Scan: Will Democratic win the House race for FL-22?

← Back to scans

Will Democratic win the House race for FL-22? — Democratic party

KXHOUSERACE-FL22-26-D · Elections · 2026-04-26

82%

Agent

73%

Market Price

+9.0%

Edge

55%

Confidence

Volume: 15,054

Spread: 3.0c

Days to resolution: 555

Markets in event: 2

Final Rationale

My independent estimate of 92.7% relied heavily on Frankel's incumbency in a D+4 district during a likely anti-Trump midterm, but I PARTIALLY ACCEPT several critiques: the upward adjustment from 0.89 to 0.927 was not well-justified (reasoning_flaw), Frankel's age 77 creates a non-trivial 5-10% retirement/health tail risk over 555 days, and the PVI shift from D+7 to D+4 in one cycle signals real district trend risk that my base rate understated. I REJECT the multiplicative independence framing in Challenge 2 since these factors are highly correlated, but ACCEPT that the weighted aggregation was inflated. The market at 73% with high orderbook depth and a recent surge already prices in the Democratic-favorable midterm signal, and I have no FL-22-specific superior information (no polls, no Cook rating, no fundraising) to justify a 20-point divergence — I settle at 82%, which preserves the structural edge from incumbency and D+4 PVI in a midterm against the president's party but respects the market and quantifies tail risks more honestly.

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 article_search

Sub-questions (Fermi decomposition)

Is FL-22 currently held by a Democrat or does it lean Democratic based on historical voting patterns and Cook PVI? (w=0.35)
Will the national political environment in 2026 favor Democrats enough to win competitive House seats in Florida? (w=0.30)
Will Democrats field a strong candidate and have sufficient funding to compete in FL-22 in 2026? (w=0.20)
Does Florida's congressional map (as redrawn under DeSantis) make FL-22 competitive enough for Democrats to win? (w=0.15)

Planner reasoning

FL-22 is a South Florida district (Palm Beach area) historically held by Democrat Lois Frankel. The key analytical framework involves: (1) Baseline partisanship — FL-22 has a strong Democratic lean given its Palm Beach coastal demographics, Jewish community, and suburban composition; if Frankel still holds the seat, this is a safe Dem district. (2) Incumbency advantage — if Frankel or another incumbent Democrat runs, the default probability is very high (85-95%). (3) National environment — 2026 is a midterm year with Trump as president; historically, the president's party loses House seats in midterms, which would favor Democrats. However, Florida has been trending Republican under DeSantis-era redistricting. (4) Redistricting impact — the DeSantis 2022 map significantly altered Florida's congressional map; FL-22's boundaries and competitiveness need verification. Temporal analysis: With 555 days until resolution, the 2026 primaries will occur in August 2026, followed by the general election in November 2026. The sequence needed for a Republican win would be: (a) Democrats field a weak candidate or no incumbent runs, (b) Republican recruits a strong challenger, (c) national environment favors Republicans despite historical midterm disadvantage, (d) Florida's rightward trend continues. The most critical reference class is Cook Political Report ratings for FL-22 — if rated Safe D, probability is ~93%; if Likely D, ~80%; if Lean D, ~65%; if Toss-up, ~50%. Given the district's historical composition, I expect Safe or Likely Democratic status, making this a high-probability Democratic hold absent unusual circumstances.

2. Research Parallel tool calls · legacy v2

Tool	Status	Time	Items
kalshi_data	OK	2.4s	-
kalshi_orderbook	OK	0.1s	-
article_search	OK	0.8s	-
web_search	OK	28.4s	-
wikipedia_lookup	OK	1.2s	-

