After a semester of covering the 2016 presidential campaign through the lens of an advertising major, I’ve been pondering one question: do campaign ads matter anymore?

This pondering was inspired by a piece featured in the New York Times blog The Upshot asking the same question. Posted on Oct. 19, it offers a pre-election look into the business of campaign advertising and whether or not there’s a correlation between ad spending and vote share.

This analysis is based on total volume of ads in key states compared to The Upshot’s daily forecasts of vote margins. There’s a lot to unpack in the piece, but the kicker at the end is what’s stuck with me since October:

Author Lynn Vavreck wrote: “All this suggests that Mr. Trump’s strategy, while efficient in terms of costs, may not be effective in terms of persuasion. He has let Mrs. Clinton dominate the ad war in competitive battleground states and it seems to be costing him votes.”

My own reporting this semester bore out the same conclusions: Hillary Clinton’s ad buys were concentrated in battleground states and were outpacing those of opponent Donald Trump.

But as the dust settled on the 2016 electoral map, it became clear that ad dominance didn’t matter.

Kantar Media, a market research firm that specializes in advertising research, and the Wesleyan Media Project, a center of study on political advertising at Wesleyan University, partnered on analysis of campaign ad spending data in 2016. Final ad tallies showed a stark divide between ad buys and election results.

The state of Florida was a prime example. According to Kantar Media data, 22,096 advertisements aired in Orlando, with 17,710 of those being Democratic ads and just 4,386 Republican ads. Tampa Bay saw a similar discrepancy: 21,631 ads with 77.1 percent of those being Democratic. While Clinton easily won the counties where those cities were located, Trump brought home the victory in the state by over 100,000 votes.

The same held true in Ohio, with the state’s three largest media markets (Cleveland, Columbus and Cincinnati) all seeing Democratic ads make up more than 70 percent of total advertisements bought. And, like Florida, Clinton won those counties but lost the state to Trump by almost 9 percentage points.

This election was a strange one. Both candidates were historically unpopular and that unpopularity actually led to fewer ads being purchased overall compared to the 2012 race. Throughout the duration of this course, we’ve studied the role of the media on presidential elections and the intimate details of an election campaign, but none of that studying has shed any light on that question: do campaign ads matter?

At the beginning of this course, I was ready to say yes – of course they matter. Advertisements have the ability to sway public opinion and capture the attention of the masses in ways other forms of communication don’t. They can doom presidential campaigns and propel candidates to historic victories.

But as the 2016 election has shown, advertisements aren’t everything. The data from this election shows that massive advantages in paid, traditional advertising does not guarantee a massive advantage in votes received. I loved the opportunity to focus on the cross-section of politics and advertising this semester, and I look forward to studying this campaign’s impact on the political advertising business in future elections.

–Brandon Carter