Anti-Obamacare Ads Are a Shot in the Dark

What works in campaigns is anyone's guess. 
Will they be in the majority next year? Photographer: Mark Wilson/Getty Images

You won't do better than today's overview of the 2014 Senate landscape by Jennifer Duffy of the Cook Political Report. Bottom line: Republicans will win seats, but control of the Senate remains a tossup.

One quibble. Duffy says, "The majority party after this election will almost certainly have 53 or fewer seats, and quite possibly just 51 or 50 seats." I'd also emphasize the possibility that all of the close races could fall in the same direction, as Sean Trende did recently. If that happens, the majority could have 55, 56 or even 57 seats. Duffy's estimate, however, is the most likely case.

Another key point. Duffy notes that "Americans for Prosperity, a super PAC that supports GOP candidates and is funded largely by the Koch brothers, has spent millions on television ads in key Senate races attacking both Democratic incumbents and House members running for the Senate for their votes in favor of the Affordable Care Act." Even if those ads prove important -- and research generally has found that the effects of early advertising dissipate long before Election Day -- there's no way of knowing whether they were effective because of the millions spent or because they attacked the ACA. We don't get to do over the 2014 cycle so we can test the effects of different themes, such as attacks on Democrats for their record on jobs, or for that matter their position on the fictional 1975 Public Affairs Act. Republican operatives undoubtedly believe Obamacare attacks are the best use of their money, but they could be wrong. And even if they're correct, the attacks may make only a marginal difference compared with a focus on some other subject.

Scholars find it difficult to determine precisely how campaign spending affects election outcomes, even with sophisticated statistical tools and excellent data. Trying to narrow that to the impact of specific types of spending (get out the vote, TV ads) is even harder. Figuring out which ad content had an effect is a guessing game.

If you're running a campaign, guessing is the best you can do. For the rest of us, I recommend extreme caution in interpreting any claims about the usefulness of any specific ad content.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.