Full rankings · Report
Midterm Seat Loss Ranking
Ranks every US president since 1926 by how many US House and Senate seats their party lost or gained at the midterm — the ballot-box verdict on their first two years. House and Senate shown as separate rankings.
Last updated November 04, 2026
A president’s first-two-year report card — the midterm
Every president’s party gets a ballot-box verdict two years into the term. This page ranks all 25 midterms since 1926 by how many US House and Senate seats the incumbent party lost or gained. The sign convention: higher is better (negative = seats lost; positive = seats gained).
As of 7:00 PM, November 4, 2026 EST. Updates nightly via Wikipedia scrape (temporary, until final certification).
Trump II’s row enters the ranked tables below once midterm_baseline.py picks up the final certified 2026 seat counts (typically 2–3 weeks post-election). The preliminary numbers above are the live indication.
US House seat swing — ranked
All 25 midterms sorted by House seats gained (positive) or lost (negative) by the sitting president’s party. Rank 1 = best outcome.
US Senate seat swing — ranked
Same 25 midterms sorted by Senate seats. The Senate picture often diverges sharply from the House — senators serve 6-year terms, so only one-third of the chamber is up in any cycle, and gerrymandered state maps don’t apply.
Both chambers combined — composite rank
Mean of each term’s House rank and Senate rank. A balanced view for readers who want one number — but the chamber-specific rankings above are the more honest reads.
Breakouts by setup
Different starting conditions produce very different midterm outcomes. Three cuts of the same data:
First-term midterms only
Every president’s first midterm, when the “referendum on the new administration” effect is strongest.
Unified government going in
Same party controls House + Senate + White House. Historically the worst outcomes — voters tend to split power.
Divided government going in
Congress or one chamber already in the opposition’s hands — less room to swing.
Methodology
What we rank. House and Senate seat swing — the net change in the incumbent party’s seat count between the pre-election Congress and the next Congress. Positive = seats gained (rare); negative = seats lost (the norm).
What we don’t include. Governorships, state legislatures, ballot initiatives, special elections. Those move markets less and have noisier historical data.
Current midterm handling. Until midterm_baseline.py picks up the final 2026 seat count, Trump II’s row is marked “Results Pending”. Once results post, the row drops into the ranking automatically.
Data source. midterm_baseline.json → historical_midterms, regenerated nightly by midterm_baseline.py from the presidential_performance database. Upstream: Clerk of the House + US Senate historical records.
Related reports
- 2026 Midterm Forecast — the pre-election model that predicts where Trump II’s row above will land once results post.
- Full-Term Ranking — how presidents rank on 24 economic + market metrics over their full term.
- Same Days in Office — apples-to-apples ranking capped at the current president’s day count.
- Back to Presidential Performance library