# How a Candidate Gets on the Ballot The least-mapped stage of American elections and the one that eliminates the most candidates. Long before anyone votes, a sequence of filing windows, fees, signature thresholds, verification procedures, challenge proceedings and constitutional tests decides whose name a voter will ever see. This map draws that sequence as a process: **eligibility → which of four tracks you are on → the filing window → the price of entry → the signature test → the removal challenge → judicial review → the disqualification question → the printed ballot.** It is the front end of the election pipeline the vault already maps at the back end. [[How a Midterm Election Is Certified]] begins after the votes are counted; [[State Election Administration Tree]] shows who administers the machinery; this map covers the months before either one starts. It is also the candidate-side companion to [[How Campaign Money Moves]] — money is the second gate, and most candidates who fail never reach it. The single structural fact this map exists to show: **the two directions run on opposite clocks and opposite standards.** Qualifying takes months — a petition drive against a deadline that in most states falls before summer, judged afterward under a test that since [[Burdick v. Takushi]] presumes the state is right. Disqualifying takes days — a challenge filed by a rival party, heard by a state board or a single judge on a calendar compressed against the ballot-printing date, and decided on whether the paperwork was clean. A challenger has to win once, at the end, on a technicality. A candidate has to win at every stage, and win early. Two consequences are visible on the 2026 ballot. In **eleven states and the District of Columbia there is no procedure at all** by which a new party can qualify — the only route is an individual independent petition, filed per candidate, every cycle, with nothing carried forward. And in nine states the most recent occasion on which a minor party or independent candidate *won* a constitutional ballot-access lawsuit is 1992 or earlier; in Mississippi it is 1969. > [!note] How to read this chart > `──▶` = the main path · `└─▶` = a branch or bypass · `◇` = a decision point · `✕` = the candidacy ends · `★` = on the ballot · `⚑` = a load-bearing stage · `[ ]` = a dated note or caveat. Stages run in the order a candidate encounters them, not in order of lethality — the two that end the most candidacies are 4 and 5, in the middle. Everything inside the fence is plain text; linked entities appear in the prose, Notes, and See also. Stage 9 is a **dated state-of-the-ballot block** and is the only part of this map that goes stale within the cycle. ``` HOW A CANDIDATE GETS ON THE BALLOT (as of 15 September 2026) │ Legend: ──▶ main path └─▶ branch/bypass ◇ decision ✕ candidacy ends ★ on the ballot │ ⚑ load-bearing stage [ ] dated note or caveat │ Scale: In EIGHT states every statewide office on the November 2026 ballot carries only a │ Democratic and a Republican nominee — AL, CA, CT, GA, LA, NM, NY, VA. In Illinois │ every legislative race does. Nationally: Libertarian Party on in 32 states, Green │ in 18, Constitution in 12, Forward in 5. │ [ Ballot Access News, 27 Aug and 12-14 Sep 2026. New Mexico turns on a pending │ Tenth Circuit injunction; Utah and Washington have no statewide office up. ] │ ■ 0. ELIGIBILITY (what the office requires before any paperwork exists) │ FEDERAL OFFICE — the Constitution fixes age, citizenship duration and inhabitancy, and │ that list is CLOSED. A state may not add to it: U.S. Term Limits v. Thornton (1995). │ └─▶ so every state restriction below is framed as a BALLOT-ACCESS regulation, never as │ a qualification — which is precisely what puts it under the test at stage 6. │ STATE OFFICE — the state sets the qualifications itself, and they can be steep. │ [ Colo. 2000, Campbell v. Davidson: a state may not require a candidate for Congress │ to be a registered voter. The line between "qualification" and "regulation" is the │ whole fight, and it is litigated one requirement at a time. ] │ ├─◆ 1. WHICH TRACK ⚑ (four different processes; they are not comparable