Yesterday we reported on recent Senate trends that looked favorable for Republicans, but new just-released polling shows a trio of races producing positive Democratic numbers. Within the last week Republican candidates were pulling ahead in Indiana, Massachusetts and Wisconsin. Now Democratic sources say those races have abruptly turned around.
In the Hoosier State, the Global Strategy Group, polling for the Joe Donnelly campaign (Sept. 8-10; 800 likely Indiana voters), reports that their candidate is leading Republican nominee Richard Mourdock by a 45-42 percent count.
Western New England University (Sept. 6-13; 545 registered Massachusetts voters), which has previously polled the Massachusetts Senate race and posted Sen. Scott Brown to a lead, now shows challenger Elizabeth Warren to be opening up a six-point, 50-44 percent advantage. This poll has a small sample and a long interview period, both negative reliability factors. Additionally, Public Policy Polling (Sept. 13-16; 876 likely Massachusetts voters) also puts Warren ahead of Brown, but by a smaller, and probably more realistic, 48-46 percent margin.
Turning to the Badger State of Wisconsin, where all post-August primary polls have shown former governor Tommy Thompson to be enjoying leads of varying sizes over Rep. Tammy Baldwin (D-WI-2), the Democrats are producing new surveys touting a different result. The Feldman Group for the Baldwin campaign (Sept. 9-12; 800 likely Wisconsin voters) gives their candidate a 50-45 percent lead over the Republican nominee. Public Policy Polling, conducting a survey for the liberal organization Democracy for America (Sept. 12-13; 959 likely Wisconsin voters) gives Baldwin a three-point, 48-45 percent slight edge. This poll also shows President Obama clinging to the barest of leads over Mitt Romney in Wisconsin, a 49-48 percent count.
The campaigns are fluid, so these snap-shot numbers could be accurate, at least in the short term. The Indiana data is internally sourced from the Democratic campaign, which is always suspect without viewing the entire questionnaire. On the other hand, the Mourdock campaign has yet to release countering data of its own. The Western New England University poll likely has a large error factor and should be discounted. The Wisconsin studies could have validity because Thompson has not yet countered two solid weeks of heavy negative advertising against him. Expect more twists and turns in all of these races before November arrives.