3. Evidence Synthesis Sonnet · 15 items · legacy v2

#	Strength	Credibility	Direction	Source	Claim	Priced In
1	STRONG	90	↑ UP	web_search	Incumbent Lois Frankel (D) currently holds FL-22, having won re-election in 2024 with 55.0% of the vote against Republican Dan Franzese.	Yes
2	STRONG	85	↓ DOWN	web_search	FL-22's Cook PVI for 2026 is D+4, a notable 3-point rightward shift from its 2024 PVI of D+7, indicating the district has become moderately less Democratic.	Yes
3	MODERATE	82	↑ UP	web_search	FL-22 covers Palm Beach, West Palm Beach, Boynton Beach, and Delray Beach — a South Florida coastal district with a Democratic-leaning demographic composition.	Yes
4	STRONG	72	↑ UP	kalshi_data	Kalshi prediction market prices FL-22 Democratic win at 73%, up 23 points over 30 days, suggesting growing market confidence in a Democratic hold.	No
5	MODERATE	72	↑ UP	article_search	A March 2026 CNN poll found Democrats hold an early advantage among 'double haters' (voters with negative views of both parties), who prefer Democrats in midterms by 31 points.	Yes
6	STRONG	85	↑ UP	article_search	Democrats won a Florida special election in a Palm Beach district that included Mar-a-Lago, a district previously won by Republicans by 19 points in 2024 — a massive Democratic overperformance signaling a favorable national environment.	No
7	MODERATE	75	↑ UP	article_search	Democrats have been 'overperforming in elections' broadly since Trump took office, consistent with the historical pattern of the president's party losing House seats in midterms.	No
8	MODERATE	70	↓ DOWN	article_search	Trump is actively attempting to influence and change midterm election administration, including pushing for redistricting and mail ballot restrictions, which could suppress Democratic turnout.	Yes
9	MODERATE	68	↓ DOWN	article_search	The Supreme Court's conservative majority appeared ready in March 2026 to limit mail-in ballot grace periods, which could reduce Democratic turnout in Florida and other states ahead of November 2026.	Yes
10	MODERATE	65	NEUTRAL	kalshi_data	Kalshi market for Democrats winning above 249 House seats is priced at only 24%, suggesting the overall national environment may not produce a massive Democratic wave sufficient to flip all competitive seats.	Yes
11	STRONG	88	↑ UP	web_search	Incumbent Lois Frankel is running for re-election in FL-22's Democratic primary alongside two other candidates (Ian Blake and Victoria Doyle), indicating a competitive Democratic primary but with an incumbent in the race.	Yes
12	MODERATE	65	↑ UP	web_search	No Republican candidate profiles have been created for FL-22 as of the research date, suggesting Republicans have not yet recruited a strong or notable challenger.	No
13	MODERATE	75	↑ UP	wikipedia_lookup	FL-22 was redrawn in the 2020 redistricting cycle as a successor to the previous 21st district, covering Palm Beach-area communities — it was not among the most aggressively gerrymandered districts under DeSantis's 2022 map.	Yes
14	MODERATE	80	↓ DOWN	web_search	The Cook PVI shift from D+7 to D+4 between cycles suggests that while FL-22 remains Democratic-leaning, Florida's overall rightward trend has modestly reduced the margin of safety for Democrats.	Yes
15	MODERATE	78	↑ UP	article_search	Democrats won a Palm Beach County state legislative special election in March 2026 in a district Republicans won by 19 points in 2024, demonstrating that even Florida's map disadvantage can be overcome by a strong national environment.	No

Information Gaps

No Cook Political Report or Sabato's Crystal Ball rating for FL-22 specifically in 2026 was found — this would be the single most important data point for calibrating probability.
No fundraising data for FL-22 2026 candidates (Frankel or Republican challengers) to assess financial competitiveness.
No polling data specific to FL-22 2026 general election matchups.
Republican candidate field for FL-22 is entirely unknown — no named challengers or profiles exist yet, making it impossible to assess GOP candidate quality.
No information on whether Byron Donalds vacating his congressional seat (to run for governor) affects FL-22 dynamics or Republican recruitment efforts statewide.
No specific information on whether Lois Frankel's incumbency position is strong or if she faces any retirement risk.
No data on Trump approval ratings specifically in FL-22 to gauge local sentiment beyond district-level PVI.

