Scan: Mitch McConnell steps down from Senate before his term ends?

date	predicted	market price	confidence
2026-09-03	14%	12%	62%
2026-08-25	16%	18%	64%
2026-08-16	19%	20%	62%

1. Planner Fable 5.1 · Fermi decomposition + tool plan

Tools chosen: polymarket_direct polymarket_related kalshi_related article_search claude_news gdelt_news wikipedia code_execution

Sub-questions (Fermi decomposition)

What is the current Polymarket YES price and how has it trended over the past 90 days?
What is McConnell's current health status — any recent falls, hospitalizations, freezing episodes, or reports of decline in the last 3-6 months?
What is the actuarial probability that an 83-84-year-old US male dies or becomes incapacitated within the ~12 months remaining before January 3, 2027?
Has McConnell or his office made any statements hinting at leaving before the end of his term, or explicitly reaffirming he will serve out the full term?
Who would appoint a successor under Kentucky law (Democratic Gov. Beshear vs. GOP-restricted appointment rules), and does that create incentives for or against early resignation?
What is the historical base rate of senators aged 80+ who announced retirement resigning or dying before their term ended?
Are there any Kalshi or other Polymarket markets on McConnell's health, resignation, or Senate departures that provide cross-market signals?

Planner reasoning

McConnell (born Feb 1942, age 83-84) has announced he will not seek reelection in 2026, so his term ends Jan 3, 2027; the question is whether he leaves early via resignation, death, or incapacity. Key drivers are his health trajectory (falls, freezing episodes, hearing/mobility issues), any signals about early departure or succession dynamics in Kentucky (Gov. Beshear is a Democrat, which affects appointment incentives and makes voluntary early resignation less likely for Republicans), and actuarial mortality for an 84-year-old male over roughly 12 months. Market price is the primary anchor; cross-venue and news research refine it.

2. Research Parallel tool calls

tool	status	time	items	summary
polymarket_direct	OK	0.2s	1	## This Market's Polymarket Data Mitch McConnell steps down from Senate before his term ends? - Current price (probability): 17.50% - 7-day price change: +3.00% - 30-day price change: -5.00% - Total volume: $1,598,445 (USD notional) - Price range: 10.50% - 76.50% - Data points: 91 days
polymarket_related	OK	4.0s	0	Scanned 100 active Polymarket markets, kept 0 matches. keyword 'McConnell': 0 markets \| keyword 'resign Senate': 0 markets \| keyword 'senator resign': 0 markets
kalshi_related	OK	3.8s	2	2 related markets / summaries. keyword 'McConnell': no matches \| keyword 'senator resign': ok \| keyword 'Senate vacancy': ok
article_search	OK	0.5s	23	Fetched 23 unique articles across 4 queries (lookback=120d). 'Mitch McConnell health': 10 articles \| 'Mitch McConnell fall hospitalized': 10 articles \| 'Mitch McConnell resign early Senate': 10 articles \| 'McConnell serve out term 2027': 10 articles
claude_news	OK	23.4s	12	## Findings - Retirement already announced (not resignation): McConnell announced his decision to retire from the Senate when his term ends in January 2027, clearing the way for an already bubbling race to replace him. He has repeatedly said he intends to serve out the full term rather than r
gdelt_news	OK	102.2s	10	GDELT: 10 articles across 2 queries (lookback=60d). 'Mitch McConnell health': error GDELT rate-limited after retries (429) \| 'McConnell resign': 10 hits
wikipedia	OK	0.2s	2	Fetched 2 Wikipedia entries (1 missing pages).
code_execution	OK	21.7s	0	## Key Findings - Baseline SSA mortality (blended age 83→84 over ~12-month horizon): ~10.7%–10.9% annualized death probability, using a time-weighted blend (≈15% of the year at age 83, ≈85% at age 84 given his Feb. 1942 birth date and the Jan. 3, 2027 term end). - **Frailty-adjusted mortality

