Scan: Will Republicans win the Senate race in New Hampshire?

← Back to scans

Will Republicans win the Senate race in New Hampshire? — Republican party

SENATENH-26-R · Elections · 2026-04-16

17%

Agent

13%

Market Price

+4.0%

Edge

45%

Confidence

Volume: 52,780

Spread: 3.0c

Days to resolution: 566

Markets in event: 2

Final Rationale

My independent estimate of 35% was significantly inflated by double-counting the open seat factor, underweighting midterm headwinds against the president's party, and over-crediting John E. Sununu's 17-year comeback attempt. After accepting the Devil's Advocate critiques, my adjusted estimate fell to roughly 22-25%. The market at 13% with very high depth (297,601 contracts) carries strong informational weight, reflecting NH's consistent federal Democratic lean, likely midterm headwinds against Republicans, and possibly emerging information about Trump's NH toxicity or strong Democratic candidates. I settle at 17% — slightly above the market because the 19-month time horizon and open-seat dynamics preserve genuine uncertainty, but much closer to the market than my original estimate given the well-identified reasoning flaws.

Price vs. prediction

How the market moved after the agent's call. Solid blue line is the actual market price; dashed line is the agent's static forecast at $t{=}0$.

$n = 58$ follow-ups

1. Planner Opus · Fermi decomposition + tool plan

Tools chosen: kalshi_data kalshi_orderbook article_search web_search wikipedia_lookup

Sub-questions (Fermi decomposition)

Will the Republican candidate win the 2026 New Hampshire Senate race based on current polling and fundamentals? (w=0.40)
Is the national political environment (approval ratings, economic conditions) favorable enough for Republicans to win a competitive Senate seat in a purple state like New Hampshire in 2026? (w=0.30)
Will the Republican candidate be significantly stronger than the Democratic incumbent (Jeanne Shaheen's seat) in terms of candidate quality, fundraising, and name recognition? (w=0.20)
Has a strong Republican candidate already announced or emerged for the New Hampshire Senate race? (w=0.10)

Planner reasoning

This question concerns the 2026 New Hampshire Senate race, which is a competitive midterm cycle contest. Key considerations: (1) New Hampshire is a purple/swing state that has elected both Republicans and Democrats to statewide office; (2) The seat in question is Jeanne Shaheen's seat — she has announced retirement, making it an open seat, which substantially changes the competitiveness calculus; (3) Open seats in swing states are highly competitive, giving Republicans a genuine shot; (4) The 2026 midterm environment will matter — as a midterm under a Republican president (Trump), historical patterns suggest the president's party typically loses seats, which would favor Democrats; (5) Candidate quality will be critical — Republicans need a credible, well-funded candidate to compete in a blue-leaning state. The temporal sequence: candidate announcements are likely already happening (April 2026), primary battles will occur summer 2026, general election in November 2026, with resolution in January 2027. The reference class of open Senate seats in purple states under a same-party midterm environment historically shows competitive races but slight Democratic lean given NH's recent presidential voting patterns. The key variable is whether Shaheen's retirement creates enough of an opening to overcome the structural midterm disadvantage Republicans may face.

2. Research Parallel tool calls · legacy v2

Tool	Status	Time	Items
kalshi_data	OK	2.1s	-
kalshi_orderbook	OK	0.1s	-
article_search	OK	0.2s	-
web_search	OK	48.7s	-
wikipedia_lookup	OK	2.2s	-