Key Uncertainties

Whether Lois Frankel wins the Democratic primary or if a weaker challenger emerges — incumbent loss in primary would significantly reduce Democratic win probability.
Whether Republicans recruit a strong, well-funded challenger for FL-22 before the June 2026 filing deadline.
Whether the Supreme Court ruling on mail ballot grace periods, if adverse, materially suppresses Democratic turnout in Florida.
Whether Florida's broader rightward trend continues to erode the D+4 PVI advantage between now and November 2026.
Whether Trump's attempts to nationalize and influence elections have a measurable electoral effect in Florida specifically.
Whether the Democratic overperformance trend seen in special elections (including the Palm Beach Mar-a-Lago district) continues through November 2026 or fades.
Potential redistricting mid-cycle if Trump pressures Florida Republicans to redraw maps before 2026.

Market Context

Price: 0.73 · 7d: +0.08 · 30d: +0.23
Volume: MODERATE · Depth: HIGH
Reliability: MEDIUM-HIGH — moderate daily volume (502 contracts), high orderbook depth (24,145), but the 23-point 30-day surge suggests a significant information event or thin early market that may have overcorrected; current price broadly consistent with D+4 PVI district with incumbent and favorable midterm environment

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

Independent estimate: 93% (78% confidence)

base rate

92%

78% conf

evidence driven

93%

78% conf

contrarian

93%

78% conf

Spread

1pp

Low disagreement

sq1: Is FL-22 currently held by a Democrat or does it lean Democratic based on historical voting patterns and Cook PVI?

97%

This is essentially a factual question. Frankel (D) holds the seat, won 55% in 2024, and PVI is D+4. Even with rightward shift, the district is clearly Democratic-leaning. Near certain YES.

sq2: Will the national political environment in 2026 favor Democrats enough to win competitive House seats in Florida?

78%

Base rate: president's party loses House seats in ~90% of midterms since WWII, averaging ~26 seats. Trump is the incumbent, so Democrats are favored nationally. Strong evidence: Democratic overperformance in special elections including a Palm Beach R+19 district flip. Headwinds: voter suppression efforts, SCOTUS mail ballot rulings. For a D+4 district specifically, a neutral-to-favorable environment is sufficient. Estimate 78%.

sq3: Will Democrats field a strong candidate and have sufficient funding to compete in FL-22 in 2026?

93%

Frankel is a 7-term incumbent running for re-election. Incumbents win primaries ~95%+ of the time absent scandal. No strong Republican recruitment evident. Very high probability.

sq4: Does Florida's congressional map make FL-22 competitive enough for Democrats to win?

88%

D+4 PVI district with Democratic incumbent — clearly winnable for Democrats. The map disadvantage is modest, not prohibitive. Mid-cycle redistricting risk exists but unlikely to specifically hurt FL-22 given its compact urban Palm Beach geography. High probability.

Show full reasoning trace

base rate: Incumbent House members in D+4 districts running for re-election win approximately 95% of the time. In midterms against the president's party, this rises further. Historically, only ~5-10% of incumbents in their party-leaning districts lose.

evidence updates: Frankel running as incumbent (+): maintains base rate. PVI shifted D+7 to D+4 (-): modest downward pressure ~3%. Democratic special election overperformance in Palm Beach (+): reinforces favorable environment ~3%. No strong GOP challenger recruited (+): ~2%. Florida rightward trend and voter suppression efforts (-): ~3%.

combination method: Weighted average gave 0.89, but I adjust slightly upward because sq1 (district lean) and sq3 (incumbent running) are the dominant determinants and both are very high. The reference class of incumbent Democrat in D+4 district during anti-Trump midterm strongly suggests 90%+.

final: Probability of Democratic hold in FL-22: ~92%. Main risks are Frankel retirement/health issue, unexpected scandal, or extreme national environment shift. Confidence is moderately high given clear incumbent advantage and favorable midterm dynamics.

