I’ve been seeing this quote going around lately from Jim Barksdale, the former CEO of Netscape. He said, “If we have data, let’s look at data. If all we have are opinions, let’s go with mine.”

I like the quote not because it strikes me incredibly astute or wise, but because it seems like it fits in with the present situation facing political analysts in this country. There’s always a place for a piece like this one from Ezra Klein that wisely counsels us to remember that there are many historical examples of presidential candidates catching fire early on only to fizzle when people got focused on electability as a top priority.

To put the history into a context that is relevant to today, there are four candidates who I thought would be the Republican nominees but who I sometimes wondered if they could really pull it off. Let me walk through them for you.

In 2000, prior to John McCain’s massive, decisive win in the New Hampshire primary, I never gave him even the slightest chance of winning the nomination. I had to briefly reconsider that assessment, but McCain acted like the dog who caught the car and didn’t seem to know what to do with his momentum. That seemed evident during his victory speech in New Hampshire and it never improved. After his ugly defeat in South Carolina, order was restored.

In 2008, it was completely unthinkable that anyone other than John McCain would win the nomination but the base hated him and he completely ran out of money and had to stop and retool his campaign. He was just lucky that Mitt Romney wasn’t a stronger alternative. No one else in the field (Rudy Giuliani, Fred Thompson, Ron Paul, Tom Tancredo, Sam Brownback, Mike Huckabee) could pass the basic laugh test. McCain won because the alternatives were so unrealistic.

If anything, 2012 was a clearer case of this. Mitt Romney was an unattractive candidate with a record uniquely unsuited for the times and mood of the Republican base. But there was no one else to choose. After flirting with every other declared candidate, the party grudgingly gave the nomination to Romney because they still cared about electability.

And, ever since 2012, and long before he indicated any intention of running, I identified Jeb Bush as the only candidate who could fit this mold. He was the only Republican in the country who could make any kind of plausible case that he might win the Electoral College in anything resembling normal circumstances. And, yes, this is a straightforward electability argument, which shows that I honor the data and am willing to use it as a guide.

Still, this electability thing is subjective and it’s not just some abstract concept that sits there fixed and immutable. There are mechanisms that are used to build up a candidate’s electability and to challenge the electability of others.

When Howard Dean caught fire, the establishment worked to convince people that he wasn’t electable. And they worked to convince us that John Kerry was electable. Right now, the Republican establishment is attempting to do the same thing with Donald Trump vis-a-vis Bush, Walker, Rubio, and, seemingly, Carly Fiorina. The Democratic establishment is rousing itself to do the same thing to Bernie Sanders for Hillary Clinton’s benefit.

What we’re talking about here, though, is a limited data set. And it’s not the case that the parties always succumb to the logic of electability. The Republicans clearly rejected this logic in 1964 when they nominated Barry Goldwater over Nelson Rockefeller. And the Democrats rejected it in 1972 when they chose George McGovern over Ed Muskie and Hubert Humphrey. In 1976, the Democrats rolled the dice on an unknown governor from Georgia and wound up winning the presidency. It’s certainly possible for a candidate who has big questions about his or her electability to win the nomination, and not completely impossible for them to go on to win in the general. We’ve seen these things happen and that’s part of the data, too.

We already know that some data models are not working. By precedent, the Republican runner-up in 2012 should be beginning this contest as the favorite. That’s what happened after 1976 when Reagan came in second place. It’s what happened after 1988 when Bob Dole came in second place. It’s what happened after 2000 when John McCain came in second place. It’s what happened after 2008 when Mitt Romney came in second place. But Rick Santorum, the runner-up in 2012, is so low in the polls that he barely registers.

I think we are really entering a phase of American politics that is unpredictable and where precedent isn’t our surest guide. Or, maybe, it’s still our best guide, but it just isn’t an adequate or reliable one.

So, that gets us to feel. I think a good analyst right now has to have a feel for the mood of the electorate, and they need to combine that with a mastery of the mechanics of how the nominations are won and an understanding of the way new election finance rules are changing the game.

This is a tall order, and it’s one that presents a daily challenge as I try to figure out what the hell is going on.

[Cross-posted at Progress Pond]

Our ideas can save democracy... But we need your help! Donate Now!

Martin Longman is the web editor for the Washington Monthly. See all his writing at ProgressPond.com