3. Evidence Brief Sonnet 5 · 7828 chars

# Current state McConnell has publicly committed to serving out his full term (through Jan 3, 2027) and has NOT resigned; his office confirmed in July 2026 he "will retire at the end of his term next year." However, he suffered a serious fall/pneumonia in June 2026, was hospitalized ~3+ weeks, and remained absent from the Capitol for roughly three months, prompting Gov. Beshear to publicly call on him to prove capacity or resign. As of mid-September 2026, GOP colleagues expected (but had not confirmed) his return to the Senate floor; no resignation has occurred. # Timeline of key events - 2024-02-28 (confirmed): McConnell announces he will step down as Senate GOP leader (not resign from Senate) effective Jan 2025. - 2025-02-20 (confirmed): McConnell announces he will not seek an 8th term in 2026 — will retire when current term ends (Jan 2027). - 2026 (unspecified, ~Feb, confirmed): McConnell hospitalized >1 week for "flu-like symptoms." - 2026-06-14 (confirmed): McConnell hospitalized; office gives no explanation ("receiving excellent care"). - 2026-07-02 (reported): CBS reports EMS had responded to his home for an "unconscious" person prior to hospitalization. - 2026-07-06/07 (confirmed): Still hospitalized 3-4 weeks later; no votes cast since June 11; office withholding details. - 2026-07-12 (confirmed): McConnell issues statement — hospitalization caused by a fall and pneumonia; says he "remains fit to serve" but not yet returning to the floor; reaffirms he will retire at term's end (not early). - 2026-07-28 (confirmed): Gov. Beshear publicly demands McConnell disclose health status or resign if unable to serve. - 2026-08 (reported): McConnell moves to rehab, then continues recovery at home; ~3-month absence from Capitol. - 2026-08-24 (rumored/reported): Resurfaced video showing McConnell "nearly motionless" fuels renewed health concern. - 2026-09-03/06 (reported): GOP colleagues (Rep. Comer) say staff expect his return around Sept 14 session; Fox host publicly says he "must return to work — or quit." - 2026-09-10/12 (reported): Still no confirmed public appearance by McConnell himself since the fall; return remains "not set in stone." # Event Will Mitch McConnell formally announce intent to leave, or otherwise vacate, his Senate seat before his term ends Jan 3, 2027? # Outcomes to forecast - Yes (resigns/vacates early) - No (serves out full term) # Kalshi market anchor Kalshi ticker itself not directly queried in raw data (kalshi_direct not returned), but Polymarket equivalent market shows **current YES price 17.5%**, up +3% over 7 days, down -5% over 30 days; volume $1.6M; historical range 10.5%-76.5% over 91 days (the high end likely reflects acute hospitalization panic in June-July 2026). Treat ~17-18% as the consensus anchor. # Sub-question answers 1. **Polymarket YES price/90-day trend** — Currently 17.5%; spiked as high as 76.5% during acute June/July 2026 hospitalization uncertainty, has since receded to 10.5%-17.5% range as McConnell stabilized and reaffirmed intent to serve out term. [polymarket_direct] 2. **Recent health status** — Serious: hospitalized June 14, 2026 for ~3-4 weeks (fall + pneumonia, confirmed July 12); ~3-month Capitol absence into September 2026; prior history includes 2023 concussion/fall, 2023 freezing episodes, 2024 fall. As of mid-Sept 2026, no confirmed public reappearance yet, though staff/colleagues expect return. [CBS, WaPo, CNN, CNBC, claude_news] 3. **Actuarial mortality probability** — Blended SSA-based estimate for ~12-month remaining horizon (age 83→84): ~10.7-10.9% baseline; frailty-adjusted (fall/health history) ~12.8-16.3%, midpoint ~14.6%. [code_execution] 4. **Statements on early departure** — McConnell's July 12, 2026 statement explicitly says he "will retire at the end of his term" (i.e., no early exit) while confirming he's not yet medically cleared to return to