burdens) │ ◇ how is the candidate seeking the ballot? │ │ │ ├─ 1a MAJOR-PARTY PRIMARY ──▶ file a declaration for the party's primary; the party's │ │ nomination is decided by voters. Lowest paperwork burden in almost every state. │ │ └─▶ open / closed / semi-closed primary determines WHO may vote in it, not who │ │ may file. That fight is live: the South Carolina Republican Party sued on │ │ 5 July 2026 to close its own primary (S.C. Rep. Party v. Bolchoz, 3:26cv-2699); │ │ the Texas Republican Party filed the same claim in 2025. │ │ │ ├─ 1b QUALIFIED MINOR PARTY ──▶ the PARTY is already on the ballot from a prior vote │ │ test or registration total; the nominee is chosen by convention or primary. │ │ └─▶ the qualification is perishable — fall below the vote or registration │ │ threshold and the party re-enters at track 1c next cycle. │ │ │ ├─ 1c NEW PARTY ──▶ qualify the PARTY by petition or registration FIRST... │ │ └─▶ ...and then, in some states, EACH NOMINEE must file a separate petition of │ │ his or her own. The two-petition structure is the quiet killer. │ │ [ Live: New Mexico Forward Party v. Toulouse Oliver, 10th Cir. 26-2145, │ │ reply brief 13 Sep 2026. The party completed the party petition, then │ │ could not gather 14,246 MORE signatures for each statewide nominee. ] │ │ ✕ 11 STATES + D.C. HAVE NO NEW-PARTY PROCEDURE AT ALL — CT, DC, IL, IN, IA, KY, │ │ NJ, NY, PA, VA, WA, WV. In those, track 1c does not exist and track 1d is the │ │ only route: a fresh individual petition, every candidate, every cycle. │ │ │ ├─ 1d INDEPENDENT ──▶ one candidate, one petition, no organization to carry forward. │ │ Usually the highest signature number on the page, and in several states higher │ │ than what a whole new party needs. │ │ └─▶ DISAFFILIATION REQUIREMENT: many states bar an independent run by anyone │ │ registered with a party within a set window before filing. Upheld in │ │ Storer v. Brown (1974) — the "sore loser" rule, now law in ~45 states. │ │ │ └─ 1e WRITE-IN ──▶ the residual route, and not universally available. Some states │ require advance declaration; some do not count write-ins at all. │ [ Ind. 1990, Paul v. Election Bd.: a state must print write-in space and count │ what is written there. 2 Jun 2026: Butch Ware, kept off the California │ gubernatorial ballot, took 22,493 write-ins — a record for that primary, and │ more than 49 of the 61 candidates who WERE printed. ] │ ├─◆ 2. THE FILING WINDOW (dates, and the officer who receives the paper) │ A declaration of candidacy, filed with a named officer, inside a window that opens and │ closes on statutory dates. Missing the close is unappealable in most states. │ ◇ is the deadline itself constitutional? │ ├─ EARLY DEADLINES ARE THE MOST-LITIGATED FEATURE IN BALLOT-ACCESS LAW and the one │ │ minor parties most often win. Anderson v. Celebrezze (1983) struck Ohio's March │ │ deadline for an independent presidential candidate. Since then: Feb too early │ │ (HI 2013), Jan too early (KY 2020, UT 2023, CA 2012), Mar too early (SD 2018, │ │ IA 2022, ID 2024), Apr too early (NM 2013), May too early (MA 1985, WY 1997), │ │ Jun too early (NV 1992), Aug too early (AK 1990). │ └─ THE FILING OFFICER is, in most states, a partisan elected official — and in 2026 │ is a named party in several of the challenges at stage 5. ──▶ see the separate │ State Election Administration Tree for who that officer actually is per state. │ ├─◆ 3. THE PRICE OF ENTRY (a fee, a petition, or both) │ ◇ fee or signatures? │ ├─ FILING FEE ──▶ permitted, but NOT as the sole route. │ │ Bullock v. Carter (1972) struck a Texas scheme charging a flat fee plus a pro rata │ │ share of election costs — assessments reached $8,900 — with no alternative. │ │ Lubin v. Panish (1974) held a state may not demand a fee an indigent candidate │ │ cannot pay without providing another way onto the ballot. │ └─ SIGNATURES ──▶ the alternative the two cases require. Which is why almost every │ state that charges a fee also publishes a petition number — and why the petition │ number, not the fee, is where the burden was relocated. │ ├─◆ 4. THE SIGNATURE TEST ⚑ (the stage that ends the most candidacies) │ GROSS