The latest survey from Susquehanna Polling and Research shows the smallest margin yet of the race for U.S. Senate in Pa. Republican Tom Smith trails Sen. Bob Casey by just 2 points, 46 percent to 44, and the race is within the poll’s margin of error.
That’s close to the 45 percent to 42 percent lead Casey had in mid September, but smaller than his 45 percent to 36 lead from SP&R’s July survey. Independent voters are tilting toward Smith 41 percent to 37.
“Smith has gained considerably when compared with our July survey, when Casey led Smith by a 45-36 margin – representing an 8-point gain for Smith with Casey’s ballot position virtually unchanged in 4 months of surveys since July,” the pollster wrote in the release memo.
Casey was up by double digits for most of the campaign, but his lead started to shrink a month ago when Smith began his TV blitz. Now, Casey’s lead is 8.3 percent according to an average of polls by Real Clear Politics, not including today’s release from SP&R.
Smith was alone on television in the Philadelphia media market for much of the last month. That likely contributed to his narrow 47 percent to 46 lead in Philly’s suburban counties.
SP&R did not release crosstabs, but the polling memo included a few other observations.
Smith needs to solidify his base. Casey, a pro-life, pro-gun Democrat, is pulling 19 percent of self-identified conservatives and is outperforming President Obama by 12 points in the conservative “T” of central Pa.
The poll shows Smith ahead in Allegheny County, near his southwest Pa. geographic base, 54 percent to 44. Meanwhile, Casey leads in northeast Pa. and Philadelphia by wide margins.
SP&R surveyed 725 likely Pa. voters from Oct. 4-6 via live telephone interviews (defined as likely by their having voted in at least one the general election in 2008, 2009, 2010 or 2011 and new voters registered since November 2011). The margin of error is plus or minus 3.7 percent.
This firm’s polls have consistently shown Republicans Smith and Mitt Romney performing above their average in other surveys, something SP&R pollster James Lee attributes to differing sample selection methods.
“Interviews are closely monitored to ensure a representative sample of Pennsylvania’s electorate is achieved based on party, geography, gender, age and other demographics; results are sometimes statistically weighted,” the firm wrote in its description of methodology.