3. Evidence Synthesis Sonnet · 15 items · legacy v2

#	Strength	Credibility	Direction	Source	Claim	Priced In
1	STRONG	78	↓ DOWN	kalshi_data	Kalshi prediction market prices the Republican winning the NH Senate race at 13%, down 2% in the last 7 days and down 1% over 30 days, suggesting the market strongly favors the Democrat.	Yes
2	STRONG	75	↓ DOWN	kalshi_orderbook	The orderbook shows HIGH liquidity (depth=297601) with no_bid at $0.44, indicating strong market confidence that the Republican will NOT win.	Yes
3	MODERATE	82	↓ DOWN	wikipedia_lookup	New Hampshire has voted Democratic in the last six presidential elections since 2004, though by relatively small margins, indicating a consistent but modest blue lean at the federal level.	Yes
4	STRONG	85	↓ DOWN	web_search	Republicans have not won a U.S. Senate race in New Hampshire since 2010, and the entire congressional delegation has been Democratic since 2017, showing structural Democratic advantage at the federal level.	Yes
5	MODERATE	72	↓ DOWN	web_search	2026 is a midterm election under Republican President Trump; historical patterns show the president's party typically loses Senate seats in midterms, which would disadvantage Republicans in competitive states like NH.	Yes
6	MODERATE	85	↓ DOWN	web_search	Kamala Harris carried New Hampshire by roughly 3 percentage points in 2024, confirming NH's modest but consistent blue lean at the federal level entering the 2026 cycle.	Yes
7	MODERATE	70	↓ DOWN	kalshi_data	The NH Governor race (2028) is priced at 77% Republican, contrasting sharply with the 13% Republican Senate probability — indicating the national/federal environment is seen as particularly unfavorable for Republicans in NH Senate races.	Yes
8	MODERATE	80	NEUTRAL	wikipedia_lookup	NH Republicans control the governorship, both state legislative chambers, and a majority in the executive council, showing the party has state-level viability but historically struggles at the federal/Senate level.	Yes
9	STRONG	88	↑ UP	web_search	Shaheen's seat is now an open seat (she announced retirement), removing the incumbency advantage for Democrats and significantly increasing competitiveness relative to a challenger-vs-incumbent race.	Yes
10	MODERATE	82	↑ UP	web_search	This will be the first open Senate election in NH since 2010 and the first for this seat since 1990, making historical comparisons difficult and the race potentially more competitive than recent cycles suggest.	Yes
11	WEAK	40	NEUTRAL	web_search	No specific polling data was found comparing Republican vs. Democratic candidates for the 2026 NH Senate race, leaving candidate quality and fundraising comparisons largely unquantified.	No
12	STRONG	87	↑ UP	web_search	Former NH Senator John E. Sununu (served 2003-2009) has announced a comeback bid for the seat, representing a credentialed candidate with prior Senate experience in this exact seat.	Yes
13	MODERATE	85	NEUTRAL	web_search	Scott Brown, a former U.S. Senator (Massachusetts) who lost to Shaheen in 2014 and served as Trump's ambassador, is also running in the Republican primary, providing a second credible but previously defeated option.	Yes
14	STRONG	87	↓ DOWN	web_search	Former NH Governor Chris Sununu (widely considered the strongest potential Republican candidate) has declined to run, removing what could have been the most competitive Republican option.	Yes
15	MODERATE	80	↑ UP	web_search	Dan Innis (state senator) dropped out and endorsed John E. Sununu, consolidating some Republican support behind Sununu ahead of the September 8, 2026 primary.	Yes

Information Gaps

No head-to-head polling data between Republican candidates (Sununu, Brown) and likely Democratic nominees was found — this is the most critical missing data point.
No information on who the Democratic candidate(s) are or their strength, fundraising, and name recognition.
No Cook Political Report or Sabato's Crystal Ball rating for this race was surfaced, which would provide expert handicapping.
No data on Republican vs. Democratic fundraising totals for this race.
No current data on Trump approval ratings in New Hampshire specifically, which would inform the midterm environment impact.
No information on the broader 2026 national political environment (economic indicators, generic congressional ballot) that would help assess midterm headwinds for Republicans.
Unclear whether John E. Sununu being out of politics since 2009 affects his current viability and whether he has maintained a political network.

Key Uncertainties

Who the Democratic nominee will be and their relative strength compared to John E. Sununu or Scott Brown.
Whether the midterm environment punishes Republicans heavily (as it typically does for the president's party), or whether unique conditions in 2026 moderate this effect.
Whether John E. Sununu's long absence from NH politics (since 2009) diminishes his competitiveness, or whether his name recognition translates to electoral viability.
Whether the open-seat dynamic is enough to overcome NH's structural federal-level lean toward Democrats and the typical midterm disadvantage for the president's party.
The degree to which the national Trump political environment specifically hurts or helps Republican Senate candidates in purple states like NH.
Whether any stronger Republican candidate could still enter the race before the September 2026 primary filing deadline.

Market Context

Price: 0.13 · 7d: -0.02 · 30d: -0.01
Volume: MODERATE — 578 average daily contracts, reasonable but not high-volume · Depth: HIGH — depth of 297,601 contracts, very deep book
Reliability: MEDIUM-HIGH — high liquidity and depth suggest informed market, but price range of 13-24% over 62 days shows significant uncertainty earlier that has now compressed toward the lower end, possibly reflecting worsening Republican prospects

4. Independent Forecaster Opus · No market price · legacy v2 Did NOT see market price

Independent estimate: 35% (40% confidence)

base rate

35%

35% conf

evidence driven

34%

42% conf

contrarian

35%

42% conf

Spread

1pp

Low disagreement

sq1: Will the Republican candidate win the 2026 New Hampshire Senate race based on current polling and fundamentals?