vs. NET is the whole discipline: a drive targets 1.5x to 2x the statutory number │ because signatures are struck for address mismatch, non-registration, duplication, │ illegibility, and circulator defects. │ ├─ HOW HIGH MAY IT GO? Jenness v. Fortson (1971) upheld 5% of eligible voters. That │ │ figure became the working ceiling: below it, courts usually sustain; above it, │ │ courts usually strike. Williams v. Rhodes (1968) had already voided Ohio's │ │ effective 15% scheme, holding a state may not shield two parties from competition. │ ├─ DISTRIBUTION REQUIREMENTS — a minimum spread across counties or districts, so that │ │ concentrated urban support cannot qualify a candidate. Repeatedly struck: │ │ Mont. 2021, Pa. 2018, and voided for independents in several more. │ ├─ WHO MAY CIRCULATE — residency bans on out-of-state circulators struck in Conn. 2016, │ │ N.J. 2021, N.Y. 2022; criminal-record bans and per-signature pay bans still │ │ contested. [ Ark., 30 Jun 2026, League of Women Voters v. Jester: a federal court │ │ struck five circulator restrictions at once — forced photo-ID display, a scripted │ │ fraud warning, advance registration of every circulator, a lifetime criminal- │ │ conviction bar, and making sponsors pay the state's legal-notice costs. On appeal; │ │ the out-of-state-circulator and per-signature-pay bans go to trial later. ] │ ├─ WHERE YOU MAY STAND — the ground has narrowed independently of any signature number: │ │ USPS interior sidewalks closed (D.C. Cir. upheld, 2012), post office visits down │ │ >50% since 2000, library visits down 20-50% since the early 2010s, the Virginia DMV │ │ closed its sidewalks in early 2025 and Arkansas followed. │ └─ ⚑ THE THRESHOLD DID NOT MOVE; THE COST OF CLEARING IT DID. │ [ A professional petitioner's good week fell from ~750 gross signatures to ~500 — │ Bill Redpath, Ballot Access News, Aug 2026, from 36 years of running drives. │ Measured in money: the Lincoln Party paid $780,000 for the 36,944 signatures │ that put one Indiana secretary-of-state nominee on the 2026 ballot. ] │ ├─◆ 5. THE CHALLENGE ⚑ (removal, and the only stage with a short clock) │ Filing does not put a name on the ballot; surviving the objection period does. │ ├─ WHO MAY FILE — in most states any registered voter. In practice, a party. │ ├─ THE FORUM — a state or county election board, or a single trial judge, deciding in │ │ days or weeks because the ballot must go to print. │ ├─ THE GROUNDS — signature validity, residency, party-registration duration, form │ │ defects, and the notarization and header rules on the petition itself. │ │ [ Ariz. 2026, Houlihan v. Lytle: unimportant petition errors cannot keep a │ │ candidate off the ballot. Alaska, 26 and 29 Jun 2026, Sullivan v. State Div. of │ │ Elections: the state may NOT remove a candidate for lacking "good faith" — there │ │ being two Republicans named Dan Sullivan on one primary ballot was not a ground, │ │ because no statute made it one. Trial court affirmed by the state Supreme Court. ] │ └─ ⚑ IN 2026 THE CHALLENGE CHANGED HANDS. For ninety years the Democratic Party filed │ most removal challenges against minor-party and independent candidates. This year │ the Republican Party filed more, for the first time on record: │ Republican (9) ..... AZ (party name change) · FL x2 (U.S. House; city council) │ IA x3 (two U.S. House; governor ticket) · KS (Sec. of │ State struck a party) · NH (Sec. of State v. an │ independent Senate candidate) · NJ (U.S. House) │ Democratic (2) ..... CA (Sec. of State struck the Green gubernatorial │ candidate) · IL (two independents, 4th district — the │ only ones of the eleven that clearly succeeded) │ [ Ballot Access News, 27 Aug 2026. Most remain pending. Before 1952 there is no │ known instance of a Republican challenge to a minor-party or independent │ PRESIDENTIAL candidate; in 2024 there were none. ] │ ├─◆ 6. JUDICIAL REVIEW (the Anderson-Burdick balancing test) │ A candidate removed at stage 5, or barred at stage 4, sues. What the court applies is │ not tiered scrutiny but a sliding scale: │ ├─ Anderson v. Celebrezze (1983) ──▶ weigh the character and magnitude of the injury │ │ to First and