New Mexico
New Mexico opened its primary to voters in no party, counted 390,218 of them as eligible, and published the number of ballots they cast on that line: zero. The turnout rate divides by all of them anyway.
What does the NON ballot style mean?
State deltas
Every state has hair. Here is New Mexico's.
The rules that make this state different from the last one, what each breaks if you assume the national default, and how the engine resolves it.
The certified primary's own ballot-style table reports 390,218 voters eligible under the non-affiliated style and 0 ballots cast on it, while the published turnout rate of 24.60% divides 346,460 ballots by all 1,408,181 eligible. Against the two major-party rolls alone the figure is 34.04%. Both numbers come from the same table, and a consumer that reads the headline rate inherits a denominator more than a quarter of which produced no ballot of the style attributed to it.
Turnout carries its denominator's composition, and eligibility by ballot style is stored separately from ballots cast so a zero row cannot be silently absorbed into a rate.
The state describes itself in its own words as a semi-open primary state where major-party registrants and unaffiliated voters may both participate. That is neither the closed model, where party registration gates the ballot, nor the open model, where anyone takes any ballot. Anything storing primary type as a boolean, or inferring participation rules from a party-registration field, cannot represent it — and the published ballot-style table shows the arrangement produces a category with eligibility and no recorded ballots.
Primary access is stored as a named regime with its own participation rule rather than as open or closed, and ballot style is kept distinct from voter registration.
Published county turnout for the same primary runs from Harding at 50.00% to Lea at 14.60%, with the statewide figure at 24.60% sitting closer to the bottom than the middle. Nineteen counties are above the statewide rate and fourteen below, so the statewide number is not a typical county and no county is a proxy for the state. Applying a statewide participation assumption per county is wrong by up to twenty-five points in either direction.
Turnout is a per-county measure with its own distribution, never applied as a statewide constant, and the statewide figure is stored as an aggregate rather than a representative value.
The 2026 general candidate list carries 77 magistrate candidates and 40 probate candidates, alongside district judges and district attorneys, and the results portal keeps Metropolitan Court Judges as a separate category for Bernalillo County alone. So the judicial tier is not one shape but four, three of them county-scoped and one of them unique to a single county. A model with a single judicial office type, or one that expects judges at state level only, cannot hold this.
Judicial offices carry a court type and a scope, and a court that exists in only one county is representable without inventing it elsewhere.
Of 1,003 local governments the Census counts in New Mexico, 769 are special districts — 77%, against 105 municipalities and 96 school districts. That share is second only to Colorado's 86% in this family, and the composition is distinctive: New Mexico's political subdivisions include acequias, the community irrigation associations that elect their own commissioners and predate the state. Modelling the local layer as municipalities captures roughly a tenth of the state's governments.
Special districts are a first-class tier with a function code, and the local layer is never assumed to be municipal.
New Mexico elects a governor, lieutenant governor, secretary of state, attorney general, treasurer, auditor and a Commissioner of Public Lands — an office with no counterpart in most states, administering nine million acres of state trust land. A separately elected Public Education Commission sits alongside them. Mapping this onto a standard statewide slate loses two offices and misfiles a third.
The statewide executive set is enumerated per state from its own candidate list rather than mapped onto a standard slate.
New Mexico's results portal runs the same ASP.NET product as South Dakota's and as the North Dakota system that carried elections through 2024 before that state migrated away mid-cycle. The contest-type codes are shared but the categories are not: New Mexico exposes magistrate, metropolitan court and Public Education Commission views that the others have no use for. A harvester written against one deployment's category list will silently miss whole office classes in another.
Contest categories are discovered per deployment rather than assumed from the platform, and the platform is recorded as a fact about the feed, not the state.
New Mexico administers 33 counties and the Census counts 33 county governments. Across the Mountain West the discrepancy appears only where consolidated city-counties exist — Colorado 64 against 62, Montana 56 against 54, Nevada 17 against 16 — and nowhere else: Wyoming, Idaho, Utah, Arizona and New Mexico all agree exactly. A reconciliation against Census unit counts is therefore safe here and unsafe in three neighbours, for a reason that is knowable in advance.
County identity comes from the state's own administration list in every state, with the Census classification stored alongside and the consolidated-city-county count recorded as the expected difference.
Federal to acequia
What New Mexico actually elects.
Seven statewide executives, a Public Education Commission, and a courthouse that puts magistrate and probate judges on the ballot in every county.
Congressional map
Where those seats are.
The 3 congressional districts in force for the 2026 cycle — not necessarily the map this delegation was elected under. New Mexico has not redrawn since the 119th Congress was seated.
| District | Population | Median income |
|---|---|---|
| NM-01 | 705,390 | $68,255 |
| NM-02 | 710,078 | $56,082 |
| NM-03 | 698,903 | $62,270 |
County ledger
One statewide feed, thirty-three counties, and a rate that describes none of them.
Turnout below is the state's own published figure per county for the certified June 2 primary, read directly from the results portal — 2,204 precincts, all fully reported. The second column is that county's distance from the statewide 24.60%, which is arithmetic on the first. Nineteen counties sit above the statewide rate and fourteen below, and the spread runs thirty-five points. Thirteen of 33 shown.
| County | Feed vehicle | Primary turnout | vs statewide | Status | Last read |
|---|---|---|---|---|---|
| Harding · highest | County HTML | 50.00% | +25.40 | Verified | 1 hr |
| Mora | County HTML | 45.75% | +21.15 | Verified | 1 hr |
| Guadalupe | County HTML | 42.91% | +18.31 | Verified | 1 hr |
| San Miguel | County HTML | 36.78% | +12.18 | Verified | 1 hr |
| Rio Arriba | County HTML | 34.31% | +9.71 | Verified | 1 hr |
| Santa Fe | County HTML | 32.48% | +7.88 | Verified | 1 hr |
| Taos | County HTML | 29.43% | +4.83 | Verified | 1 hr |
| Los Alamos | County HTML | 26.75% | +2.15 | Verified | 1 hr |
| Bernalillo · Albuquerque | County HTML | 26.24% | +1.64 | Verified | 1 hr |
| Sandoval | County HTML | 25.21% | +0.61 | Verified | 1 hr |
| San Juan | County HTML | 21.55% | -3.05 | Verified | 1 hr |
| Dona Ana · Las Cruces | County HTML | 17.03% | -7.57 | Verified | 1 hr |
| Lea · lowest | County HTML | 14.60% | -10.00 | Verified | 1 hr |