33%

Base rate: NH has not elected a Republican senator since 2010 (Kelly Ayotte). In the last 6 Senate races (2010-2022), Republicans won 1 out of 6. That's ~17% base rate. However, this is an open seat, which historically increases competitiveness significantly — open seats swing more than incumbent-held seats. Open seat adjustment pushes up by ~10-12%. NH's federal lean is modestly Democratic (Harris +3 in 2024), but it's genuinely competitive territory. The midterm headwind for the president's party pushes down by ~5% — historically the president's party loses ~2-3 Senate seats in midterms, and competitive seats are where losses concentrate. Net assessment: ~17% base + 12% open seat - 5% midterm headwind + 5% for general competitiveness of NH = ~33%.

sq2: Is the national political environment favorable enough for Republicans to win a competitive Senate seat in a purple state like NH in 2026?

30%

Base rate: In midterm elections, the president's party typically faces headwinds. Since 1934, the president's party has lost an average of ~4 Senate seats in midterms. In competitive/purple states, this effect is amplified. Looking at recent midterms under first-term presidents: 2018 (Trump) saw Democrats gain 2 Senate seats despite a tough map; 2010 (Obama) saw Republicans gain 6. The 2026 map is more favorable for Democrats than 2024. Trump's approval has historically been polarizing in NH. The national environment being unfavorable for Republicans in a purple state pushes this down. Without specific Trump NH approval data, I estimate ~30% chance the environment is favorable enough for a Republican win in NH specifically.

sq3: Will the Republican candidate be significantly stronger than the Democratic candidate in terms of candidate quality, fundraising, and name recognition?

40%

Base rate for open seats: candidate quality is roughly balanced in open-seat races, so ~50% baseline. John E. Sununu is a credible candidate with prior Senate experience, which pushes up slightly. However, Chris Sununu (the stronger candidate) declined to run, pushing down by ~5%. John E. Sununu has been out of politics since 2009 — 17 years is a very long absence, pushing down by ~5%. We don't know the Democratic nominee, which is a major uncertainty. Democrats will likely recruit a strong candidate for an open seat in a winnable state. Scott Brown as an alternative has already lost statewide in NH (2014), making him weaker. Net: 50% - 5% (no Chris Sununu) - 5% (long absence) = ~40%.

sq4: Has a strong Republican candidate already announced or emerged for the NH Senate race?

60%

John E. Sununu has announced and is a former senator — this is a credible, experienced candidate. He won this exact seat in 2002. That's a strong signal, pushing up significantly from a neutral 50%. However, Chris Sununu declining removes the strongest potential candidate (pushes down ~10%). John E. Sununu's 17-year absence from politics is a meaningful concern (pushes down ~5%). Dan Innis's endorsement of Sununu and consolidation of support pushes up ~5%. A contested primary with Scott Brown could weaken the eventual nominee (pushes down ~5%). Net: 50% + 15% (former senator) - 10% (no Chris) - 5% (long absence) + 5% (consolidation) - 5% (contested primary) = ~55-60%.

Show full reasoning trace

base rate: Republicans have won 1 of 6 NH Senate races since 2010 (~17%). Open Senate seats in purple states are more competitive, with the out-party winning roughly 50-55% of the time historically. But NH's federal lean is Democratic, and 2026 is a midterm under a Republican president, creating headwinds. The reference class of open Senate seats in blue-leaning states during midterms unfavorable to Republicans suggests ~30-40% Republican win probability.

evidence updates: Open seat (Shaheen retirement): +12-15% from base. John E. Sununu candidacy: +5% (credible but rusty). Chris Sununu declining: -5% (removed strongest candidate). Midterm headwinds for president's party: -5%. NH's consistent blue lean at federal level: -3%. No polling data available: uncertainty widens estimate toward 50% slightly.

combination method: Weighted average of sub-questions, with sq1 (direct fundamentals assessment) weighted most heavily at 40%. Sub-questions capture overlapping but distinct aspects of the race outcome.

final: NH is a genuinely competitive state, and the open seat makes this more competitive than recent Senate races there. However, the structural Democratic lean at the federal level, likely midterm headwinds against the president's party, and the absence of the strongest Republican candidate (Chris Sununu) all weigh against Republicans. Without polling data and with significant unknowns about the Democratic nominee and national environment, I estimate approximately 35% probability of a Republican win.

