Skip to main content
What Harry Got Wrong In 2015

I f-ed up a lot in 2015. It’s bound to happen when you’re predicting things for a living and involved in approximately 200 articles, chats, podcasts and videos per year. I also got a lot right. But that, or the volume of work, doesn’t excuse my errors.

So what exactly did I screw up?

One of the biggest errors I think I made in 2015 was concentrating too much on who would win the Republican nomination and not so much on who would drive the conversation.1 You can see that clearly with Donald Trump (who I still don’t think will win the Republican nomination). Not only did I dismiss Trump’s candidacy on many occasions (here, here, here and here, for example), I was sometimes eager to do so in order to confirm my original belief that he wouldn’t win.

But winning the race and affecting the race are different things. Trump may lose in Iowa, collapse in New Hampshire and disappear from the national stage. If that — or something like it — happens, much of the sound and fury surrounding Trump in 2015 will seem silly in retrospect. But regardless of where Trump ends up once voting starts, he has certainly been the dominant voice in the Republican nomination race so far.

One big reason I didn’t take Trump seriously was because he entered the race with relatively poor favorability ratings, and putting too much stock in those was another semi-major mistake I made in 2015. My research showed that past nominees started their campaigns either well-liked or not well-known; Trump was the opposite: well-known and disliked. But I underestimated how much voters’ impressions of even well-known candidates can change. Trump’s favorability ratings among Republicans, while still mediocre, improved a ton.

You can see something similar with Chris Christie. I poohpoohed his chances early and nearly as often as I did Trump’s. I still don’t think Christie is at all likely to win the GOP nomination, but Christie, like Trump, was able to regain at least some modicum of popularity among Republican voters, which has made him competitive in New Hampshire.

What else did I mess up? We’ve talked a lot about the endorsement primary — the competition for support from party officials. What I didn’t seem to emphasize enough at times is that just because someone doesn’t win the endorsement primary doesn’t mean that candidate can’t be competitive. It just means he or she is less likely to win. Gary Hart, for example, was crushed in the Democratic endorsement race in 1984, yet fought Walter Mondale until the last primary. A more recent example: Rick Santorum emerged as the main opponent to Mitt Romney in the 2012 Republican primary despite a severe lack of support from elected Republicans.

Put another way, I didn’t communicate that the endorsement primary doesn’t predict vote percentage as much as it predicts winners. Many candidates who had the second-most support in the endorsement primary didn’t come close to winning the nomination (Howard Baker in the 1980 Republican campaign, John Glenn in the 1984 Democratic campaign and Phil Gramm in the 1996 Republican campaign, for example). Potential isn’t the same as potential realized. This year, that was true for Scott Walker, and it will probably be true of Jeb Bush as well.

Moreover, the party actors have a say in the primary only when they decide to. It was my belief at the beginning of the primary process that Ted Cruz would be a dead duck because most of his fellow GOP officeholders don’t like him. What I didn’t anticipate back in March (before Trump entered) was that the party actors might have too many favored candidates still in the race (Bush, Christie and Marco Rubio) to pick a favorite and too many they didn’t like (Cruz and Trump) to form a united front before the primaries started.

What I can take heart in (and I hope my readers can too) is that I can learn from these missteps and apply their lessons in the future. In fact, I’ve already acknowledged that Christie has a shot in New Hampshire and that Cruz has a shot at winning the nomination. Only a fool sticks to his original viewpoint when the facts change and it becomes clear that his hypothesis was off-base.


  1. Funny enough, I did do this on the Democratic side, despite some bumps along the way.

Harry Enten is a senior political writer and analyst for FiveThirtyEight.