How to Think about Voting

Bryan T. McGraw & Timothy W. Taylor

Winter 2024

One striking feature of the last few election cycles was the revelation for some that voting is not an especially simple act. It should be straightforward: Figure out which candidate or party best fits one's political views or interests, and vote accordingly.

But what if neither candidate nor party even remotely aligns with one's views? What if all, or at least all the mainstream, candidates take positions that a voter finds deeply unacceptable? What if that voter finds some things to like (and dislike) about one candidate or party and other things to like (and dislike) about another, and it proves difficult — maybe impossible — to rank the two? How can a perplexed citizen feel confident voting at all in these circumstances?

Citizens of a representative democracy must undertake moral and political work that should not be considered easy or simple. The act of voting, like so many civic responsibilities, is neither. In choosing who should represent and exercise political authority over the public, voters must make any number of complex, ambiguous, morally fraught judgments that genuinely matter, even if they are usually immaterial in terms of wielding real political power.

Americans could thus benefit from a new framework for thinking about voting, and not just for elections in which they find themselves especially bewildered (or worse), but more generally for any set of electoral choices.


To start, Americans need to think about what they are doing when they vote.

Assume that balloting is taking place in a relatively sizeable jurisdiction where thousands are voting for a chief executive or legislator, and that the candidates are mostly well-connected to particular political parties with a reasonably (though not entirely) coherent set of moral-political commitments. In those sorts of circumstances, what are voters actually doing?

The temptation is to think of voting as a means of achieving the best possible outcome — call this "strategic voting." This is not an implausible way of understanding voting; indeed, the person with the most votes in an election usually wins, and to the degree that one's vote contributes to that victory, a voter has helped make a desirable outcome (according to his preferences, at least) occur. This is the way most people think about voting. But it presents problems.

The first is that in any decent-sized jurisdiction, one voter's capacity to effect a particular consequence is so small as to be statistically insignificant. Consider the approximately 4.8 million people in Michigan voting in the 2016 presidential race. Donald Trump ended up winning the state by a mere 0.3%, or around 11,000 votes. Even in such a close race, an individual's vote is just not significant. And that's in the event that one's candidate or party wins; if he loses, it's hard to say that one's vote mattered at all, except maybe as a small part of a supposed "statement" about what "some" people think or support.

Strategic voting faces other issues. When voters seek to accomplish some political end by supporting a candidate or party, they often won't have any reasonable assurance that the outcome they are voting for will, or even can, occur. Candidates and parties have been known to mislead voters about their intentions, after all. They also overpromise, underestimating the practical and institutional hurdles standing between them and their stated policy goals. When they do genuinely want to enact sweeping (or even smaller) reforms, doing so is much harder than they imagine.

Barack Obama learned this after vowing to shut down the military prison at Guantanamo Bay. Despite his efforts to fulfill that promise, he failed. He encountered institutional constraints, and he eventually determined that it was not worth spending his political capital on the issue when he had other priorities. Even when politicians are being genuine — and they are often not — voting to effect policies by choosing one candidate over another is at best a highly uncertain game.

In short, one's individual vote does not (really) matter, and strategic voting is a fool's errand. That citizens have a difficult time seeing these problems with strategic voting is understandable. Voters like to imagine themselves as having significant amounts of political power — they make up "the People," after all, and in democratic systems, the people have the final say, right? Except that it is almost never "the People" as a whole who have a say, and majorities almost never determine the final outcome. Even if it appears that they do, majorities are ephemeral.

In the end, claiming that "we" have the final say in much of anything smacks of hubris. For Christians like the authors of this essay, such sentiment runs afoul of the reality of God's providential care of the world and His actual final say. Something similar is true for those who don't share our faith: Framing voting decisions solely in terms of their consequences and strategically trying to bring about one outcome over another puts people in danger of supposing that they somehow control and order History, effectively making themselves objects of their own worship. This sort of idolatry puts human beings in the position of denying reality. The truth is that history always escapes our grasp; we are but bit players in its execution.

Our analysis here might sound like a counsel of despair, an Eeyore-like take on voting: "It doesn't really matter anyway." To the contrary, we believe that thinking aright about voting offers Americans real freedom in exercising their responsibilities as democratic citizens. If the seductive mirage of strategic voting is set aside, voters no longer carry the whole weight of responsibility for an election's outcome. They are thus free to exercise their civic responsibilities — including voting — sincerely, to think about it as shaping their moral character and as an expression of their best sense of what they hope America's social and political order can look like.

Most good-willed people endeavor to live what some have called an "integrated" moral life, meaning that they try to act in such a way that their lives make sense as a moral whole. Although it is nearly impossible for humans to never compromise or get caught in various moral contradictions, most people strive to avoid them. People sometimes face genuine moral dilemmas — situations in which all options appear to involve morally questionable or even evil outcomes. But otherwise, they generally attempt to act in ways that cohere with their other basic or considered moral views.

