How Voters Actually Decide: Ideology, Loyalty, And The Limits Of Persuasion

The idea that voters choose candidates like diners picking from a menu—scanning policy positions, weighing them against personal preferences, and selecting the closest match—is intuitive but deeply misleading. Political scientists call this misconception the Strategist’s Fallacy. G. Elliott Morris named it in October 2025: the belief that voters hold positions and match them to candidates as if ordering an entree. The reality, as decades of research show, is messier. Voters do not think in left and right, at least not in the way campaign strategists assume. Understanding what an "ideological voter" actually is—and how researchers measure it—reveals why elections turn on loyalty, identity, and a tiny sliver of genuinely persuadable voters.

For most people, ideology is not a coherent belief system. Early research by Philip Converse found that only about one voter in nine reasons ideologically—meaning they can consistently apply abstract concepts like liberalism or conservatism across issues. Later work by Christopher Kinder and Cindy Kalmoe found that share has barely moved in the decades since. Morris puts the figure at 10 to 15 percent. That small group does think in left and right, and they vote accordingly. But for the other 85 to 90 percent, ideology is less a map than a vague mood. They may hold strong views on specific issues—taxes, abortion, immigration—but those views often conflict with each other, shift over time, or simply do not line up neatly with a party label. When researchers measure ideology, they typically use survey questions asking people to place themselves on a seven-point scale from very liberal to very conservative. But most respondents cluster near the middle, and their answers often do not predict their actual positions on individual policies. In short, ideology is a weak predictor of voting behavior for most people.

What predicts voting far better is party loyalty. Partisanship works like a tribal identity, learned early and reinforced by family, friends, and media. Voters ask not "What does this candidate offer me?" but "Who is my team?" This is why campaign persuasion has such a hard time moving people. In a landmark analysis of 49 field experiments, Joshua Kalla and David Broockman found that the best estimate of campaign persuasion in a general election is zero. Ads, canvassing, and mailers rarely change minds; they may mobilize or demobilize voters, but they do not convert them. This finding undercuts the entire premise of the Strategist’s Fallacy. If voters do not think in left and right, and if they are not swayed by arguments, then campaigns are less about persuasion than about turnout.

This does not mean swing voters do not exist. They do, but they are a small and shrinking group—often less ideological and less informed than partisan loyalists. True swing voters are not moderates weighing competing visions; they are people with weak attachments, low information, or cross-pressured identities. They may be persuaded, but only at the margins, and usually by events, not ads. The distinction between moderates versus progressives is a real one within parties, but it matters mostly in primaries, where the small ideological minority has outsized influence. In general elections, the broader electorate is not choosing between left and right; it is choosing between familiar and unfamiliar, comfortable and risky.

What political science says about how voters decide, then, is that elections are mostly a function of fundamentals: the state of the economy, national conditions, and the partisan baseline. Morris’s two-input fundamentals model forecast Harris at about 48 percent of the two-party vote in 2024; she got 49.3 percent. On conditions alone, Trump wins about nine times in ten. The three decisive states turned on 229,766 votes—a tiny margin that reflects not persuasion but the normal churn of a close race. Campaigns matter, but mostly at the edges.

Even so, the disputes left open moderation debate persists. Some argue that parties should move to the center to capture swing voters; others say base turnout is what matters. The research suggests both camps are partly wrong. Because voters do not think in left and right, moving ideologically may do little to attract persuadable voters, who are not paying close attention anyway. Meanwhile, legislators in both parties overestimate how conservative their constituents are, as David Broockman and Christopher Skovron found. That misperception pushes candidates to extremes that alienate the very moderates they think they need. WelcomePAC’s report Deciding to Win recommends a 2028 Democratic nominee willing to break with progressive advocacy groups—but whether that is a winning strategy depends on whether any such break actually reaches the small number of voters who are truly undecided. The Strategist’s Fallacy is not just an academic curiosity; it is a recurring trap for anyone who believes elections are decided by ideas. They are decided by identity, conditions, and a few thousand people who are not thinking about ideology at all.