Can past elections improve on August’s shares? Washington’s 31st District, 2026
Can past elections improve on using August’s vote shares?
The tested formula did not improve the estimate.
The test followed Republican candidates in 302 past races against Democrats. Carrying each candidate’s August vote share forward unchanged produced an error score of 2.15 percentage points. A straight-line model fitted to other election years scored 2.17. The score gives larger misses more weight; lower is better.
Read the full comparison, charts and calculation record
The first article shows why choosing the candidate from November’s result changed the apparent performance of the formula.
August 2026 vote counts
| Race | Candidate | Party preference | Votes | Two-candidate total | Share |
|---|---|---|---|---|---|
| State Senator | Tamara Stramel (D) | Prefers Democratic Party | 15,638 | 34,221 | 45.70% |
| State Senator | Phil Fortunato (R) | Prefers Republican Party | 18,583 | 34,221 | 54.30% |
| State Representative Pos. 1 | Drew Stokesbary (R) | Prefers Republican Party | 19,674 | 33,973 | 57.91% |
| State Representative Pos. 1 | Stephen Szczurko-Walton (D) | Prefers Democratic Party | 14,299 | 33,973 | 42.09% |
| State Representative Pos. 2 | John Bielka (D) | Prefers Democrat Party | 14,915 | 33,954 | 43.93% |
| State Representative Pos. 2 | Joshua Penner (R) | Prefers Republican Party | 19,039 | 33,954 | 56.07% |
Vote counts (CSV) · Vote-count and historical-comparison code
Data and code
Official source files and prediction-test code (ZIP) · Calculation and record checks
Sources and production
Silphium is Penner Strategy’s research system. These studies use AI-assisted analysis and code-generated calculations and charts. The linked review records describe the calculation checks and remaining limitations.