I used to take the view — and I think I’ve been pretty consistent about this — that special elections to Congress, while interesting on their own merits, should more or less be ignored in terms of their national implications.
I still probably err toward that side now. The campaign press is likely to exaggerate the impact of special elections (as it does with most day-to-day events). If I were interviewed on television and had the time for just one five-second comment, it might be spent on pointing this out.
But that’s an oversimplified position. Special elections, unlike polls, are “hard” data — real people voting on real candidates and real issues — and can provide us with some insight if we look at their results very carefully.
Since we have the luxury of space here, let me make several quick points that pertain both to the contests in New York and Nevada on Tuesday night, and to the spin that you’re likely to hear from Republicans and Democrats afterward, depending on the results.
1. Blaming the candidate is often a poor excuse, as it appears to be here for Democrats.
Candidates can matter: Would Republicans have lost the Senate race in Delaware last year — they lost it in a landslide, in fact, by 17 points — had they nominated Michael N. Castle rather than Christine O’Donnell? Polls before the race suggested that Mr. Castle would have won.
Harry Reid ran a brilliant campaign to save his Senate seat in Nevada, but with an approval rating only in the high 30s, he probably would not have accomplished that had he been up against a better candidate than Sharron Angle.
Even in the case of the Massachusetts special election in 2010, there was room to assign some of the blame — certainly not all of it — to the Democratic candidate, Martha Coakley, as well as to give some of the credit to the strong personal qualities of Scott Brown.
But these are exceptional cases: Ms. O’Donnell and Ms. Angle were inexperienced candidates who were far outside the political mainstream in their states. Ms. Coakley insulted the Red Sox — in a race in Massachusetts.
I don’t think Assemblyman David I. Weprin, the Democratic candidate in the New York race, comes all that close to that bright-line test.
The grievances against him are fairly petty: for instance, that he misstated the size of the national debt — a gaffe, to be sure, but one of a minor magnitude that most candidates would expect to make at least once or twice over the course of a campaign. Or that Mr. Weprin is a bad dancer. (The Republican candidate, Bob Turner, made a couple of gaffes as well).
Mr. Weprin is not a world-beating candidate, but he’s been elected to other offices several times before and he holds positions on the issues that square well with his district. And it’s not as if Mr. Turner — a 70-year-old who has never held elected office before, who raised very little money and who lost by 20 points in 2010 — is the Queens equivalent of Scott Brown.
The polls did show that Mr. Weprin had favorability ratings inferior to those of Mr. Turner — but sometimes that’s a symptom of a losing campaign rather than one of its causes.
2. New York’s Ninth Congressional District has highly unusual demographics, with a set of local issues that are unlikely to extrapolate well to the rest of the country.
On the other hand, while all Congressional districts have their quirks, New York’s Ninth is especially unusual.
First, there are the local issues — Barack Obama’s positioning toward Israel, Mr. Weprin’s endorsement of a plan to build a mosque and Muslim cultural center in Lower Manhattan, and possibly gay marriage — that will resonate more in Queens than they will in the rest of the country.
Roughly 40 percent of voters in the Ninth District are Jewish, 20 times the rate in the country as a whole. Moreover, and perhaps more important, many of those voters are Orthodox Jews, who often have starkly different political viewpoints than Reform or secular Jews, and who are extremely rare in the United States outside a few spots in the New York region.
There’s also the fact that the district was already behaving unusually in 2008. Despite having a 37-point edge in party registration, Mr. Obama won the election by only 11 points there — barely better than the seven-point edge he had nationwide. I doubt that there was any district in the country, perhaps outside a few remnants of the “Solid South,” where so many enrolled Democrats voted against Mr. Obama.
Mr. Obama’s unpopularity is no doubt a huge factor in this race. But certain types of critiques are likely to be disproportionately resonant in this particular district compared with others.
3. If the polls are right, the result in Nevada should be as troubling to Democrats as the result in New York.
Although the special election in Nevada has gotten less attention, in some ways it might be the more appropriate race for drawing national implications.
Sure, Nevada’s Second Congressional District has a few oddities: it’s had fast population growth and crashing housing prices. But its demographics are otherwise fairly “normal” and heterogeneous. And in contrast to the New York district, it seems to be a place where Democrats were making a lot of progress: Mr. Obama lost it by less than a full percentage point in 2008, whereas John Kerry was beaten there by 16 points in 2004.
Democrats probably weren’t going to be favored in this district, which is still somewhat Republican-leaning. But a double-digit loss, as seems possible based on the polls, is a decidedly subpar result for them.
In addition, the national implications notwithstanding, Nevada is a pivotal state in both the presidential and Senate elections next year.
4. Academic studies find that special elections do have some predictive power, especially taken as a group.
A study by David Smith and Thomas L. Brunell on special elections to the House since 1900 finds that they do anticipate general election results to some extent: there is a modest, but statistically significant, correlation between how many seats change hands during special elections and which party does best in the next midterms.
The connection does not always work. In advance of the 2010 elections, Democrats won seven of the nine special elections to the House, with an even score as far as seats changing parties. (Democrats lost Hawaii’s First Congressional District to Republicans, but won New York’s 23rd, both under somewhat unusual circumstances with multicandidate fields.) Then they lost 63 seats in November.
Nor is it clear that special elections have predictive power above and beyond other factors, like economic performance and generic ballot polls. Still, they seem to provide kernels of useful data.
5. Special elections have more predictive power if you look at the margin of victory and not just the result.
Mr. Smith and Mr. Brunell have another finding that echoes something I’ve frequently said about special elections: They have more predictive power for national elections if you look at the margins of victory rather than the winners and losers. If Mr. Weprin pulls out a victory by 1,000 votes, we should basically draw the same implications as if he had lost by 1,000 votes, even though the headlines will be much different.
6. Special elections can also affect the behavior and morale of the parties, and they may do so here.
But that assumes that parties are as cool-headed about the results as we’re attempting to be here. And they aren’t. A loss in New York’s Ninth District, after a rough couple of months for Democrats, will only worsen their morale. It could discourage other Democrats from running.
And it could affect campaign strategy in other races. Republicans might conclude that Israel is a vulnerability for Mr. Obama, for instance, and make it a bigger part of their campaigns. That may or may not be a good strategy for them — but it could affect the landscape of the campaign all the same.
7. Special elections won’t provide us with much insight into the degree of anti-incumbent sentiment.
This is axiomatic: Barring unusual cases, incumbents aren’t on the ballot in special elections. So their results are likely to tell us more about how open-seat races might play out than those in which there is an incumbent on the ballot.
In some ways, in fact, a poor result for Democrats might tell us more about the Senate picture, where there are a number of key open-seat races (including in Nevada), than it would about the House, where the Democrats have dozens of fertile Republican incumbent targets. A “throw the bums out” mentality could still produce some gains there even if Democrats were having a rough night otherwise.