Professors: How the Electoral College concentrates candidate attention and why it matters

The outcome of the 2024 presidential election will likely hinge on what happens in seven key states: Arizona, Georgia, Michigan, Nevada, North Carolina, Pennsylvania and Wisconsin. Together, around 48 million voters live in these states, comprising just 18% of about 262 million potential voters in the American electorate. Because of the Electoral College setup, and specifically the winner-take-all feature of American elections in all states but two (Maine and Nebraska), both presidential campaigns will focus much of their attention on these battlegrounds. 

This strategic targeting has tangible consequences. Analytics firm AdImpact estimates that 76% of television ad spending from Super Tuesday through Aug. 23 was concentrated in these seven states. Data compiled by The Washington Post shows that Republican presidential nominee Donald Trump and running mate JD Vance have made a total of 38 campaign appearances since mid-July, with 29 of these in battleground states, while Democratic presidential nominee Kamala Harris and running mate Tim Walz made 25 appearances, of which 16 were in the battlegrounds. Over the same period, these four candidates made a combined total of no appearances in New York, one in California, two in Texas and two in Illinois, four states that together include 30% of the total U.S. population.

Numbers like these raise important questions about how the rules governing our presidential elections influence the relative importance of voters in specific states. Put plainly, campaigns might weigh the priorities and preferences of voters in Georgia and Pennsylvania more heavily than those of voters in California and Texas. One need look no further than Harris’ Pennsylvania-driven support of fracking or Trump’s Nevada-driven proposal to end federal taxation of tips to see the policy consequences of this imbalanced attention. And because battleground state voters will garner the lion’s share of television ads, campaign visits and grassroots mobilizations efforts this fall, turnout in battleground states tends to be higher on average (by about 5 percentage points) than in non-battleground states. In contrast, the vast majority of Americans who cast ballots outside this prized subset of states will mostly be relegated to spectator status between now and Nov. 5. 

However, it is unclear whether this is a recent development in American presidential elections caused by polarized partisan politics and microtargeted outreach, or whether it is something that’s been around a long time, caused by the unique structures of presidential selection rooted in our Constitution and honed by strategic candidates and their campaigns.

To better understand the nature and consequences of Electoral College plans, we spent years combing through presidential libraries and candidate archives to unearth the strategy documents and candidate records that reveal how campaigns sized up states throughout the modern era. Our recently published book explores Electoral College strategies over 18 elections from 1952 through 2020 and reveals that presidential campaigns have always developed detailed and sophisticated targeting strategies to win the Electoral College, at least over the last seven decades or so. In this regard, 1952 isn’t much different from 2024.

The reason is straightforward: Winning presidential elections requires a campaign to carefully distribute its limited time and money to amass a majority of electoral votes (currently 270 are needed to win). When state-level outcomes seem to be a foregone conclusion, the campaigns consider those states out of play, and they are usually ignored. Mainly the states targeted with precious resources, such as television advertising or candidate appearances, are those that are at risk of going to the other side but are also critical for a winning coalition. 

Our review of historical campaign records shows that although they might not have had the data and tools that David Axelrod or Karl Rove had at their disposal, top campaign operatives from the 1950s and 1960s were just as smart and sophisticated about how to deploy their assets as campaigns are today. 

And while each election has its own dynamics, we find that some states are battlegrounds routinely. For example, Pennsylvania has been a strategic target for Republicans in every one of the past 18 elections, while Democrats have likewise targeted Pennsylvania in every year but 1968. Ohio has been targeted 17 times since 1952 by both Democrats and Republicans, while Wisconsin has been a strategic state 17 times for the Democrats and 15 times for the Republicans. It stands to reason that these states might be overrepresented when it comes to policy concessions from candidates.

Conversely, Hawaii has never been a battleground state, and Wyoming and Rhode Island have been strategic targets only once. Illinois falls somewhere in the middle: classified as a battleground state 13 times by the Democrats and nine times by the Republicans between 1952 and 1988, but not once by either side since. Because of this strategic neglect, this sixth most populous state has received little attention from presidential candidates over the last eight elections.

We also find that the more populous states — which deliver more electoral votes — used to be classified as battlegrounds more frequently than they are today. California, New York and Texas were usually classified as battlegrounds by campaigns between 1952 and 1972. But this tendency declined from 1976 through 2000 and disappeared altogether from 2004 through 2020. This is partly because these states are not as competitive as they once were and partly because the cost of campaigning in large states has increased exponentially over the past 50 years.

