An oft-discussed factoid from the 2012 elections: Nationally, Democratic candidates for House seats won 1.7 million more votes than Republicans. But despite this, Republicans won a solid 33-seat majority in that chamber. There are a number of forces driving this -- redistricting is certainly one of them. Other have argued that the country's geography has an inherent Republican bias -- Democratic voters tend to be clustered in urban areas, where they are more likely to "waste" votes with large majorities.

But exactly how exceptional was the 2012 result? I wanted to find out, so I used election returns from the House clerk to see how the House popular vote compared the seat distribution going back to 1946. First, caveats -- the national popular House vote is a bit of an abstraction. People vote for candidates, not parties. Incumbency also plays a role in these vote totals -- popular incumbents often generate weak opposition, which can lead to lopsided counts when opposition voters simply don't show up.

Still, it is useful to situate the famous 2012 factoid in the context of what's come before, as we will certainly hear more about it leading up to this year's election. To that end, I plotted the difference -- in seats -- between the expected number of Democratic seats in the House (based on the national vote share) and the actual number of seats below. Majority-Democratic Houses are shaded blue, and majority-Republican Houses are red.

What stands out here isn't necessarily the Democratic deficit in 2012, but the overwhelming Democratic advantage in nearly every election up until the 1990s. Democrats were short 20 seats relative to their vote share in 2012. But in 1976, for instance, Democrats had a surplus of 45 seats -- they received 57 percent of the vote, but won 67 percent of the seats. Democratic advantages of 20 seats or more were common through most of the '60s, '70s and '80s.

It's easy enough to pinpoint exactly when this changed -- in 1994, with the Gingrich revolution. Inasmuch as this represented an abrupt and long-lasting upending of the House status quo, this really was a revolution. Democrats went from a 29-seat surplus to a 10-seat disadvantage in the space of two election cycles.

So in absolute terms, the Democratic discrepancy in 2012 is well within the bounds of what came before. The difference, of course, is that Democrats won a majority of votes but did not win a majority of seats. That has real-world consequences, as the majority party sets the chamber's agenda. Think of how different a Pelosi-led House would be from the Boehner House we've had since 2013.

And 2012 was only the second time a party won the popular vote but not the majority -- the other time was in 1996, and again the outcome favored Republicans. But then, there was only a 10-seat discrepancy in Republicans' favor.

To my eye these numbers cast some doubt on the notion that an urban/rural bias favoring Republicans is behind the Democrats' recent underperformance in the House. The share of people living in urban areas increased only slightly between 1970 and 2000 -- certainly not to any degree that would fully explain the dramatic change in Democrats' prospects in the 1990s.

It's also important to note that the Republican revolution of 1994 wasn't about gerrymandering. Districts were redrawn in 1991, which meant that new boundaries were in place for the 1992 elections, when Democrats still held a 29-seat advantage. That said, the lopsided seat totals in 2012 may have been partially due to Republican gerrymandering, but we'll want to keep a close eye on vote totals going froward to know for sure. If Democrats continue to win the popular vote but lose the House majority, it will be a good indicator that our post-2010 political landscape is starkly different than what came before.