the floor. No qualifying "leave early" statement found. [CNBC, claude_news] 5. **Kentucky succession law** — 2024 KY House Bill 622 stripped Gov. Beshear's appointment power; a vacancy would trigger a special election rather than gubernatorial appointment, removing the political incentive/fear that Republicans would lose the seat to a Democratic appointee, and thus removing one disincentive against resignation. [claude_news/The Hill] 6. **Base rate for senators 80+ post-retirement-announcement** — Not directly available (Wikipedia "died in office" list page missing); no specific historical base rate was retrieved. 7. **Cross-market signals** — Kalshi's own McConnell market not found via keyword search; comparable Kalshi markets: "Will Trump resign during his term" at 22% YES; "Which SCOTUS justices resign — Alito" at 64%. No Polymarket related markets found (0 matches). [kalshi_related, polymarket_related] # Key facts (high-confidence, factual) 1. [Wikipedia] McConnell (b. Feb 20, 1942) announced Feb 20, 2025 he will not seek re-election in 2026; term ends Jan 3, 2027. 2. [CBS/CNN/WaPo] Hospitalized June 14, 2026; office disclosed July 12 cause was a fall + pneumonia; no votes cast since June 11, 2026. 3. [CNBC/claude_news] July 12, 2026 statement: "will retire at the end of his term next year" — explicit reaffirmation of full-term completion, not early exit. 4. [claude_news/The Hill] KY HB622 (2024) removes governor's appointment power; special election required for vacancy. 5. [Polymarket] Current YES price 17.5%, having fallen from a peak of 76.5% during acute crisis. # Cross-market signals - Kalshi related: "Trump resign" market at 22% YES (different individual, limited comparability); "SCOTUS Alito resign" at 64% (different profile). - Polymarket: This market itself is Polymarket-sourced; 17.5% YES, volatile (10.5-76.5% range), no other directly related Polymarket markets found. - Sportsbook implied: none available. # Analyst opinions and speculation - Fox commentator (Sept 2026) publicly said McConnell "must return to work — or quit," reflecting mounting media pressure narrative. [aol.com/independent] - Gov. Beshear (D) has twice (July 28, Sept 10) publicly pressed McConnell to prove fitness or resign — political pressure but no legal authority to force resignation. - GOP colleagues (Comer) expect return, not resignation, as of early September 2026. # Directional lean per outcome - **Yes (resigns early)**: Supported by advanced age (84), documented fall/pneumonia, ~3-month absence, opaque health disclosures, public pressure to resign, elevated actuarial mortality/incapacity risk (~15-21% combined per code_execution model), removal of KY appointment disincentive. - **No (serves full term)**: Supported by explicit July 12, 2026 statement reaffirming intent to complete term, GOP colleague expectations of imminent return, historical pattern of McConnell resisting resignation calls despite health scares (2023-2024), and Polymarket pricing having declined from crisis peak back toward ~17.5%. # Gaps / unknowns - No confirmed public reappearance by McConnell as of latest data (mid-September 2026); resolution of his September return is pivotal and unresolved in the research. - No historical base rate data retrieved for 80+ senators completing vs. not completing term after retirement announcement. - Kalshi's own direct YES price for this specific ticker was not returned in raw research (only Polymarket data available as anchor). # Calibration anchors - Polymarket YES price (anchor): 17.5%, recent range 10.5%-76.5% over 91 days. - Actuarial/model estimate: ~15-21% (midpoint ~18%) combined probability of death, incapacity, or resignation before Jan 2027. - Precedent: McConnell has survived multiple serious health scares (2023 concussion, freezing episodes, 2024 fall) without resigning, suggesting strong personal resistance to early departure absent incapacitation.

