If you were trying to design the worst way to vote, you might:
- Force voters to say the least possible amount – name just one candidate, and say nothing about how much you like or dislike any of the others.
- Make it reward voters for not voting for whom they really want.
- Make it operate, over time, in such a way as to diminish your number of choices to the minimum – only 2 (or 1, meaning no choice at all).
- Make it easy for fraudsters to invalidate ballots ("overvoting").
FIGURE illustrates a hypothetical ballot on|
a score voting machine's display screen.
There's a better way: score voting (also known as "range voting").
You've all seen score voting in action as the Olympic scoring system. Judges give the competitors scores and the highest average score wins. Similarly, in a score voting election, voters would give the candidates scores, and the one with the highest average would win.
Score voting permits voters to express their opinions about any number of candidates (not just one). It eliminates the "spoiler," "wasted vote," and "candidate cloning" problems. All candidates compete on a level playing field, whether Democrat, Republican, Independent, or other. It's simple enough to run on all of today's voting machines and to be used by kindergarteners.
- Each vote consists of a numerical score within some range (say 0 to 99) for each candidate. Simpler is 0 to 9 ("single digit score voting"). Voters may also indicate "X" or "NO OPINION" if they have no opinion about a candidate. Such votes don't affect that candidate's average.
Andrews 0 1 2 3 4 5 6 7 8 9 NO OPINION Benson 0 1 2 3 4 5 6 7 8 9 NO OPINION Carey 0 1 2 3 4 5 6 7 8 9 NO OPINION Davis 0 1 2 3 4 5 6 7 8 9 NO OPINION Elbert 0 1 2 3 4 5 6 7 8 9 NO OPINION
- The candidate with the highest average score wins... except
- If desired, the candidates can pre-agree that each is to be awarded T fake votes using score S (for some pre-agreed values of S and T) before the election begins. The highest average based on both the real and fake votes, wins. This is fair since S and T are the same for every candidate. (Why do that?)
- To educate the public about the advantages of score voting and comparative disadvantages of other systems (especially the USA's embarrassingly poor current "plurality" system),
- To lobby for the adoption of score voting.
- To assist anyone looking to use Score Voting for an election, whether for public office, student elections, or anything else.
But CRV also supports research into (and is open-minded about) all voting systems. Indeed, CRV, for some purposes, even advocates other systems. CRV is nonpartisan.
Virtues of Score Voting (& criticisms) (history) (comparison chart) (criteria)
- EXPRESSIVE: You give information, about all candidates not just one.
- QUANTITATIVE: Your vote not only says that you prefer candidate A over B, but also by how much.
- ENCOURAGES HONESTY NOT STRATEGY: Your score for candidate C in no way affects the battle between A vs. B. Hence, you can give your honest opinion of C without fear of "wasting your vote" or hurting A. You never have an incentive to betray your favorite candidate by giving a higher score to a candidate you like less. (This may sound like an obvious and easy criterion, but very few other voting systems obey it!) The decreased importance of "convincing voters you can win" (as opposed to "convincing them you are the best") may decrease the importance of cash.
- UNAFFECTED BY CANDIDATE CLONING: Consider the situation where A has "clones" A2 and A3. In the old "plurality voting" system, the clones "split the vote" and lose. In the "Borda voting" system, a party assures victory merely by running enough clones. In contrast, in Range voting, A is neither harmed nor helped – no more bitter enmity between similar candidates.
- BREAKS STRANGLEHOLD OF 2-PARTY DOMINATION SO VOTERS GET MORE CHOICES: Plurality & "Instant Runoff Voting" yield 2-party domination. "Approval voting" improves over those systems by allowing voters to vote for all the candidates that they approve – with no built-in penalty for approving a third-party candidate. But range voting is even better, since it empirically yields much greater support for 3rd parties than approval voting. Having more parties should also decrease the importance of gerrymandering, and increase interest & turnout.
- CURRENT VOTING MACHINES CAN HANDLE RANGE VOTING: That can't be said of IRV and Condorcet.
- THE "BEATS-ALL" CANDIDATE WINS: A beats-all candidate is one who would beat every other candidate in one-on-one races. Counterintuitively, in many voting systems "beats all" candidates do not necessarily prevail. (Indeed, a "lose to all" candidate can easily win under our current system.) But in range voting (under certain assumptions about strategic voter behavior) the beats-all candidate always wins.
- MAXIMIZES PLEASANT SURPRISE: A voter is "pleasantly surprised" when the election winner is better than she expected. It is a theorem that range voting (under certain assumptions about strategic voter behavior) maximizes the number of pleasantly surprised voters.
- FEWER "SPOILED BALLOTS" and LESS FRAUD: In plurality voting, an accidental hanging chad or overvote can cause your vote to be discarded. With range voting, it causes only part of your vote to get discarded (converted to "no opinion") with the rest – your opinions of all the other candidates – still operational. There's no such thing as an "overvote" since expressing opinions about all candidates is the idea; there's no wrong way to fill in a ballot. Score voting also reduces risk of TIES.
- USING OBJECTIVE CRITERIA, RANGE VOTING IS BEST: For the mathematicians... Range voting has the lowest BAYESIAN REGRET among all common single-winner election methods. (Found from massive computer simulations for all 720 tried combinations of the following: number V of voters and N of candidates with V>N>2, several kinds of "utility generators", several amounts of "voter ignorance," and with either "honest" or "strategic" voters.)
- HUGE POSITIVE IMPACT: "Democracy" improved over monarchy and feudalism. We can express that improvement as a number by using the Bayesian regret yardstick and assuming, e.g, that those nondemocratic systems were at least as good as "random winner." Measurements on this same yardstick suggest that switching from plurality to range will cause at least as much further regret reduction.
- When you consider "INDIRECT EFFECTS" invisible to Bayesian Regret ("direct" effects happen in one election, indirect ones over a sequence of many elections) that usually increases Score Voting's edge over competing methods. (Examples: 2-party domination and "nursery effect.")
- UNBIASED: Plurality and IRV both have natural pro-extremist biases, while approval has pro-centrist bias. But score-voting has little or no discernible such bias.
- NATURE USES RANGE VOTING: Honeybees & ants have run trillions of elections over the last 20-50 million years to make hive relocation decisions. Under severe evolutionary pressure, they came up with... range voting! (Cf. Bonobos.)
- VOTERS WANT IT! Referenda and poll studies show that voters want score voting and would enact it by landslide (≈60%). However they do not want various other systems including IRV.