By the 2020s, the list of battleground states includes a mix of small (Nevada), medium (Georgia) and large (Pennsylvania). In this sense, the hope among the nation’s Founders that the Electoral College would prevent larger states from dominating the selection of the country’s chief executive has been realized, at least in elections since 2000.

If some of these larger prizes are no longer as competitive as they once were, one might think that campaigns would need to target more states overall. But our research shows that the number of battleground states has declined. Between 1952 and 2004, roughly 15 to 20 states were considered competitive in most elections. In the current era, the number is more like eight to 12 and seems to be dropping. Just seven battleground states in 2024 might be an abnormally low number, but it is not out of line with recent trends. 

People attend a campaign event with Republican presidential nominee Donald Trump on Sept. 7, 2024, in Mosinee, Wisconsin. A recent poll has Trump trailing Democratic nominee Kamala Harris in the battleground state. (Scott Olson/Getty)

Our research also shows how lopsidedly concentrated levels of campaign attention affect what happens on Election Day. On the one hand, the influence of TV ad spending and candidate appearances should fluctuate over time: Money is more available and transportation easier in recent elections relative to the 1950s. Campaigners today also have more data and more sophisticated methods for communicating with voters than in earlier years. On the other hand, voters today also have stronger attachments to their preferred political party, and connecting with voters in a complex, distracting digital world is much harder than it was during the golden years of television. 

We find that the typical effects of campaigns on Election Day vote totals are small, but potentially decisive. For example, media market-level data from 1988 to 2020 shows that a $100,000 shift in TV ad spending (in 2020 dollars) boosted a candidate’s share of the two-party vote by a little more than one-half of a percentage-point on average. Television advertising effects were larger between 1976 and 2000, when a $100,000 differential boosted a candidate’s vote share by a single percentage point, but from 2004 to 2020, the estimated impact fades to almost zero.

Likewise, a single presidential candidate appearance boosts the market-level vote by about one-quarter of a percentage point in the 1988-2000 period, and by about one-tenth of a percentage point in the 2004-2020 era. 

Even though the vast amount of candidate time and ad dollars spent on battleground states moves only a small number of voters on Election Day, we show that such small effects are probably enough to be pivotal when elections are decided by razor-thin margins. With nearly $500 million spent on presidential campaign advertising so far this year, even minuscule effects can move enough voters in key places to shift a state from blue to red or back again. Among the 18 elections held between 1952 and 2020, the outcome of seven would have flipped with a shift of fewer than 60,000 votes in just the right states.

This year, the primacy of battleground states and the obsession with Electoral College strategy among our political leaders have been on full display. One of the factors that probably convinced President Joe Biden to drop out of the race is that he did not see a pathway to 270 electoral votes. With seven battleground states in play and four leaning strongly to Trump (Arizona, Georgia, Nevada and North Carolina), Biden needed to hold all his base states (including Maine, Minnesota, New Hampshire, New Mexico and Virginia) and sweep the “blue wall” states of the upper Midwest. The ascension of Harris to head the Democratic ticket was almost assuredly due in part to her access to the Biden-Harris campaign war chest, which would allow her to advertise on TV in key states, and to her ability to stump aggressively in critical media markets, which Biden’s age and physical limitations precluded. 

For better or worse, Electoral College strategies have a clear and profound influence on American politics and policies. The supersized influence of battleground states in U.S. presidential elections is likely to persist as long as the Electoral College framework is in place. Should America decide one day to dismantle the Electoral College in favor of an alternative such as the national popular vote, it is possible that presidential candidates might distribute campaign attention more evenly. Even so, reforms along these lines could create new biases and incentives to concentrate resources in other ways. In the end, there may be no optimal solution. But that doesn’t mean we should stop looking for a better one.

Daron Shaw, professor at the University of Texas at Austin; Scott Althaus, professor at the University of Illinois Urbana-Champaign; and Costas Panagopoulos, professor at Northeastern University, are authors of the recently published book “Battleground: Electoral College Strategies, Execution, and Impact in the Modern Era.”

Submit a letter, of no more than 400 words, to the editor here or email letters@chicagotribune.com.

Related posts