One of the more persistent assumptions about present-day presidential elections is that they are primarily base elections. One way of expressing what this means is simply to quote from a David Atkins piece from this past weekend.

General elections are won by turning out the people who already agree with you ideologically, but only show up to vote every other election when they really feel inspired to, but otherwise feel that politics is a waste of time that doesn’t change anything dramatically [or] affect their daily lives.

Another way of phrasing this is that presidential elections are won by motivating the people in your base who need some external motivation to show up and vote. You can do this by getting them excited or you can do it by making them afraid or you can do it by making personal contact with them, or some combination, but you’re really not trying to persuade anyone to change their ideological point of view.

I think if you look at the last four presidential elections, you can take this analysis and plug it in to each of them individually to see how the parties performed. And I think this has some explanatory power.

Going back to 2000, the election was so close that you can plausibly argue that turning any of these dials just slightly could have changed the outcome. I mean, setting aside the fatal design of the Palm Beach County butterfly ballot, and the votes cast for Ralph Nader, and the strategies employed in the recount, and the Supreme Court’s decision to intervene decisively on Bush’s side, it’s possible to envision Gore winning simply by being a little more exciting or by instilling a little more fear in Democrats or by having done a slightly better job of voter contact. As is always the case when an incumbent president is leaving after eight years, the opposing party was more motivated to vote in 2000, and Gore needed to do just a little more to counteract that.

In 2004, it’s easy to see that the Republicans did an outstanding job of getting out their base in Ohio, and that turned out to be decisive. This was done in part by using the threat of gay marriage as a cudgel that aroused fear in social conservatives. Anti-gay measures were placed on the ballot, and that alone may have made the difference.

The two Obama elections were a bit different from the two Bush elections, especially the first one in 2008. In 2008, Barack Obama did a good job of actually converting many Republicans. You could see this most visibly in some of the endorsements he got from people like Colin Powell, and the Eisenhower grandkids, and William F. Buckley’s son. But the phenomenon was much more widespread, as the Bush administration ended so ignominiously that it was impossible for many people to ignore its failures. McCain and Palin were also poor campaigners whose performance was hard to defend. In the end, I’d argue that Obama’s superior organization and voter contact wound up padding his victory by adding some states, like Indiana and North Carolina, to his column, but he mainly won by winning the argument and not simply by better mobilizing a preexisting base.

When he sought reelection in 2012, it was more of a reversion to the 2000 and 2004 style election, only this time with the Democrat coming out on top. The main difference was that the election wasn’t nearly as close as the earlier ones were, but that was partly explained by Romney doing a comparatively poor job of motivating his base.

In all of these elections, though, the loser was able to top 45% of the vote. McCain-Palin did the worst, managing only 45.63%, and that looks to a lot of people like a hard floor.

But is it really a hard floor?

Think about when we began talking about Red States and Blue States. That dichotomy wouldn’t have made any sense in the 1980’s. New Jersey voted for Poppy Bush in 1988. In 1984, Reagan won 49 states, losing only Minnesota and the District of Columbia.

In a way, it’s unfortunate that Stanley Greenberg profiled the Democrats of Macomb County, Michigan that year, dubbing them Reagan Democrats. It’s a shame not because it was bad analysis. It was insightful analysis that still has explanatory power today. But it wasn’t just socially conservative, patriotic autoworkers who rejected Walter Mondale and the Democratic Party in 1984. It was the people of Massachusetts and Rhode Island, perhaps the two most reliably Democratic states not named Hawai’i.

Whatever you thought about him or think about him in retrospect, Ronald Reagan won the argument against the Democrats in 1980 and, especially, in 1984. He didn’t win by exciting more preexisting Republicans, and he didn’t win simply by making people afraid of his opponents. He won by creating Republicans who had previously been Democrats, and not just among white working class union voters. He also won the youth vote, with reverberations that are still being felt today.

Now, some people will argue that things are different today. The parties are more ideologically stable and predictable, and voters are much less likely to split their tickets. The way people get their news is different, with more folks getting a steady supply of right-wing ideology and little else. Demographics have changed, too, and populations have sorted. These factors, in combination, could provide a higher floor for each party so that they can lose the argument and still be assured of getting somewhere in the neighborhood of 45% of the vote.

I don’t discount that argument at all. I think it would be very hard for anyone to win 49 states in our current political environment, no matter how obviously superior they were or how blatantly incompetent their opponent. But that doesn’t mean that every general election is a base election. And I suspect, although it is not yet certain, that 2016 will not be a base election.

The way I like to envision this is as if the two parties are like tectonic plates. If you think about the San Andreas Fault, it’s made of a Pacific Plate and a North American plate that are locked together. Most of the time, the intersection is stable. When the pressure builds up sufficiently, there can be some slippage along the fault line, but this usually only moves things an inch or two at a time. Once in a while, though, and precisely when is hard to predict, there is enough pressure built up to cause a major slippage and a large earthquake.

Back in the 1960’s through the 1980’s, the plates weren’t really locked at all, but were gliding all over the place as the parties did a slow motion realignment. But, starting in the 1990’s, they melded together and we’ve been in this 45-45 political universe ever since.

When you look at the gridlock in Congress and the resulting dissatisfaction of the American electorate, this is analogous to the pressure that builds up in a stable system.

I don’t want to take this analogy beyond where it can go, but what’s happened in the Republican Party with the nomination of Donald Trump is clearly a sign of weakening. For these two plates to stay stable, they need to maintain roughly equal strength, and if one them starts to crumble, there should be the possibility (even the likelihood) of major slippage along the fault line. And that’s when a political earthquake can happen.

I’ve been looking for signs of this for a couple of years now, long before Trump came onto the scene. And I’ve identified a lot of warning signs along the way, some of which have already resulted in some significant quakes like Eric Cantor losing a primary and John Boehner giving up the Speaker’s gavel.

So, in summary, I am not convinced that this election will be won by “turning out the people who already agree with you ideologically.” I think there is a strong possibility that this election will be won more decisively than any election since 1988. But, like predicting earthquakes, this is an imprecise science. The quake I’m seeing could come in 2020. The plates could slip in the opposite direction from what I’m expecting. After all, there are weaknesses and fissures developing on the Democrats’ side, too.

What I feel confident about is that the pressure has built up to such a degree that a major quake is in our near future. Gridlock cannot stand forever without one side winning the argument with the American people.