This week Rasmussen Reports dropped two polls of Wisconsin, both conducted on Aug. 15. Both polls had very good news for Republicans. One showed Mitt Romney barely ahead of President Barack Obama in Wisconsin, 48-47 percent, the other showed newly minted Republican Senate candidate Tommy Thompson up 11 percentage points over Democrat Tammy Baldwin, 54-43.
The bad news for Republicans is that these polls were done by Rasmussen, which this blog has noted before tends to have results biased toward Republicans. Can this poll be believed?
Rasmussen shows Thompson outperforming his party's national candidate, Romney, by 6 points. On the surface, this isn't absurd. This means six percent of the electorate would vote for both Thompson and Obama. Thompson was popular when governor, so voters may do this. Can any corroboration for this be found in other data?
Marquette University Law School's poll conducted Aug. 2-5, which was before Tuesday's primary, asked about the same matchups Rasmussen polled. Marquette found Thompson leading Baldwin 48-43 percent and Romney trailing Obama 45-50 percent. This shows Thompson also outperforming Romney but only by 3 points.
What seems likely is Rasmussen is again predicting too much support for the GOP (Rasmussen is the only pollster showing Wisconsin this close in the presidential race) but that Thompson will outperform Romney in the state.
Perhaps more concerning for Baldwin partisans is that both polls show her underperforming the president by 6 points. It's quite possible that Baldwin can win while polling worse than Obama, but she can't poll that much worse if she's going to keep this Senate seat in Democratic hands.