ensemble: {'ensemble_count': 3, 'perspectives': {'base_rate': {'p_yes': 0.35, 'confidence': 0.35}, 'evidence_driven': {'p_yes': 0.34, 'confidence': 0.42}, 'contrarian': {'p_yes': 0.35, 'confidence': 0.42}}, 'spread': 0.01, 'individual_p_yes': {'base_rate': 0.35, 'evidence_driven': 0.34, 'contrarian': 0.35}}

5. Devil's Advocate Sonnet · legacy v2 Market price revealed

Reasoning Flaw Math Error Asymmetric Evidence Missing Info

Challenges

The base rate calculation appears to underweight the historical record. The forecaster notes Republicans won 1 of 6 NH Senate races since 2010 (~17%), but then applies adjustments that nearly double this to 33-35%. The open seat adjustment of +12-15% seems too generous given that NH's federal lean has been consistently Democratic. Open seats in blue-leaning states during midterms unfavorable to the incumbent president's party don't historically swing +12-15% toward the out-of-power party — that figure seems more appropriate for truly neutral states.
The 'midterm headwind' adjustment is applied inconsistently. The forecaster notes 2026 is a midterm under a Republican president (which hurts Republicans), but only applies a -5% adjustment. Historically, the president's party loses ground in midterms, and since Republicans are the president's party in 2026, this should be a LARGER negative adjustment for Republican Senate prospects, not a smaller one. The forecaster seems confused about the direction — midterm headwinds hurt the PRESIDENT'S party (Republicans), so this should be a much stronger downward factor.
The sub-question weighting and combination method is poorly specified. The forecaster says sq1 gets 40% weight but doesn't specify weights for sq2-sq4. More importantly, sq4 (whether a strong candidate has emerged) is not directly about winning probability — it's an intermediate factor that should feed into sq1 and sq3, not be combined as a separate probability. Including it as a separate term inflates the combined estimate.
The forecaster treats the contrarian perspective as essentially identical to the base rate and evidence-driven perspectives (spread of only 0.01), which defeats the purpose of ensemble diversification. A genuine contrarian view might consider scenarios where the Republican wins (e.g., a major Democratic scandal, economic collapse benefiting the out-party narrative) OR where the Republican does even worse (e.g., a nasty primary, Trump toxicity in NH suburbs). The ensemble is not actually providing diverse viewpoints.
The 'open seat' adjustment seems to double-count. The forecaster gives +12-15% for the open seat in the base rate section, then treats Shaheen's retirement as a separate [STRONG][UP] evidence update. These are the same factor being counted twice.
The forecaster acknowledges massive information gaps (no polling, no Democratic candidate identified, no expert ratings, no fundraising data, no Trump NH approval) but doesn't adequately widen uncertainty toward the prior/base rate. With this many unknowns 566 days out, the estimate should lean more heavily on structural factors (NH's blue lean, midterm dynamics) rather than candidate-specific speculation about John E. Sununu's viability.
John E. Sununu's candidacy gets a net positive adjustment, but the evidence for this is weak. He lost re-election in 2008 by 6 points in a wave year, has been out of politics for 17 years, and the political landscape of NH has shifted considerably more blue at the federal level since his tenure. A comeback after 17 years is historically very unusual and rarely successful.

Suggested adjustment: -10pp

Multiple reasoning flaws inflate the estimate: (1) double-counting the open seat factor, (2) underweighting midterm headwinds against the president's party, (3) over-crediting John E. Sununu's candidacy despite a 17-year absence, (4) the ensemble lacking genuine diversity. The structural factors — NH's consistent federal Democratic lean, midterm dynamics hurting the president's party, absence of the strongest Republican candidate, and no identified weak Democratic opponent — all point to a probability lower than 35%. Adjusting down by ~10 percentage points to approximately 25% better reflects the fundamentals while maintaining appropriate uncertainty given the 566 days to resolution and massive information gaps.

Market comparison: Forecaster: 34.7%, Market: 13.0%. The market has high liquidity and depth (297,601 contracts), suggesting informed pricing. The 21.7% divergence is large. While the market can be wrong, the forecaster's reasoning contains genuine flaws (double-counting, inconsistent midterm adjustment, thin ensemble) that inflate the estimate. The true probability likely lies between the market and the forecast — the market at 13% may be somewhat low given the genuine uncertainty 566 days out and the open-seat dynamic, but the forecast at 35% is too high given the structural Democratic advantages and reasoning errors identified.

6. Aggregator Opus · final calibrated forecast

17% (45% confidence)