Our Voting System Is Broken — The Problem with Plurality¶

Voting 101 · foundational¶

The diagnosis — the "why fix anything at all?" that comes before the spoiler-effect mechanism and the STAR pitch. Four honest questions decide whether Choose-One Plurality is worth replacing: is it prone to failure, have the failures been frequent, were they serious, and are they getting worse? This page answers all four, then shows why STAR settles on a single ballot the one question Plurality can't.

Prefer it spoken? The narrated pitch is What's so good about STAR; the slide version is the Why STAR Voting deck.

Want the problem broken into separable pieces instead? The problem, in four causes splits it into four independent failures — the closed primary, the single mark, vote splitting, and a count that ignores what you wrote — each with its own fix, and a grid showing which reform fixes which. This page argues that Choose-One is worth replacing; that one maps where the influence is lost.

Is the system actually broken?¶

The fundamental problem with Choose-One Plurality isn't really in dispute: the rules let a candidate who is the least preferred choice of a majority be declared the winner. A majority can be against you and you can still take office — just by leading a divided field. That plainly violates majority rule, and nobody seriously argues it's democratic. The only open question is whether it's worth doing something about — so take the honest version of the skeptic's four questions in turn.

"But usually the front-runner really is the favorite, right?"¶

Usually — yes, and it's worth conceding plainly. When you have richer ballots to check against, the plurality front-runner usually turns out to be the broader choice too: in Australia's ranked elections the first-preference leader goes on to win about 90% of the time. But flip that around — roughly one race in ten is decided differently once you look past first choices. And here's the catch: a Choose-One ballot gives you no way to tell which kind of race you're in. From the marks on the ballot you can't know whether your plurality winner is the majority's real choice or an accident of vote-splitting. The ballot itself hides the answer — and that hidden answer is exactly what STAR puts back on the table.

How often, and how serious? — the two-party mask¶

If it's only one in ten, why has nobody noticed? Because for decades two dominant parties masked it. Squeeze almost every race into two big tents and you rarely see a third candidate split the vote — the flaw was real but invisible. That has been changing. The moment serious independents show up — Perot-style or otherwise — we snap right back to plurality winners instead of majority winners. In 1992, with Perot in the race, Bill Clinton won an outright majority of the popular vote in exactly one state — his home state of Arkansas (plus the District of Columbia). Every other state was carried with a plurality, not a majority: a president elected while 49 states out of 50 never gave any candidate more than half their vote.

That's not a fluke of one strange year — it's structural. The instant voters have more than two real choices (which is exactly what most people say they want), Choose-One starts handing out minority winners and punishing the voters who dared back a third option. That's vote-splitting and the spoiler effect, and as the two-party grip loosens it's getting more common, not less.

So: prone to failure, yes; serious when it happens, yes; and trending worse. Three yeses — which is why it's worth fixing.

Worked demo. 01_political_left_split (.yaml) — a 60–66% coalition splits and Choose-One hands the seat to a candidate the majority ranked last; the engine's [Vote-splitting check] block says it in numbers.

How STAR fixes it¶

STAR removes the two things that break Plurality. First, you're never punished for honesty: you score every candidate 0–5, so backing your favorite and a compromise can't split your own side. Second — and this answers the skeptic's real question — STAR doesn't just hope the winner has majority support, it checks. The automatic runoff is a head-to-head majority test between the top two, and the preference matrix shows exactly who beats whom. The very thing Choose-One hides — "is this winner actually the majority's choice?" — STAR prints right on the result, from one single ballot.

Under Choose-One you can't know. Under STAR you read it off the result. That's the whole difference between hoping for majority rule and measuring it.

Worked demo. 04_star_wars_vote_split (.yaml) — Choose-One elects Vader (40%, the candidate 60% ranked last); STAR elects Leia, the majority's real choice, and the [Divergence from STAR] line shows Choose-One (Plurality) disagreeing with STAR while STAR matches the head-to-head (Condorcet) winner.

The one-liner¶

"Choose-One can crown the candidate a majority likes least, and the ballot itself won't tell you when that's happened. STAR refuses to split your vote, then proves the winner has majority support — on one ballot, in plain sight."

Where this fits in the overall teaching¶

Level: 101 · for voters — foundational. This is the diagnosis; read it before the spoiler-effect mechanism and the STAR pitch.
Pairs with: the spoiler effect and the split_voting/ demos — this page argues that the system is broken; those show how. Then What's so good about STAR delivers the fix.
Terminology: keep it Choose-One / Plurality for a public audience; the point is method-accurate for Plurality specifically. (Reworked from a classic pro-reform passage that originally leaned on IRV; here the resolution is STAR's majority runoff + preference matrix, which answer "is the plurality winner the real winner?" directly, on one ballot.)

Cross-references: - GLOSSARY — "Choose-One / Plurality / First-Past-The-Post," "Plurality / minority winner," "Vote splitting," "Spoiler effect," "Majority finish." - Why STAR Voting — Part 1 problem bullets. - What's so good about STAR — Segment 1 (the spoiler). - LINKS.md → Full Deck 2025 ("PROBLEM: VOTE-SPLITTING," "THE VOTING DILEMMA"), Torrance LWV (Arend) ("SPOILER EFFECT").