If this is accurate, then voting should be thought of not in terms of its consequences per se, but as one aspect of an integrated moral life. That is, voting should be considered an expression of something about one's moral convictions, and citizens should work to make their votes congruent with those convictions. We call this "sincere voting."

Voting sincerely does not mean citizens are merely expressing themselves, as if voting were a kind of vanity project in which people have the opportunity to demonstrate their moral bona fides to anyone who will listen. It is not, in other words, electoral virtue signaling. Instead, citizens expressing themselves sincerely via their votes means they are saying something about how they would like their country (or state, etc.) to look or act. Given who we are morally speaking, in voting we are (or should be) demonstrating that we think this person or that party would best direct our political community toward its proper ends.

Voting is thus more akin to a speech act than an exercise of power. To vote in ways inconsistent with one's broader moral convictions is to speak and act in a manner at odds with oneself; it is to live and act duplicitously. Again, times arise when such moral dilemmas appear inescapable, but those are extraordinary circumstances in which choices demand a high level of justification. Sometimes one really does have to ally with the Soviets to beat the Nazis. But those times are few, and Nazis are not around every political corner.

One oddity in thinking about voting this way is that its expressiveness is not always (or even often) public. Voting is confidential, and though communication is dominated by social media — where too many people tell the world too much — the primary audience in voting-as-expression is oneself. Indeed, given the anonymity of voting, no outside audience can decipher any virtues demonstrated through one's ballot; only an individual voter can understand (insofar as he can know) why he made the judgments he did and what moral sensibilities those judgments are meant to reflect.

When voters do attempt to explain their votes to others, it is not uncommon for their explanations to be misunderstood or misconstrued. But to the degree that citizens understand themselves and their voting as expressing their sincere sense of what they prefer their community to look like, they will in turn find themselves shaped by those same votes. That is, insofar as voting reflects one's moral convictions, how citizens vote will work in turn to shape those same convictions, if for no other reason than the calling humans feel to live lives that are something like an integrated whole.

People can (and do) live with tremendous hypocrisies. But if they vote in ways contrary to what they claim to be their core convictions, the next step likely involves smoothing out those contradictions — and perhaps adjusting those purportedly foundational moral beliefs to match their voting.

Conceiving of voting as an exercise of self-development may strike some as self-centered or narcissistic. Yet the notion that how people act profoundly shapes whom they become is quite reasonable. Imagine a voter who decides to support a candidate or party even though doing so appears to violate one of that voter's core convictions. That he might retroactively justify his vote by realigning his own positions with the candidate's or party's views is not surprising. To declare that voting could cost a person his soul might suggest too much. But the willingness to compromise one's deepest commitments by voting strategically can indeed damage it.

So far, this discussion has remained abstract, and perhaps not all that helpful to those perplexed about how to vote in any particular election. Who, after all, goes into the voting booth committed to defying his basic moral convictions? How does this argument in favor of sincere voting help citizens reach decisions about whom to vote for?

We can start with the premise that as part of citizens' moral identity, they should have a sense of what a morally decent society looks like. The obvious corollary to that view is that they should vote in a way that expresses and advances those same convictions. This means at least voting in a manner that contributes to the creation of a society ruled by a constitutional order — one that secures a reasonably just set of social, political, and economic institutions that can persist over time. People will of course disagree about what constitutes a just society, but they should nonetheless vote in a way that is consistent with their own sense of what their moral integrity demands. To paraphrase former president Obama, in voting, we should be the sort of just society we seek, at least on this earth.


Citizens who seek to vote based on their vision of a just societal order nonetheless face the same problem mentioned above: How should they think about voting when neither candidate nor party presents as acceptable?

Often times in these cases, citizens become strategic voters and embrace a "lesser of two evils" approach. A voter might say: This issue is more important than that issue, and because I agree with this candidate on the more important issue, I will go ahead and vote accordingly, even though I find myself seriously disagreeing with him on several of his other policy positions.

That is not a terrible way to think, and if given a choice between two candidates who are more or less satisfactory on a range of issues but sharply disagree on one important one, the presence of a candidate with a view deemed deeply immoral in that one policy area makes the decision relatively straightforward. But what about a situation in which neither candidate is acceptable on what voters consider crucial issues? In this case, just trying to rank candidates based on their views on important questions may not be enough to responsibly guide voters' choices — they need something a bit more systematic to help. They need what we call an evaluative framework. This framework should function on at least two levels: the constitutional and the political.

At the highest level, citizens' goal should be to vote in such a way that expresses their sense of how their society should be governed in its most fundamental way. That is, they should evaluate candidates (or parties) based on how they will shape and develop America's constitutional order — the basic institutions and practices, both formal and informal, that set the terms and conditions for everything else.

