won by 244. Absentee ballots still uncounted could change the outcome, but Coker doubted it. Those voters are usually "upscale" and tend to go "against the incumbent," he said.
Right again. After absentee votes were counted, Ecker won by 450 votes.
Coker was clearly strutting his stuff.
Three months earlier he had tipped a reporter off that his polls showed Executive Sid Kramer would lose to Neal Potter in Montgomery County. "And when that happens, it will send a message throughout the state that incumbents can be beaten," he said at the time. "You're gonna see some big surprises."
In Howard County, "Ecker will win the 2nd and the 5th (council) districts, and Bobo will win the 3rd and the 4th," Coker predicted in August.
His other August predictions were that Republican Darrel Drown would upset incumbent Angela Beltram in the 2nd Council District and that Republican Marty Madden would wrest a delegate seat from the incumbents in District 13B. Coker also said Republican Christopher J. McCabe would "run well" against Democratic state Sen. Edward J. Kasemeyer.
Election night happened just as Coker had said it would -- except that McCabe ran so well against Kasemeyer that he won. Coker had told McCabe on Election Day that he would lose. It was the pollster's only miss.
"I'd rather tell somebody they're in trouble and have them pull it out than the other way around," he said.
When he first came to Columbia eight years ago, Coker was a member of the Young Democrats and ran for and lost a seat on the local Democratic Central Committee.
Since then, his Mason-Dixon Opinion Research firm in Columbia has gained a reputation for accuracy, but Coker's Democratic colleagues never came calling. Or when they did, they expected him to do things "for free," just like when he was a Young Democrat, he said.
It was the local Republican Central Committee that showed up with money in hand this year -- although it was still "half" of what he usually gets, Coker said. He accepted because of the "fun" of doing a local political campaign.
Ecker's race tickled him most.
"In the last eight years, I've polled, 800, 900, 1,000 races of all kinds for both parties all over the country. I've seen other races in which leads larger (than Bobo's) changed by 30 points. Chuck's trouble was convincing people he could win."
Accordingly, Coker met privately in June with some of Ecker's "potential contributors" to let them know his findings.
Although Ecker trailed Bobo by 18 points, she showed herself vulnerable because only 45 percent of the electorate said then that they would vote for her, Coker said. Also, her negatives were 25 percent, which if "shaken up," could get a lot higher than that, he said.
Contributors were wary, but Ecker wasn't. He lent his campaign $30,000.
What Ecker needed first, Coker said, was name recognition. Running in the primary against businessman Gilbert E. South, the party's standard-bearer for county executive in 1986, was "the best thing that could have happened to Chuck Ecker," Coker said. "It built his name recognition to the point where he could attack, and it allowed the voters to see he was a credible candidate."
Finding out where to attack was Coker's job. His usual set of polling questions was followed by a set of "repairing" questions to determine what attacks moved the most voters.
The repairing questions, a standard technique in market research, worked like this. If a voter indicated a preference for Bobo, the voter was then asked a series of "Did you know?" questions about Bobo administration spending, Bobo's stance on growth prior to her becoming executive, and her campaign contributions from developers. Voters were then asked in light of the repairing information, if they would still vote for Bobo.
Information gleaned from those questions formed the basis of Ecker's newspaper ads and cable television commercials, Coker said. "Every little thing had a role. It clicked. And what you got was a pretty sweeping outcome," he said.
Democrats in the past were fortunate enough to have strong candidates to pull them through, Coker said, but they "are always one step behind the curve" in terms of tactics.
Four years ago, "They did the same thing they had always done and had only one (incumbent) picked off, so this year they were out waving at cars." Four years hence, they will be using television commercials and pollsters just like the Republicans did this time, he predicts.