A woman casts her vote in 2014 in Thornton, Colo. (Jack Dempsey/Associated Press)

During the course of a presidential campaign, we get very worked up about momentary controversies that end up having little or no real impact on the way the race turns out. There isn’t necessarily anything wrong with that — the final result isn’t the only thing that’s meaningful, and the playing out of the process tells us a lot about who we are and what we ask of our leaders (for better and for worse). But it’s also good to step back from the daily circus of the campaign and look at the broad forces that shape the ultimate outcome. There’s an interesting analysis of a year’s worth of polling data out today from Reuters about party identification, and it shows how deep a challenge the GOP faces in the fall. Here’s an excerpt:

  • In 2012, Democrats made up 44.7 percent of party-affiliated likely voters, compared to 39.1 percent Republicans, a difference of about 6 percentage points, according to the analysis of 87,778 likely presidential voters polled leading up to the 2012 presidential election. The results have a credibility interval of plus or minus 0.3 percentage points.
  • Three years later, that lead had grown to nine points, 45.9 percent to 36.9 percent, according to the analysis of 93,181 likely presidential voters polled in 2015. The results in 2015 have the same credibility interval as 2012.
  • Among Hispanics who are likely presidential voters, the percentage affiliated with the Republican Party has slipped nearly five points, from 30.6 percent in 2012 to 26 percent in 2015. Meanwhile, Hispanic Democrats grew by six percentage points to 59.6 percent.
  • Among whites under 40, the shift is even more dramatic. In 2012, they were more likely to identify with the Republican Party by about 5 percentage points. In 2015, the advantage flipped: Young whites are now more likely to identify with the Democratic Party by about 8 percentage points.
  • Meanwhile, black likely voters remain overwhelmingly Democratic, at about 80 percent.

Gallup data show a similar movement in party identification; four years ago they showed the two parties even in support, while in Gallup’s most recent polling, Democrats outnumbered Republicans by 5 points. This is important because cross-party voting at the presidential level has virtually disappeared. For decades, Democrats had a wide advantage in party ID, but Republicans regularly won the White House because significant numbers of Democrats would sometimes vote for a Republican president (Republican voters, though smaller in number, were more loyal). But that no longer happens. Here’s a chart I made using American National Election Studies data showing how partisans have voted since 1952:


In 2012, 93 percent of Democrats voted for Barack Obama, and 93 percent of Republicans voted for Mitt Romney. For almost everyone who had a party affiliation, all the back and forth of the campaign — the debates, the ads, “47 percent,” “you didn’t build that” and all the rest — didn’t change anything.

Of course, not everyone has a firm party identification that never changes from year to year. There are a few people who will watch the campaign and decide that they support the Republican candidate, and therefore they’re now willing to call themselves Republicans. And it isn’t surprising to see a movement toward the Democrats among Hispanic voters, when for months the Republican candidates have competed to see who’s more enthusiastic about building walls along our southern border and tossing out immigrants.

None of this means that a Democratic victory is guaranteed. There are still enough independent voters to shift the election one way or the other, and they’ll be influenced by who the two nominees are. If there’s an economic downturn next year, it will make a Republican victory much more likely. A large terrorist attack could also have a significant impact on the vote. A closely divided country with high levels of party loyalty means that every presidential election is going to be close. When Barack Obama beat John McCain by 7 points, it was the modern equivalent of a blowout. Unless one party does something crazy like nominate Donald Trump, we’re unlikely to see a margin like 1984 (an 18-point win for Reagan) or 1972 (a 23-point win for Nixon) for an awfully long time. And when the race is close, small things can matter.

It’s also important to note that this analysis concerns only the presidential race — even as Republicans struggle to figure out how to win the White House, they still control Congress and a majority of governor’s offices and state legislatures. While the president’s party usually suffers down-ballot (since people blame the president for whatever’s going wrong and take out their national frustrations in more local races), Democrats have lost an unusually large amount of ground during the Obama years, and it remains to be seen how much they can win back. So Republicans can reasonably argue that while they have a presidential problem, in the larger political picture they’re actually doing quite well.

Which is true enough. But for the next 10 months, everyone is going to be consumed with the presidential race, and rightfully so. And Republicans are starting off with a significant disadvantage.