Proportional representation, three ways. Paste each party's vote total — one party per line — set the number of seats, and this divides them up under the three methods real elections use: D'Hondt (highest averages, /1 /2 /3…), Sainte-Laguë / Webster (highest averages, /1 /3 /5…) and largest remainder (Hare or Droop quota). It shows all three side by side so you can see where they agree and where the method changes who gets the last seat. Everything runs in your browser; nothing is uploaded.
Write each line as Name: votes (a colon, comma or just a space before the number both work). Thousands separators are ignored, so 1,234,567 and 1234567 are the same.
The threshold excludes any party below that share of the valid vote before seats are shared out (0 = no threshold). The quota choice only affects the largest-remainder column.
In a proportional-representation election, parties win seats roughly in proportion to their votes — but seats are whole numbers, so a rule is needed to decide exactly how the fractions round. Different countries use different rules, and the rule genuinely changes the result, usually at the margin of the last seat or two. This tool runs the three most common families on the same numbers so you can compare them.
Four parties, 8 seats: A 100,000 votes, B 80,000, C 30,000, D 20,000 (230,000 total). Under D'Hondt the seats fall A 4, B 3, C 1, D 0 — the largest party A takes an extra seat and the smallest party D gets none. Under Sainte-Laguë and largest remainder (Hare) the same votes give A 3, B 3, C 1, D 1 — party D wins the final seat instead of party A. That single seat moving from the biggest party to the smallest is exactly the "large-party bias" people mean when they compare D'Hondt with Sainte-Laguë. Load the example above to see all three columns.
Both D'Hondt and Sainte-Laguë hand out seats one at a time. Every party has a current "average" = its votes divided by (its seats so far counted through the method's divisor sequence). The next seat goes to whichever party has the highest average right now; that party's divisor steps up and the process repeats until all seats are gone. Largest remainder is different: it does the division once with a fixed quota, then mops up the leftover seats by remainder size.
Which method is "fairest"? There's no single answer — Sainte-Laguë and largest-remainder track vote share most closely, while D'Hondt trades a little proportionality for a tendency toward stable, larger blocs. The point of showing all three is to see the trade-off on your own numbers.
What does the threshold do? Many systems require a party to clear, say, 5% of the vote to win any seats. Set the threshold and parties below it are dropped before allocation, and their votes no longer count toward the quota or averages.
Can I use it for non-elections? Yes — the same maths apportions anything indivisible in proportion: delegates, committee seats, even rounding a budget into whole units.
Is anything sent to a server? No. The whole calculation runs in your browser, so you can paste real figures safely.
This is one of a set of free voting calculators here: Borda count, ranked-choice / instant-runoff, Condorcet, STAR, and a compare-all-methods tool for single-winner elections. This page is the multi-winner, party-seat counterpart.