We analyzed four voting systems and their benefits in day-to-day use. You can test them yourself in our voting simulator. Definitions of common voting terms and more information about the systems can be found in the resources page.
Plurality
This method is the best for any situation involving only two candidates, or when the voting process needs to be quick.
Ranked
Ranked voting is best in situations when the voter’s personal preferences need to be recorded, such as when people need to be sorted into groups.
Approval
This system is best for situations where people do not have enough information to choose or rank an option, or if the options could result in the exclusion of some group members.
Score
Score voting is perfect in situations where something or someone is being judged/reviewed. Any situation that requires each candidate to be evaluated individually would also work well.
Plurality
Plurality voting, commonly known as First Past the Post, is the most common voting system. It is used everywhere in the world, in all sorts of situations. In the United States, people are pushing for Plurality to replace the Electoral College system (The National Popular Vote Interstate Compact).
How to Count
Voters vote for one candidate each, and the candidate with the most votes wins.
| Pros | Cons |
|---|---|
| Requires only one choice | Trends toward two main parties |
| Always picks the candidate with the most overall votes | Causes strategic voting |
| Counting is easy and fast | Vulnerable to spoiler effect |
Conclusion
This method is the best for any situation involving only two options, or when the voting process needs to be quick.
All other systems eventually use Plurality in two candidate elections.
Something to note is that because this system is susceptible to the spoiler effect, voters may strategically choose not to mark their first choice on the ballot.
Voters may vote for one of the majority candidates rather than the minority candidate they prefer to prevent a majority candidate that they really dislike from winning.
Ranked Voting
Ranked voting is the most common alternative voting system to FPTP. IRV is gaining popularity in public elections and is currently used in some city council and state primary elections in the U.S. It was also used to elect legislators in Australia, Canada, Fiji, Hong Kong, and Papua New Guinea. Borda Count is less common but is used in some elections in Slovenia, Iceland, Nauru, and Kiribati.
How to Count
Voters rank the candidates from favorite to least favorite by marking which candidate is their first choice, which is their second choice, and so on. There are a few different voting systems that count ranked ballots, and while each system has its own flaws, any of them can be chosen to count votes using Ranked voting.
Instant Runoff Voting (IRV)
Instant Runoff is also called Ranked Choice voting and the Alternative Vote (although there are many more alternative voting systems). In IRV, first place votes are counted in the same way as FPTP. Then, the candidate with the least votes is eliminated, and the votes of those who voted for this candidate as their first choice go to whomever they voted for as second choice. This procedure gets repeated until one candidate has at least 50% of the vote.
| Pros | Cons |
|---|---|
| Reduces vulnerability to spoiler effect | Does not pass monotonicity criterion |
| Removes need for second rounds of voting like primaries and runoffs | Complicated and nontransparent counting method because voters’ second choices can become more varied among more candidates |
Borda Count
With Borda, the position of each candidate on each ballot gets added to the candidate’s total. For example, if a candidate was the first choice, third choice, and fourth choice of three different voters, the candidate’s total would be eight. The candidate with the smallest total is the winner.
| Pros | Cons |
|---|---|
| Simple to calculate because it only requires addition | May create a reverse spoiler effect |
| Removes need for second rounds of voting like primaries and runoffs | Causes strategic voting |
Condorcet Method
Condorcet simulates FPTP elections between every pair of candidates based on voters’ choices. For each pair of candidates, the number of voters who have ranked one over another are counted. In the example ballot image to the right, candidates Red, Yellow, and Green are ranked. Between Red and Green, Red has a higher ranking. If a majority of the rest of the ballots in this election also rank Red higher than Green, Red would win the matchup. The same is applied to the Yellow and Green pair and the Red and Yellow pair. The candidate who wins the most matchups wins the whole election.
| Pros | Cons |
|---|---|
| Reduces vulnerability to spoiler effect | Takes a long time to calculate on paper with lots of candidates because it requires a lot of pair ups |
| Passes favorite betrayal criterion | May fail to find a winner by creating a cycle, where no one candidate wins the most matchups. |
Conclusion
Ranked voting is best in situations when the voter’s personal preferences need to be recorded.
This works in situations where people need to be sorted into groups according to their preferences.
If one group has too many members, some voters can be sorted into other groups based on their second or third choices.
Thus, the voters are evenly split into groups that they want.
Something to note is that because spoiler effect may be present in some of the Ranked voting systems, voters may rank a candidate that is not their first choice as
their first choice. This may be because the voter prefers one majority candidate over another even though neither is their first choice.
Approval
Approval voting is not that common in general elections. It is currently being used in Fargo, ND and St. Louis, MO in the United States, as well as small towns in Western Europe. Approval voting is one of the only systems in which the number of candidates selected is up to the voter.
How to Count
Voters mark all the candidates they “approve of” or support. The candidate with the most approval votes/who is marked by the most people wins the election.
| Pros | Cons |
|---|---|
| Reduces vulnerability to spoiler effect | Preferences aren't indicated, so it does not satisfy the later-no-harm criterion |
| Prevents two-party systems | May not pass majority criterion, which can cause bullet voting |
| Counting is easy and fast because it uses the same, familiar ballot of Plurality | May pit candidates of the same party against each other to prevent vote splitting |
Conclusion
This system is best for situations where people do not have enough information to choose or rank an option,
or if the options could result in the exclusion of some group members.
Approval would be most useful in casual settings rather than public elections, when the decision has a small or temporary impact.
Something to note is that because this system is susceptible to the reverse spoiler effect, voters may not vote for options that they want but are not their
first choice in fear that they are reducing their first choice candidate's chance of winning.
Remember, voting for all of the candidates on the ballot using this system is the same as voting for none because each candidate gets the same amount of points from you.
Score
Score voting, also known as range voting, is a point-based system. Score voting is not in widespread use in any public elections, although it is quite similar to the systems judges at the Olympics use and the product rating systems found in e-commerce.
How to Count
Voters individually rate each candidate, often on a scale of one to five or one to ten. The rating of each candidate on each ballot is totalled and the candidate with the highest sum wins. This system seems similar to the Borda Count, but because each candidate is rated irrespective of the others, more than one candidate can have the same rank/score.
| Pros | Cons |
|---|---|
| A candidate’s score is not affected by the other candidates they go against | Vulnerable to reverse spoiler effect, leading to strategic voting |
| Fights against tyranny of the majority | A candidate well-liked by a minority may defeat majority candidates that are not as polarized |
Conclusion
Score voting is perfect in situations where something or someone is being judged/reviewed. Any situation that requires each candidate to be evaluated individually
would also work well.
Something to note is that because this system is susceptible to the reverse spoiler effect, voters may rank candidates that are not their first choice lower than
they really believe in fear that they are reducing their first choice candidate's chance of winning.
For example, they may give all candidates except their first choice the lowest possible score (usually zero or one) and their first choice the
highest score (usually five or ten) even if they those numbers don't reflect their true feelings.