One does not have to agree with Aristotle that politics is "architectonic" in any strong sense to understand that a society's basic institutions will influence everything else. A constitutional order, more so than social or political institutions more generally, is hard to change, and if candidates or parties declare an intent to damage or warp (or refuse to reform) that order, this presents a good reason not to vote for them. Conversely, if a candidate or party appears genuinely committed to reform that will move America's constitutional order in a more just direction, that might be a reason for voters to override preferences they might have for another party's or candidate's policy positions.

But what if both candidates pose a fundamental threat to the constitutional order — say, through either communist or segregationist platforms? Neither of the candidates would embrace a basic moral vision that could in any way be thought just, and no one, in our view, should vote for either person. Because citizens are free to vote according to their conscience, they can also choose not to vote — to abstain when the choices available are clearly immoral.

Despite the catastrophizing rhetoric that pervades today's politics, most of the people's electoral choices (thankfully) are not of this sort. Instead, most of the public's disagreements occur on the level of policy differences. Policy disputes can range from the fairly trivial (should the marginal tax rate be 39% or 39.5%?) to the more substantial. And the point at which a policy difference turns into a constitutional one can be difficult to discern.

A good rule of thumb might be to think about how to change or reverse course on a particular issue. If it would require a sea change in the political order that would be all but impossible to reverse, it is a constitutional issue. Recognizing same-sex marriage certainly counted as such an issue, as did the 1964 Civil Rights Act. These shifts would be difficult to overturn as a matter of political procedure because they reshaped the underlying social order such that one cannot imagine going back.

Policy choices, on the other hand, might indeed be "sticky" and thus tough to change, but they could at least plausibly be altered with a different party in control of Congress or a different president. Think here of the ways in which former president Donald Trump pursued a new direction in U.S. trade policy — a shift that President Joe Biden has largely embraced, but one that could be backtracked with a different Congress and president.

Voting based on policy differences among candidates or parties is much more difficult and complex than voting with regard to constitutional issues, for at least two reasons. First, aside from direct popular votes on referenda, citizens do not actually vote for policies as such, but instead for candidates or parties, and those candidates and parties do not have preferences on just one policy, but rather on a whole bundle of them. Finding a candidate or party whose preferences exactly match one's own is rare, especially at the national level. Voters might find themselves preferring one candidate's stances on some policies but another candidate's stances on others, and being perplexed as to how to choose between them.

Second, precisely because voters select candidates and not policies directly, they end up voting in effect for one political party over another. Democratic politics is an inherently partisan affair, and political parties are a necessary and even beneficial part of the process. But times may arise in which voting for one candidate means essentially endorsing a party whose broad inclinations one finds disreputable. This is especially true when voting at different levels of government, where the parties themselves vary depending on the region — a Republican in Texas might have rather different priorities than a Republican in Massachusetts, for instance.

These kinds of difficulties multiply the closer one looks at voting. Exercising the responsibilities of democratic citizens simply cannot be reduced to some formula or algorithm — that sensibility is a profound mistake. It implicitly supposes that there must always be a "clean," straightforward answer to the nation's political dilemmas, that citizens are thus morally bound to vote in just one particular way, and that those who think differently must therefore be malicious.

Some policy preferences are certainly difficult to defend morally, and citizens might sometimes act out of malice. But most political disagreements are not the result of malevolence; they are just that — disagreements born of differing judgments about complex and difficult questions. No such algorithm can correspond to the messy reality of the political world.


If, as we contend here, sincere voting is morally superior to strategic voting in guiding an individual's preference at the ballot box, are there ways to make it easier to vote sincerely? The 2016 U.S. presidential election provides an instructive example to start with.

In 2016, the two major-party nominees, Hillary Clinton and Donald Trump, had the lowest favorability of any candidates in presidential polling history. Despite this, third-party vote share did not fare much better than historical averages; fewer than 5% of voters selected a candidate from a minor party. Moreover, if states in which the vote was close are compared to those in which Clinton or Trump won by large margins, there is no statistical difference in third-party vote share. In other words, voters acted strategically even in states that were so overwhelmingly red or blue that no presidential candidate (or vice-presidential candidate, for that matter) bothered to step foot in them.

Why would millions of citizens vote strategically when they say they dislike the candidates and when their individual votes are inefficacious? Many likely internalized their fifth-grade civic instruction to vote as if their choice were decisive. Others simply vote for a candidate or party almost by inertia: That's just how they vote. But voting patterns also reflect institutional design, and the country's electoral laws could be reformed to incentivize more sincere voting.

Although numerous candidates are listed on a ballot, American voters in almost all jurisdictions may select only one name. And when a voter selects his preferred candidate, it is impossible to know which of the remaining candidates was his second or third choice — or his absolute last choice. The ballot, in a sense, is like a wet cloth soaked with information. American elections as they currently operate squeeze only a few drops of data from the ballot. What if instead they clenched the ballot and drained a torrent of voter preferences from it?

