There were five races that we were tracking closely over the course of the evening -- and I've already seen analysts drawing flimsy conclusions from each of them.
Pennsylvania -- Democratic Senate primary
The results: Joe Sestak defeats Arlen Specter, 54-46.
The conventional wisdom: This was a stunning repudiation of the Democratic establishment.
The reality: Certainly, Specter had the support of a lot of Very Important People, including the President, many unions, and the mayors of Philadelphia and Pittsburgh. But in many cases, it seemed perfunctory. The White House elected not to send either Barack Obama or Joe Biden to the state in the closing days. The unions were nominally supporting Specter, but were concentrating their cash in Arkansas and elsewhere. As Sestak began to emerge as the superior general election candidate, their support grew even more tepid. This was an important win, and the netroots progressives who championed Sestak's campaign deserve all the credit in the world. But something can be dramatic without being especially surprising. Joe Sestak is a mainline, lunchpail Democrat who defeated a very unpopular Republican-turned-Democrat who ran an awful campaign and who Pennsylvania Democrats weren't used to punching their ticket for. No huge shock there.
Kentucky -- Republican Senate primary
The results: Rand Paul defeats Trey Grayson, 59-35.
The conventional wisdom: This was a stunning repudiation of the Republican establishment.
The reality: Because of Paul's impressive 24-point margin of victory, almost any explanation you might proffer probably contains some element of truth. But for all his libertarian and tea-party dressing, Paul in fact ran on a fairly conventional, conservative platform. He's pro-life, anti-gay marriage, anti-immigration ... there are only the faintest hints of libertarianism here. This was probably a good thing for him because Kentucky, which has traditionally been socially conservative but economically moderate, is pretty much kitty-corner to the libertarian side of the political quadrant. This was actually very clever, in a lot of ways -- Paul's last name (and decision to affiliate himself with the tea party) gained him national attention and fundraising and earned media, but to people in Kentucky, he ought to have been a very comfortable choice who was somewhat more fresh-faced than his rival. The branded product beat the generic one.
Paul might have some trouble in general election, especially after somewhat underwhelming turnout in the primary (Democratic turnout was actually 60 percent higher, although Democrats enjoy a substantial registration advantage in Kentucky). But that's more because of his inexperience and standoffishness and less because of his platform.
Kentucky -- Democratic Senate primary
The results: Jack Conway defeats Daniel Mongiardo, 44-43.
The conventional wisdom: What? The Democrats had a primary too?
The reality: Yes, they did -- and it was the closest race of the night, with Conway taking an early lead from a substantial advantage in Louisville and Lexington but Mongiardo, who was stronger in rural areas, nearly closing the gap by the end of the night. Conway had polled somewhat better in general election matchups so this is a result most national Democrats will be pleased with -- although the way that Attorney Generals have been going lately, maybe he'll prove to be a bit jinxed.
Arkansas -- Democratic Senate primary
The results: Blanche Lincoln and Bill Halter head to overtime. Lincoln has 45 percent of votes counted so far tonight, and Halter 43 percent, but a majority was required to avoid a run-off.
The conventional wisdom: Lincoln spent too much time hanging out in the middle of the road and got run over.
The reality: There are parallels between what Rand Paul accomplished in Kentucky and what Bill Halter did in Arkansas. As I mentioned earlier, Kentucky is not a particularly good state for real libertarians. Likewise, Arkansas is not an especially good state for netroots progressives, who are mostly white, liberal, and college-educated, whereas the state's Democratic primary electorate is 61 percent non-college, 64 percent non-liberal, and contains a fair number of black voters.
Halter endeared himself to national progressives and to unions with his vocal support of the public option, giving him money, momentum and media attention. But to Arkansasans, he was a relatively familiar face (as the sitting Lieutenant Governor) who ran a relatively non-ideological campaign, railing against corruption, bailouts, and wishy-washiness, as challengers of all political persuasions are doing. Halter came out against cap-and-trade, on the other hand, and tried his best to avoid taking a position on contentious social issues.
Certainly this is a rough environment for moderates, but Lincoln made matters worse by drawing unnecessary attention to herself on health care, and by picking the wrong issues to moderate upon: yes on TARP, no on the public option is a set of positions that very few rank-and-file Democrats (or voters of any kind) will share. And she was a very incumbent-y incumbent in an environment where incumbents are not popular.
Of course, we should not yet be speaking about her in the past tense; Lincoln could still win the run-off. But I suspect that the presumably superior enthusiasm of Halter's voters will pay off for him in three weeks. Turnout was actually not bad in Arkansas -- in fact, it slightly exceeded turnout in the 2008 Presidential primary -- but I don't know if Blanche Lincoln is the sort of person for whom people are going to get up off the couch to vote for twice in one month.
Pennsylvania 12th Congressional District -- Special election
The results: Mark Critz (D) defeats Tim Burns (R), 53-45.
The conventional wisdom: A big, clutch win for Democrats.
The reality: Neither outcome would have been surprising here. The polling showed a toss-up, and the district (with a PVI of R+1) is close to the national median. There's a lot of variance in open-seat elections for the House; even in an environment like 2008, Democrats would have had about a 30 percent chance of losing this seat, and even in one as relatively poor for them as 2004, they would have had about a 40 percent chance of winning it.
Still, the 8-point margin of victory was surprising. As I wrote yesterday morning: "It's really only if one of the candidates wins by middle-to-high single digits ... that [PA-12] might tell us something", and Critz met that threshold.
Republicans have some decent excuses; they may have been harmed by the fact that there was a contentious Democratic Senate primary occurring at the same time, for instance, and the DCCC seems to have a peculiar knack for winning special elections. The Democratic candidate ran against his party's health care bill! But make no mistake: there are garbage cans being kicked, and consultants being sworn at, at NRCC headquarters right now. And the Republicans may need to engage in some self-reflection about whether nationalizing the race will be the optimal strategy in each of 50 distinct states and 435 distinct Congressional Districts.