One of the most positive things about the rise of the blogosphere, IMHO, is the rise of what might be called "applied political science": ordinary people (as well as political professionals) applying actual data to political trends, rather than merely opining, with the goal of affecting actual elections or policies, rather than merely for the sake of knowing things.  A good example can be found at Swing State Project recently, concerning the success rate of Congressional candidates on their second try for office:

Unfortunately, there wasn't a particularly high success ratio: of the 14 races that were considered competitive where the Democratic candidate was making a second run, only three four made it over the finish line (Dan Maffei, Eric Massa, Mary Jo Kilroy, and Larry Kissell). These candidates seemed to benefit from a perfect storm of traction from a repeat run, and running against weakened opponents (a different opponent for Maffei in the wake of Jim Walsh's retirement and a bungled GOP recruitment, and befuddled, unlikable opponents for Massa and Kissell). (On the following table, * indicates a different opponent in 2008.)

The others seemed to falter, either in the face of a nutty GOPer but too red a district (Brown, Wulsin, probably Esch) or an uncontroversial 'moderate' incumbent with a strong hold on a suburban district (Seals, Burner, Feder). By contrast, because of the confluence of swing districts and craptacular opponents, NY-25, NY-29, and NC-08 seem like races we likely could have won with or without a returning opponent (although the prospect of a Maffei rematch may have caused Walsh's retirement)... which isn't to say that we should avoid rematches, simply that it may not provide as much of an advantage as conventional wisdom currently holds.

Also, I can't help but notice one troubling pattern: the male reruns improved on their 2006 numbers. The female reruns declined. If you look at the names above from the 2004-06 cycles, you see the same pattern (Nancy Boyda excepted). I won't attempt to psychoanalyze that, but it's disappointing nonetheless.

It's worth reading in full.  I would love to see some of these guys get in touch with the more academic side of the poli-sci blogosphere (e.g. The Monkey Cage or Crooked Timber)--I suspect there might be some reinvention of the wheel, and even if not I suspect there's a lot of commonality that could be profitable to both groups.