Proportional STAR uses the same STAR ballot as single-winner STAR. What the ballot asks of voters is also the same: rate favorite candidates high, opposition candidates low or not at all, and favorable candidates in between. What is different, and it makes all the difference, is the formula that operates under the hood, a formula named Allocated Score. In multimember districts, Allocated Score is the mathematical engine that delivers seats to candidates in proportion to how voters scored them on the STAR ballot. This article covers the basics of the Allocated Score engine without getting bogged down in the accounting details.

The best way to see how it works is to watch it work. We'll walk through a simplified election — three seats, four candidates, sixty voters — and introduce the key ideas as they come up naturally.

The Ballot

Here is the ballot for our example district. Four candidates are running for three seats. Before reading any further, take a moment to look at the Party Endorsements column. One row is different from the others.

Congressional District 7 — 3 Seats · STAR Ballot
Candidate Party Registration Party Endorsements Rating
Martinez Democratic Democratic ⓪ ① ② ③ ④ ⑤
Chen Republican Republican ⓪ ① ② ③ ④ ⑤
Olson Working Families Working Families, Green ⓪ ① ② ③ ④ ⑤
Patel Green Green ⓪ ① ② ③ ④ ⑤

Olson's row shows two endorsements — Working Families and Green — even though Olson is registered with only one of those parties. That second endorsement is a signal. It tells Green voters: our party believes this candidate represents our interests. It is visible on the ballot before anyone votes, so voters can factor it into their scores. This is what a coalition looks like at the ballot box.

Patel, Green's own candidate, shows only one endorsement — Green. No other party endorsed Patel. We'll see why that asymmetry matters when the seats start filling.

The Setup

Sixty ballots are cast. The voters sort themselves — not by registration, but by how they scored. Here are the four groups and their sizes:

Dem 24 voters
Rep 15 voters
WF 13 voters
Grn 8 voters

Three seats, sixty voters. The price of a seat — called the quota — is total ballots divided by the number of seats: 60 ÷ 3 = 20. Every seat costs the same. This price is never checked before a seat is awarded — it only matters afterward, when the cost is processed.

Every voter starts the count at full influence. Think of it like the fuel gauge in your car: full at the top, empty at the bottom. When your candidate is selected and the cost is processed, your gauge drops — you spent some of your influence on representation you received. The gauge tracks what you have left to spend on the seats still to come.

Ballot weight — start of count
Dem
1.00
Rep
1.00
WF
1.00
Grn
1.00

How the Count Works

Each round, the Allocated Score formula does two things — and keeping them separate is the key to understanding everything that follows. Think of it as two windows at a service counter, each handling a different job:

🪑
Window 1
Selection
The candidate with the most support is awarded the seat. No mention of price — just, you have the most support, this seat is yours.
⚖️
Window 2
Accounting
The cost is processed. The quota is charged against the voters who backed the winner. Their gauges drop. Whatever influence remains is their change.

The person at the first window never touches the price. The person at the second window never decides who won. Selection and accounting are two different operations, and they happen in that order — every round, without exception.

The Election

Seat 1 of 3Martinez — Democratic
Dem 5★
QUOTA
price: 20
CUSP
surplus above the line ↑
Martinez Democratic
Window 1 — Selection

The count sums the support behind each candidate. Democratic voters — the largest group at 24 — all gave Martinez their highest score. No other candidate comes close. Martinez is selected and awarded the first seat.

Window 2 — Accounting

Now the cost is processed. The price of a seat is 20, and 24 voters backed Martinez. More than enough. The beaker — the accounting tool — fills with Democratic support past the quota line, so the cost is covered entirely within this one group. All 24 voters share the charge equally: each one's gauge drops by the same amount.

Because there were more voters than the price required, nobody is fully spent. Each Democratic voter gets change — a small fraction of their influence carries forward. Their gauge drops from 1.00 to about 0.17. Reduced, but not empty.

Ballot weight — after Seat 1
Dem
0.17
Rep
1.00
WF
1.00
Grn
1.00
Seat 2 of 3Olson — the coalition seat
WF 5★
Green 4★
QUOTA
price: 20
CUSP
WF fully allocated ↓
Green is the cusp ↑
Olson Working Families + Green
Window 1 — Selection

Working Families' 13 voters all gave Olson five stars. Green's 8 voters — honoring their party's endorsement — gave Olson four stars. Together, that is 21 voters' worth of full-weight support behind Olson. Republican's Chen has 15 voters at full weight. 21 is more than 15. Olson is selected and awarded the second seat.