4. Forecasters Fable 5.1 + GPT-5.6 Sol · in parallel

claude-fable-5-1

46% conf

Yes	17%
No	83%

The Polymarket consensus sits at 17.5% YES after receding from a 76.5% crisis peak, and McConnell's July 12, 2026 statement explicitly reaffirmed he will retire only at term's end, consistent with his history of resisting resignation calls through the 2023-2024 health scares. Only ~3.5 months remain in the term, so frailty-adjusted mortality/incapacity risk over the horizon is roughly 4-6%, with voluntary resignation under political pressure (Beshear, media, three-month absence, and KY's special-election succession removing a GOP disincentive) adding perhaps 10-12%. Colleagues expected a mid-September return rather than resignation, and no announcement had occurred as of the latest data. I stay essentially at the market anchor, ~17% YES.

gpt-5.6-sol

68% conf

Yes	18%
No	82%

The 17.5% Polymarket YES price provides the best consensus anchor, and its retreat from the hospitalization-driven peak indicates the market largely credits McConnell’s stabilization. His serious fall, pneumonia, prolonged absence, advanced age, and lack of a confirmed public reappearance keep early vacancy risk material, with the actuarial/incapacity estimate broadly supporting a high-teens probability. Conversely, his explicit July 12 commitment to retire only at the end of his term, colleagues’ expectation of his return, and his history of resisting resignation after health scares favor No. Kentucky’s special-election law modestly reduces the political barrier to resignation but does not outweigh his stated intent.

5. Devil's Advocate Sonnet 5 · red-team critique

Both forecasts converge tightly on the Polymarket anchor (17-18%) but neither flags that Kalshi's own price was never retrieved ("Kalshi ticker itself not directly queried... kalshi_direct not returned"), so they are implicitly assuming Polymarket = Kalshi consensus without justification.

Neither forecaster meaningfully engages with the single most pivotal unresolved fact in the brief — "No confirmed public reappearance by McConnell as of latest data (mid-September 2026)" — which the brief itself calls "pivotal and unresolved"; if he fails to reappear by the Sept 14 session, that should meaningfully shift probability upward, yet both treat it as a minor footnote.

Forecast 2 cites "the actuarial/incapacity estimate broadly supporting a high-teens probability" without rescaling for the shortened ~3.5-4 month remaining horizon, while Forecast 1 correctly notes mortality/incapacity risk over this window is closer to 4-6% — this is a real inconsistency between the two forecasts that isn't reconciled or flagged.

Neither forecaster separately models death-in-office vs. voluntary resignation vs. forced incapacity, even though the brief provides distinct actuarial mortality figures (~10.7-10.9% baseline, frailty-adjusted up to 16.3%) that should be decomposed rather than folded into a single vague "resignation risk" bucket.

Forecast 2's stated confidence (0.68) is inconsistent with its own rationale, which lists substantial countervailing uncertainty (unclear health status, no reappearance, prior resignation-resistance history) — this overconfidence isn't justified by the evidence, especially given the unresolved Sept 2026 return status.

Neither forecaster addresses the explicit data gap noted in the brief — "no historical base rate data retrieved for 80+ senators completing vs. not completing term after retirement announcement" — meaning both are implicitly relying on a single anecdote (McConnell's own 2023-2024 resilience) as their reference class rather than acknowledging this as thin evidence.

6. Aggregator Fable 5.1 · final calibrated forecast

17% (57% confidence)

Yes: 17% No: 83%

Both forecasters landed at the Polymarket anchor (~17.5%) and the critique does not expose a shared directional flaw, though it correctly notes the actuarial figures must be rescaled to the ~3.5-month remaining horizon: frailty-adjusted mortality/incapacity over that window is roughly 4-6%, not high-teens. The residual ~10-12% comes from voluntary or pressured resignation given a three-month absence, no confirmed public reappearance as of mid-September, Beshear/media pressure, and Kentucky's special-election law removing the GOP disincentive to vacate. Against this, McConnell's explicit July 12 reaffirmation that he will retire only at term's end, colleagues' expectation of an imminent return, and his consistent refusal to resign through prior health scares (2023-2024) anchor the No case. Since Yes plausibly includes death or incapacity vacating the seat, I keep Yes near the market at 17% rather than moving below it, and see no evidence beyond what the market has priced to justify moving above it.

Pipeline Timing

Total pipeline time: 206.5s

Per-tool research timings shown in the Research section above.