Fourteenth Amendment rights against the precise interests the state │ │ asserts, and ask whether those interests make the burden necessary. │ ├─ Burdick v. Takushi (1992) ──▶ converts it to a scale. A SEVERE burden must be │ │ narrowly tailored to a compelling interest. A "reasonable, nondiscriminatory │ │ restriction" is sustained by the state's important regulatory interests alone. │ │ ◇ which branch? ── the case is usually decided the moment that word is chosen. │ ├─ Munro v. Socialist Workers Party (1986) ──▶ the state need not produce EVIDENCE that │ │ the harm it is regulating against would occur. Legislative judgment suffices. │ └─ Timmons v. Twin Cities Area New Party (1997) ──▶ a state may ban fusion, and its │ interest in "the stability of the political system" — read as the two-party │ system — is a legitimate one to weigh against the burden. │ ✕ NET EFFECT: a framework that is formally balancing and functionally deferential. │ [ In 9 states the most recent constitutional ballot-access WIN by a minor party or │ independent is 1992 or earlier: Miss. 1969 · La. 1979 · N.D. 1980 · Okla. 1984 · │ Mass. 1985 · Alas. 1990 · Ind. 1990 · Fla. 1992 · Nev. 1992. Compiled by Richard │ Winger; the 2026 entries are Ariz. (Houlihan) and Wis. (Strange v. Elections Comm). ] │ ├─◆ 7. THE DISQUALIFICATION QUESTION (14th Amendment, Section 3) │ The one exclusion ground that is federal, constitutional, and currently inoperative. │ ├─ Trump v. Anderson, 4 Mar 2024 ──▶ all nine Justices agreed that responsibility for │ │ enforcing Section 3 against federal officeholders and candidates rests with │ │ CONGRESS, not the states. A state may not strike a federal candidate under it. │ ├─ STATE OFFICES ──▶ states retain concurrent authority to enforce Section 3 there. │ └─ ◇ has Congress legislated the procedure Section 5 requires? ── NO. │ ✕ so the clause is valid, the enforcer is identified, and the mechanism does not │ exist. The gate has an operator who has never been given a key. │ ├─◆ 8. THE PRINTED BALLOT (what survives onto the page still varies) │ Ballot ORDER (rotation, incumbent-first, or party-in-power-first) · the PARTY LABEL and │ whether a candidate may claim one · INCUMBENCY and OCCUPATIONAL designations · whether │ write-in space is printed at all · and the ballot-printing date, which is the real │ deadline every clock above is measured against. │ └─◆ 9. STATE OF THE 2026 BALLOT [ dated block — 15 September 2026 ] Statewide offices with ONLY D and R nominees ......... 8 states (AL CA CT GA LA NM NY VA) Legislative races with ONLY D and R nominees ......... every one, in Illinois New-party qualification procedure does not exist ..... 11 states + D.C. Libertarian Party ballot status ..................... on in 32 states Green Party ......................................... on in 18 states Constitution Party .................................. on in 12 states Forward Party ....................................... on in 5 states Party challenges to minor/independent candidates .... 9 Republican, 2 Democratic Pending at the Supreme Court ........................ RNC v. Eakin, No. 25-962 — documentary proof of citizenship to register; set for argument in OT 2026-27. It is a REGISTRATION case, not a ballot-access case, but it is the gate one step upstream of every signature on every petition above. [ Party ballot-status counts and the challenge tally: Ballot Access News, 27 Aug 2026. Single-party-monopoly counts: Ballot Access News, 13-14 Sep 2026. ] ``` ## Notes - **Why the two clocks matter more than any single threshold.** No stage on this map is unusual on its own; a signature requirement, a filing deadline and an objection period all have obvious administrative justifications. The asymmetry is what they add up to. A petition drive is a months-long capital expenditure with no refund, and it is reviewed *afterward* under a standard that in most cases asks only whether the state had an important reason. A challenge is filed in an afternoon, costs the challenger nothing but a lawyer, and is decided before the printer's deadline on whether the header on page 40 was formatted correctly. The 2026 record at stage 5 shows both parties using the instrument; the point of the map is that the instrument exists on only one side of the process. - **The 2026 party flip is a fact about tactics, not about doctrine.