Idaho

The certified canvass reports Registration at Cutoff, Election Day Registrants and Total Registered separately for all 44 counties, and they reconcile exactly — 1,024,753 plus 11,087 equals 1,035,840. Elsewhere in this corpus the equivalent gap has to be reconstructed, guessed at, or is simply unavailable. Here the risk inverts: a consumer that reads only Total Registered silently absorbs 11,087 voters who were not on any pre-election roll, and one that reads only the cutoff figure misses them entirely. Two defensible turnout rates exist and the state publishes the inputs for both.

Statewide, election-day registration is 1.07% of the roll. Minidoka County is 18.66% — 1,952 registrations against a total of 10,459 — and Custer is 8.67%. The next county down is Twin Falls at 2.26%, and the floor is Benewah at 0.36%. So a statewide same-day rate describes almost no individual county, and any per-county model calibrated on the average is off by an order of magnitude in the two places it matters. This page reports the distribution and does not offer a cause for it.

Each of Idaho's 35 legislative districts elects a senator plus two representatives distinguished only as Seat A and Seat B, giving 105 seats that all run every two years. So a district identifier does not identify a seat, a chamber does not have halves to stagger, and Seat A and Seat B are not geographic subdivisions — they are labels on two seats with identical constituencies. Anything keying a legislative office on district number alone collides two representatives, and anything expecting staggered senate terms finds none.

The certified canvass is organised as office-by-party — United States Senator appears three times, once each for the Democratic, Libertarian and Republican primaries, and Governor likewise. Across 115 offices there were 216 contests: 112 Republican, 100 Democratic and 4 Libertarian. So contest count is not office count, an office may have anywhere from one to three primaries, and a minor party can contest some offices and not others. Counting rows in the canvass overstates the ballot by nearly double.

Idaho publishes results through 44 separate per-county instances of the same election-night product, addressed individually — ada-county-id, adams-county-id, and so on through the alphabet. There is no single statewide results endpoint; a statewide picture requires 44 requests against 44 jurisdictions, each with its own identifiers and its own official flag. The certified statewide record lives elsewhere again, in a canvass PDF. The same vendor serves Virginia as one statewide instance, so the product shape says nothing about the state's shape.

The 2026 audit records that Ada County — the state's largest, at 301,671 registrations — does not sort its absentee or early ballots by precinct but by batches, so precinct-level voter history could not be audited there beyond the election-day distinction. Elsewhere the audit lists precincts and batches interchangeably in one line: 1610, 1801, 1802, then EV 7 and similar. A precinct in Idaho is a reporting unit for some ballot types and not others, and which depends on the county.

The post-election audit narrative states that Idaho voters cast 312,140 votes in the 2026 primary; the certified canvass county table totals 312,139 ballots cast. The words differ — votes against ballots cast — so the two may be measuring different things, and the audit itself documents a single-ballot discrepancy in Ada precinct 1610 where a voter was issued a nonpartisan ballot and left without casting it. A pipeline that treats any official figure as authoritative without recording which document it came from cannot reconcile a one-unit difference, or explain it.

Idaho elects a governor, a lieutenant governor, a secretary of state, a state controller, a state treasurer, an attorney general and a superintendent of public instruction — each on its own ballot line, each with its own primaries. The lieutenant governor is nominated independently rather than running as a governor's pick, which is the opposite of the arrangement in several neighbours, and Wyoming next door has no lieutenant governor at all. The state controller has no counterpart in most states.