Nonprofit, nonpartisan journalism. Supported by readers.

Community Voices is generously supported by The Minneapolis Foundation; learn why.

On Nate Silver, the limits of political moneyball, and why Trump may win

What political moneyball misses are three important factors: candidate quality, mood of the country, and the “politainment” quality of American politics.

Trump is a weak candidate, but he has the benefit of it being an anti-establishment year at a time when Clinton is the poster child for the establishment.
REUTERS/Jonathan Ernst

It’s time to admit it: Trump may win.

schultz portrait
David Schultz

Increasing and begrudgingly, the establishment politicians, pundits, and analysts are beginning to realize that Trump may actually win the presidency. Nate Silver, whom too many people put too much political stock in, is now saying that Clinton is favored 58.8 percent to win, down from dramatically larger percentages even just a couple of weeks ago. It’s nice to see that Silver finally is getting in the range of my assessment, which has said Clinton has a 55 to 60 percent chance of winning. But even then, I may be exaggerating her chances and would put it at 50+ percent — barely break even.

Silver came to fame with applying the logic of moneyball to politics – successfully using his algorithms to call 99 of 100 states in the last two presidential elections. Silver is smart, but it would not have taken an Einstein to call at least 90 of the 100 states. This is what the logic of my book “Presidential Swing States” is all about – showing how because of the Electoral College, partisan voting, and party alignments, the elections were over before they started in all but 10 states. Moreover, in the last two election cycles, one could have also eliminated a couple of other states from the swing-state category, giving one about 94 states that would have been easy to predict. Throw in the relative stability of polling and getting to 99 is not so hard.

Article continues after advertisement

Why is all this worth mentioning in connection with Trump and the 2016 elections?  First, despite 2016 being a unique election year, there are still many forces that make it a relatively normal election that again is reducing the outcome to only a handful of swing states. Still in play are the core 10 that have consistently been in play for the last seven election cycles, such as Ohio, Florida, North Carolina, and Iowa, but additionally a few others such as Michigan and Pennsylvania may now be  flip-able. There is evidence that partisanship is still a factor driving how people vote, with few real swing voters moving from party to party. Voters are coalescing around Clinton to some degree but more so for Trump, thereby producing a normal election pattern that suggests is it still a few swing voters in a few swing counties in a few swing states that will determine who gets to 270 electoral votes and wins the presidency. All this bodes well for the Nate Silvers of the world.

The uniqueness of 2016

Additionally, when one looks at Trump versus Clinton, traditional wisdom hands it to Clinton. Until recently ahead in the polls, she has a better-run campaign, more money, and has insulted far few people than Trump. Yet this is where the uniqueness of 2016 kicks in, and where the limits of political moneyball appear.

First, there is no such thing as an Electoral College lock for the Democrats who think demographics is destiny. Statistically voters may be presupposed to vote a certain way — but you need to get them to vote. Trump supporters are passionate and will show up; Clinton’s are not. She relies on many voters who are mercurial at best when it comes to voting, and she has done little to address the lack of enthusiasm many have for her. She has yet to seal the deal with the Bernie Sanders supporters and liberals, simply assuming that running to the center as a Republican much like her husband did will result in these people having nowhere else to go and therefore they will vote for her. The year 2016 and the millennials are very different from 1992 and the baby boomers. Even African-Americans who loved  Obama in 2008 and 2012 may not come out the same way in 2016.

Second, polling is more complicated now than before. Cellphone technology, polling costs, defining likely voters, and other issues all complicate this year’s predictions. Many media outlets are cutting costs on polls. Take, for example, the Sept. 18 Star Tribune poll, with results from 625 respondents and landlines constituting 69 percent. A good poll should have at least 1,000 respondents and nearly 70 percent cellphones. This is a flawed poll. Silver’s predictions are only as good as the polls and he has blown several predictions this year, consistently overestimating Clinton’s strengths. Face it: Clinton often polls better than she performs, while Trump performs better than he polls.

Third, what political moneyball misses are three important factors: candidate quality, mood of a country, and the “politainment” quality of American politics. No matter how good a campaign, some candidates are simply not good. Clinton is a weak candidate and does not resonate well. Factors such as likablity are missed in political moneyball.  

The anti-establishment benefit

Yes, Trump is also a weak candidate, but he has the benefit of it being an anti-establishment year at a time when Clinton is the poster child for the establishment. And Trump understands the politainment aspect of contemporary politics that is increasing post-truth (candidates do not tell the truth and the public does not expect it), post-rational, post-issues, and simply pop culture sound-bite driven. Image is everything; content is nothing.

Put this all together and the traditional political pundits, politicians, analysts, and Nate Silvers of the world are missing a tremendous amount about politics in 2016. They and traditional political scientists also miss the importance of how politics is moving marginal numbers of swing voters in a few swing counties in a few swing states and therefore the issue is not how Trump or Clinton appeal to large numbers of people but only how to move a few people. Aggregate analysis misses subtle shifts.

Right now logic dictates Clinton still should win, but the reality may be that the public should prepare for a Trump presidency because there are many reasons to think he will win.

David Schultz is a Hamline University professor of political science and the author of “Election Law and Democratic Theory” (Ashgate, 2014) and “American Politics in the Age of Ignorance” (Macmillan, 2013). He blogs at Schultz’s Take, where a version of this piece first appeared.   


If you’re interested in joining the discussion, add your voice to the Comment section below — or consider writing a letter or a longer-form Community Voices commentary. (For more information about Community Voices, email Susan Albright at