** [[Richard Winger]] has tracked minor-party and independent challenges since 1985, and his August 2026 count — nine Republican, two Democratic — is the first year on record in which Republican challenges outnumbered Democratic ones. Most of the nine are still pending as of this writing and may yet fail. What the flip indicates is a changed calculation about which third-party candidacies draw from which major party, not any change in what the law permits. Both counts are Winger's; no official body publishes a tally. - **"No procedure" means exactly that.** The eleven states and D.C. at stage 1c have not set a high bar for new-party qualification — they have not written one. A group wishing to appear on the ballot in Illinois or New York files an independent petition for each individual candidate, each cycle, and nothing about clearing it in one cycle reduces the burden in the next. This is a structurally different regime from a state with a difficult petition, and it is frequently reported as though it were the same thing. - **What this map deliberately leaves open.** (1) Whether the collapse in signature-gathering productivity described by Bill Redpath is measurable in any systematic dataset — it is drawn from one experienced practitioner's estimate across 36 years, reported in a newsletter, and no agency collects the figure. It is carried here with his name attached and no false precision. (2) Whether the nine pending 2026 Republican challenges succeed at a rate comparable to the historical Democratic ones; there is no baseline. (3) Whether a state may bar an independent candidacy under [[Storer v. Brown]] when the candidate's party disaffiliated *from them* — an inverted sore-loser question that no reported decision squarely reaches. - **A note on Section 3.** Stage 7 is drawn as a live stage because it is invoked in filings every cycle, not because it currently operates. [[Trump v. Anderson]] resolved the forum question unanimously and left the enforcement question untouched: Section 5 gives Congress the power to legislate the procedure, and Congress has not. Reporting that treats the decision as having *narrowed* Section 3 misreads it — the Court relocated the power rather than shrinking it, and the relocation is what makes it inert. The related [[Fourteenth Amendment]] and [[January 6 Capitol Attack]] pages carry the substantive history; this map carries only the procedural position. - **Officeholder verification.** This map names no current officeholder and therefore required none. Every named case, date and count was verified on 15 September 2026 against the source identified in the bracket beside it. The Supreme Court entry at stage 9 (No. 25-962) is a docket number, verified as granted for the October Term 2026 argument calendar; nothing about its outcome is asserted. ## See also - [[How a Midterm Election Is Certified]] — the same pipeline, after the votes - [[State Ballot Position Crosswalk]] — the next question after who is printed: in what order, who decides it, and what the position is measurably worth - [[State Election Administration Tree]] — who administers the filing window and the challenge - [[How Campaign Money Moves]] — the second gate, reached only by candidates who clear this one - [[Ballot Access]] — the concept, and the vault's entry point for the underlying doctrine - [[Anderson v. Celebrezze]] · [[Burdick v. Takushi]] — the balancing framework at stage 6 - [[Williams v. Rhodes]] · [[Jenness v. Fortson]] · [[Storer v. Brown]] — the threshold cases at stages 1 and 4 - [[Bullock v. Carter]] · [[Lubin v. Panish]] — the filing-fee cases at stage 3 - [[Trump v. Anderson]] — the Section 3 forum question at stage 7 - [[Richard Winger]] — the source stream for the 2026 counts on this map - [[Elections Clause]] · [[Fourteenth Amendment]] — the constitutional footing - [[Libertarian Party]] · [[Green Party]] · [[Constitution Party]] — the parties counted at stage 9 - [[Foundations Hub]] · [[2026 Midterms Hub]]