Ranked-choice voting is one such law that not only allows the voter to express more preferences, it also encourages sincere voting. In a ranked-choice vote, voters may rank each candidate from their first to last preference. Votes are then tallied, but only first-preference votes are counted to determine if a candidate wins a majority. If no majority is secured (which is often the case), the last-place candidate is dropped and all of his votes are reallocated to his voters' second preferences. If there is still no candidate with a majority, the new last-place candidate is dropped and his votes are reallocated to the candidate next on the ranked list. This process continues until a candidate wins a majority of the votes.

Under these sorts of electoral rules, voters don't vote "against a candidate"; they can simply rank their most despised candidate last. A voter is then compelled to weigh the remaining candidates to provide a preference ranking. Strategic voting might still lead a voter to rank the remaining candidate with the best chance of winning first. However, once the reviled candidate is ranked last, a voter is not punished for voting sincerely among the remaining candidates. Because the candidates frequently fail to secure a majority of votes in the first round of counting, a voter is free to rank a minor-party candidate first, knowing that if his preferred candidate finishes last, his vote "survives" to be reallocated to his next preference. Voting for a minor candidate under ranked-choice voting is thus less likely to be a "wasted" vote than doing so under plurality rules.

Another strength of ranked-choice voting is that it frees voters from having to strategize about how to ensure that their least-preferred candidate loses. If a majority of voters concur on their least-preferred candidate under this system, that individual cannot win the election in any scenario. Plurality elections, by contrast, cannot guarantee such an outcome. In the 2015 general election in the United Kingdom, for example, fewer than 40% of British voters favored the Conservative Party, but it gained a majority with 51% of seats in Parliament. Under plurality rules, if a majority of voters agree on which candidate (or party) is their last preference but cannot agree on their first preference, it is possible for the least-preferred candidate to win the election.

Ranked-choice voting has a real potential to moderate electoral outcomes. Given that candidates under existing plurality rules win simply by accruing the most votes, they do not have much of an incentive to pursue coalition-building across their field of competitors. In fact, if a candidate is polling favorably with less than a majority of voters, he might strategically sow discord among the other candidates to split the majority. Ranked-choice voting, on the other hand, enables a candidate to win office by appealing to a competitor's voters and asking to be listed second in their rankings. Such moderating forces would be a welcome change for America's polarized politics.

Expanding the use of ranked-choice voting is the most straightforward reform that would incentivize sincere voting. In 2016, residents of Maine simply replaced their plurality election laws via ballot initiative. Other proposed changes, such as shifting to the system of proportional representation used in the Netherlands, would require significant amendments to the Constitution to implement. More novel proposals, such as allowing citizens to cast their vote either for or against a candidate on the ballot, have also gained attention. While permitting votes against candidates would promote sincere voting, without evidence from actual elections, we are reluctant to endorse any proposal that might increase political polarization among the citizenry.


Voting is not an easy task — nor should it be. Although Americans' individual votes rarely appear to have an impact, voting remains one of their most important civic responsibilities. Rather than adopting a disengaged cynicism, citizens have the freedom to vote sincerely and make their voting an expression of their most basic moral commitments. When they must make difficult choices between competing candidates, the general framework outlined here can help them differentiate the merits of a candidate's policy positions from whether that candidate is likely to upend the constitutional order.

No single reform, of course, can fix what ails American politics. But making voting — that quintessential democratic act — a bit easier morally would surely be a step in the right direction. Implementing ranked-choice voting is one promising proposal that could make it easier for citizens to vote sincerely and help repair some of the more dangerous divisions in our democracy today.

To say that citizens must make difficult, complex, and fallible judgments about their voting choices, and that many of those choices will have trade-offs and unintended effects, does not turn voting into a tragic act. Instead, this approach recognizes not only the tragedy of operating in a finite world of moral and material scarcity, but also the freedom that voters have to make those judgments without the existential worry that everything always depends on them.

Americans have responsibilities as citizens and are called to exercise them with wisdom and charity. These include affirming with their votes as best they can a vision for how their political communities can thrive. They do not bear the burden as individuals of always knowing fully what to do, and so they are free to make the best choices they can and recognize that their fellow citizens are doing likewise. Maybe in that freedom Americans can find a bit of wisdom about not just constitutional or policy choices, but also how to live together when they each know just a little and all disagree over so much.

Bryan T. McGraw and Timothy W. Taylor are associate professors of politics and international relations at Wheaton College.


from the


A weekly newsletter with free essays from past issues of National Affairs and The Public Interest that shed light on the week's pressing issues.


to your National Affairs subscriber account.

Already a subscriber? Activate your account.


Unlimited access to intelligent essays on the nation’s affairs.

Subscribe to National Affairs.