PARTICIPANTS IN THIS TEST
—in the November general election
TURNOUT IN THIS TEST
—of current registered voters
REGISTERED VOTERS
—active and inactive registrations
The historical records are real; the 2026 figures are hypothetical model outputs. The test estimates election participation, which can differ from votes cast in the Assessor contest. Registered party does not establish candidate preference.
THE HISTORICAL ASSUMPTION
2018 is the main comparison
The weights are a test assumption, not a statistically selected optimum. Historical rates are applied to today’s registration for each party. This blend has not been independently validated.
Compare alternative weights
| 2018 / 2022 | Participants | Turnout |
|---|
These are sensitivity scenarios, not the limits of a probability interval.
01 / THE ELECTORATE
Expected turnout by registered party
Estimated composition of the participating electorate. Party colors describe registration only.
EXPLORE AN ASSUMPTION
What if overall turnout changes?
Move the county total to see how the party estimates change while keeping the same historical participation pattern. This is a scenario, not a probability interval.
WHAT THIS TEST EVALUATES
Historical checks, explicit assumptions
Retrospective checks compare estimated party participation with recorded historical participation. The 2026 scenario explores a 70% weight on 2018 and 30% on 2022. These checks do not establish the accuracy of that future scenario. This study does not estimate candidate vote shares or chances of winning.
02 / THE HISTORICAL RECORD
Participation since 2018
| Election | Republican | Democratic | Independent | Libertarian | Participants |
|---|
2018 receives the larger weight in this test, with 2022 supplying the remaining weight. The 2020 and 2024 presidential elections provide context and a separate historical check; they receive no weight in this blend.
03 / WHAT HAS BEEN CHECKED
How well did the party split hold up?
These checks estimate each party’s participation in a later election using the previous election of the same type. The later election’s actual county total and registration are supplied to the check. This tests how turnout is divided among parties; it does not validate the new 2018/2022 blend or a complete advance forecast.
See each party’s error and the comparison baseline
| Check | Republican | Democratic | Independent | Libertarian |
|---|
The small Libertarian group had larger errors: about 15.4 points in 2022 and 6.8 points in 2024. Treat its estimate with extra caution. These two retrospective checks are limited evidence and are not an untouched validation set.
04 / WITHIN THE COUNTY
Precinct turnout under this test
All-party scenarios start with the statewide model’s precinct pattern and adjust to this test’s county total. They are not separate 2018 precinct fits. Party estimates above are county totals and are not assigned to individual precincts.
| Precinct | Registered | Expected participants | Expected turnout |
|---|
05 / SOURCES & ASSUMPTIONS
A transparent starting point
No calibrated uncertainty interval is available for the party estimates. No polling, absentee-request trends, or voter-removal model is included.