Window 2 — Accounting

Now the accounting shows something new. The beaker fills from the bottom, heaviest tokens first. Working Families' five-star support settles in — 13 voters. But 13 falls short of the price of 20. The quota line is still above them. Their entire contribution is consumed. Working Families voters are fully allocated — their gauges drop to 0.00. They got their representative.

The remaining cost — 7 more units — falls to the Green tier above. Green's 8 voters gave four stars, so they sit in the next tier up. This is the cusp: the tier where the line cuts through. Seven of the eight voters' worth of support is needed, so all 8 Green voters share that cost equally. Their gauges drop to about 0.13. Reduced, but not empty — they still have a small amount of influence for the final round.

This is the endorsement at work. Without Green's 8 voters, Working Families' 13 alone would trail Republican's 15. Chen would be selected for this seat instead of Olson. The endorsement visible on the ballot — Green listed alongside Working Families in Olson's row — is what made this seat possible.
Patel's position: Green's own candidate is still running and losing. Green's voters directed their support to Olson instead. They came to the election without the numbers to win a seat on their own — 8 voters against a price of 20 — and chose to spend their influence where it would count.
Ballot weight — after Seat 2
Dem
0.17
Rep
1.00
WF
0.00
Grn
0.13
Seat 3 of 3Chen — Republican
Rep 5★
Dem
Grn
last seat fills itself
Chen Republican
Window 1 — Selection

Republican voters have been at full weight this entire time — they haven't backed any winner yet. Their 15 voters' worth of untouched support is the largest pool of remaining influence. Chen is selected and awarded the final seat.

Window 2 — Accounting

Two seats have already been paid for at 20 each — that is 40 of the original 60 units of weight consumed. Exactly 20 remain in the district: one quota. Whoever is selected for the last seat is sitting on exactly one quota of remaining influence. The last seat fills itself — there is no way to fall short, because the leftover weight is the price.

Ballot weight — after Seat 3 (election complete)
Dem
0.00
Rep
0.00
WF
0.00
Grn
0.00

The Delegation

Seat 1
Martinez
Democratic
Seat 2
Olson
Working Families
+ Green Coalition
Seat 3
Chen
Republican
Patel (Green) — not elected. 8 voters against a price of 20. But Green's endorsement of Olson determined who was selected for Seat 2 — they shaped the delegation without winning a seat of their own.

1 Democratic · 1 Working Families + Green · 1 Republican. The three seats reflect the distribution of voter support across the district. Every gauge started full and ended empty — every voter's influence was spent, and every seat was paid for at the same price.

What the Formula Never Needed to Know

Look back at what just happened and notice what the Allocated Score formula never asked about. It never asked which party a voter belonged to. It never asked about race, language, geography, or income. It never drew a line on a map. It never assigned seats to parties in advance.

The formula did one thing, three times: it counted the support behind each candidate, selected the one with the most, and processed the cost. The groups that formed around each winner were not declared or registered — they emerged from the scores on the ballots. Democratic voters became a "group" because they scored Martinez the same way. Working Families and Green voters became a coalition because they scored Olson the same way. Republican voters became a group because they scored Chen the same way.

The proportionality of the outcome — one seat per twenty voters, three groups each large enough to cover the price — was not determined by a thumb on the scale or a backroom deal. It was a consequence of the arithmetic. Equal price, equal treatment, repeated until every seat is filled. That's how Proportional STAR counts every vote and makes every vote count.

Allocated Score. The mathematical formula that powers Proportional STAR. It selects candidates, charges the quota, makes change, and repeats.
Selection. Each round, the candidate with the most support is selected for the seat. Most support wins.
Price of a seat (quota). Total ballots ÷ seats. Charged after a candidate is selected, never checked before. Every seat costs the same.
Window 1 — Selection. Awards the seat to the candidate with the most support. No measuring, no price — just selection.
Window 2 — Accounting. Processes the charge. Fills the beaker, calculates change, updates the gauges.
Ballot weight (the gauge). Your remaining influence. Starts at 1.00 (full). Drops when your candidate is selected and the cost is processed. At 0.00, you're fully allocated.
Making change. Weight reduction after a seat is won. The group retains surplus influence but at reduced strength — like getting change from a purchase.
Coalition endorsement. When a party endorses another party's candidate. Visible on the ballot before voting begins, so voters can factor it into their scores.