Many scholars, myself included, have long argued that political parties control presidential nominations.

The rise of insurgent candidacies in the 2016 race, however, has led some to concludethat this time is different, that the theory is wrong or that it may be days away from being proved wrong.

Given the vigorous debate, let’s go through what the theory is, what it isn’t, and why it’s important.

More coverageOpinion and analysis about the 2016 presidential campaign

Statistician Nate Silver explained the theory this week in a very useful column on his website, FiveThirtyEight, about “The Party Decides,” the much-cited book that lays out one version of the idea of party control. For Silver, the book’s thesis boils down to: “You ought to pay attention to what influential people who care about a party nomination are doing, since they can have a lot of say in the outcome.”

The “influential people” Silver is talking about are those I call “party actors” — the politicians, campaign and governing professionals, formal party officials and staff, donors and activists, and party-aligned interest groups and media who make up U.S. political parties. They care deeply about nominations because these define parties, and determine not only the choice of candidate, but also the party agenda and priorities.

For my money, saying parties “can have a lot of say in the outcome” understates their role: Parties have determined nominations from the 1980s through 2012. But other party scholars would probably be more comfortable with the more modest claim.

advertisement | advertise on newsday

An important distinction is that we’re talking about nominations, not pre-Iowa polls or individual primaries.

So the Washington Post’s Dan Drezner is wrong to say that the theory predicts that Hillary Clinton will win Iowa and New Hampshire and that those primaries are an “easy test.” The test is whether Clinton, who has overwhelming support from party actors, wins the Democratic nomination.

Individual primaries are relatively unpredictable. That’s why everyone agrees that parties often fail to control nominations in congressional or gubernatorial races.

Sign up for The Point

Go inside New York politics.

Sometimes a candidate catches fire at the right time, and even a fairly unified party is unable to overcome the publicity that can generate. And that can happen in individual presidential primaries, too.

The difference at the presidential level is that the sequential process — week after week after week of primaries and caucuses — allows the party time to regroup and recover. For example, Pat Buchanan won in New Hampshire in 1996, Jerry Brown upset Bill Clinton in Connecticut in 1992, and Newt Gingrich won in South Carolina in 2012, but none came close to being nominated.

Indeed, this theory about parties isn’t mainly a tool to predict which candidate will win a nomination. More than anything, it describes how political parties control political institutions, including the presidency, with nominations being one important way that parties constrain politicians.

From the point of view of party actors, which candidate is nominated isn’t as important as ensuring that any plausible nominee is firmly attached to the party, its people and its policy program. That any nominee will produce what Richard Skinner calls a “Partisan Presidency” like those of Ronald Reagan, George H.W. Bush, Bill Clinton, George W. Bush and Barack Obama.

That’s why Drezner also is wrong to point to Republicans seemingly warming to Trump this month as proof that a Trump nomination would be “not fatal” to the theory.

advertisement | advertise on newsday

There’s no reason to believe that Trump, if elected, would have any loyalty to the Republican Party. If he wins the nomination, it will be entirely on his own, with a campaign organization almost entirely staffed with loyalists, not the usual Republican campaign professionals.

He really is, even more than Ted Cruz or Bernie Sanders, a total (party) outsider. His nomination would signal complete failure for “The Party Decides,” although exactly what kind of failure, and what party scholars would take from it, is a complicated question.

Whatever a Trump nomination would say about political science theories, however, the far more important point is that it would be a failure of the Republican Party, signaling unpredictable but potentially very serious changes to what that party has been for the last 40 years or more.

Jonathan Bernstein is a Bloomberg View columnist covering U.S. politics.