Matthew Yglesias asks what advice can be drawn from Red State, Blue State for people who want to understand how to influence election outcomes. As Yglesias points out, the default assumption among political scientists is that candidates and parties are usually doing what they're doing for a reason and that there probably aren't any low-hanging fruit or they would've already been grabbed.

On the other hand, if political candidates are taking their advice from Dick Morris and Mark Penn, well . . . maybe there are some useful things they could learn from our book.

Most of the free advice in our book comes in chapter 9, where we provide evidence in support of the following points. I wouldn't say we prove any of these claims, but the evidence is certainly consistent with them, and they make sense to me:

1. Given where the Republicans were positioning themselves (circa 2004), Democratic presidential candidates have been pretty well situated ideologically. Republican presidential candidates, however, could gain something like 2% of the vote by moving to the center on economic issues.

2. Moving to the center is also worth about 2% of the vote for congressional candidates. For congressmembers, though, 2% isn't so important--very few congressional races are close, nowadays.

3. Individual presidential candidates are not so important in determining the vote, and vice presidential candidates even less so. My advice here is for parties to pick the vice presidential candidate whom they think would the best president in the event that the succession were to happen; forget about the purported electoral benefits of one candidate or another.

4. The change in economic circumstances from year 3 to year 4 of the presidential term is highly predictive of the presidential election outcome. This is no secret, and so we can understand the president's economic policy choices from this perspective.

None of the above is big news (although the asymmetry of item 1 was interesting; it popped out of our analysis of 2004 National Election Study data; item 2 makes sense but is misunderstood by people who've learned just a little bit of political theory about the median voter; and items 3 and 4 appear to be unknown to Dick Morris).

I agree with Yglesias that some of our more detailed findings, while interesting and even at times cool, don't quite fall into the "news you can use" category. For example, income predicts how you vote in Mississippi more than in Connecticut. Fine. How does that affect your campaign pitch? Either way you have to do some combination of swinging voters, getting your side to turn out, and getting the other side not to vote. Similarly, I don't know what a campaign would directly do with the information that churchgoing predicts Republican voting much more among the rich than the poor. Run abortion-focused political ads on the Golf Channel rather than on wrestling shows? Maybe.

Aaron Swartz is closer to the mark, I suspect, when he suggests that political activists might make better use of my methods (as demonstrated in my books on regression modeling and Bayesian analysis) rather than the specific findings presented in Red State, Blue State.

That said, I think I can--and have--offered some potentially helpful advice. Yglasias asks, "Is there electoral advice for, say, the now-out-of-power Republicans lurking somewhere in this book?" Actually, a few months ago I wrote an article based on Red State, Blue State with advice for conservatives after the 2008 election. So, yes, I think there is some relevant advice, even if it's not quite a nuts-and-bolts recommended campaign strategy.

In writing that article, I didn't see myself as a player in a zero-sum game--offering advice to Republicans so as to help them do better against the Democrats--but rather as a participant in a larger process, making both sides more aware of public opinion. I'd like to feel that the more public-minded politicians and activists on both sides of the aisle benefit from a clearer picture. Especially if the alternative is the confused or manipulated statistics you sometimes see coming from brand-name political consultants or partisan talking heads who have every incentive to exaggerate.