ensemble: {'ensemble_count': 3, 'perspectives': {'base_rate': {'p_yes': 0.92, 'confidence': 0.78}, 'evidence_driven': {'p_yes': 0.93, 'confidence': 0.78}, 'contrarian': {'p_yes': 0.93, 'confidence': 0.78}}, 'spread': 0.01, 'individual_p_yes': {'base_rate': 0.92, 'evidence_driven': 0.93, 'contrarian': 0.93}}

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

Overconfidence Missing Info Reasoning Flaw Asymmetric Evidence

Challenges

The forecaster's base rate of '95% for incumbents in D+4 districts' may be overstated. Historical incumbent re-election rates of ~95% apply to ALL incumbents including those in safe seats; for D+4 districts specifically, the rate is closer to 85-90%, especially when PVI has been trending against the party (D+7 → D+4 in one cycle is significant).
The combination method is opaque and arguably inflates the result. With sub-probabilities of 0.97, 0.78, 0.93, 0.88, treating these as roughly independent multiplicative factors yields ~0.59, while a strict 'all must be true' framing gives a much lower number than 0.927. The forecaster acknowledges weighted average gave 0.89 but then 'adjusts upward' — this adjustment is not well-justified and looks like motivated reasoning.
sq2 at 78% may be too high as a standalone factor but the way it's incorporated essentially ignores its downside. If the national environment is NOT favorable (22% chance), Democrats winning a D+4 district where PVI just shifted 3 points right becomes meaningfully harder. The forecast doesn't properly account for this conditional risk.
555 days to resolution is a LONG time. Many things can happen: Frankel retirement (she's 77 years old — non-trivial retirement/health risk for a 7-term member), scandal, redistricting changes, candidate recruitment changes. The forecaster identifies these but doesn't quantify them adequately — even a 5-8% combined tail risk from these factors alone would pull the estimate down.
The market at 73% with high orderbook depth and a recent 23-point upward surge suggests informed traders see meaningful risk the forecaster may be underweighting. While the market isn't ground truth, a 20-point divergence with no clearly-superior information on the forecaster's side warrants reconsideration. The forecaster has no FL-22-specific polling, no Cook rating, no fundraising data — significant information gaps acknowledged but not reflected in confidence.
Frankel's age (77) is a notable retirement/health risk over 555 days that isn't explicitly quantified. Base rate for House members 75+ retiring or having health events over an 18-month window is non-trivial (5-10%).

Suggested adjustment: -10pp

The forecast of 92.7% is overconfident given: (1) acknowledged information gaps (no Cook rating, no polling, unknown GOP field, no fundraising data), (2) the unjustified upward adjustment from 0.89 weighted average to 0.927, (3) inadequate quantification of Frankel's retirement/health risk at age 77 over 555 days, (4) the meaningful PVI shift from D+7 to D+4 indicating real district trend risk, and (5) failure to properly condition on the 22% scenario where national environment isn't favorable. A range of 0.80-0.85 better reflects genuine uncertainty.

Market comparison: Forecaster: 92.7%, Market: 73.0%. The 20-point divergence is large. Market has been trending up sharply (+23 over 30 days), suggesting it's already incorporating Democratic-favorable signals the forecaster cites. The forecaster has no FL-22-specific superior information (no polls, no Cook rating, no fundraising data) to justify being 20 points above a liquid market with high orderbook depth. Some divergence is justified by structural factors (D+4 PVI, incumbent, midterm dynamics), but the magnitude suggests overconfidence.

6. Aggregator Opus · final calibrated forecast

82% (55% confidence)

7. Entry Decision Opus Trader

Side

YES

Position Size

$750

Contracts

987

Trader Confidence

55%

Trade Rationale

The forecast indicates a +9% edge on YES (82% vs 73% market) for a Democratic incumbent (Frankel) in a D+4 district during a likely anti-Trump midterm. However, the forecaster confidence is moderate (0.55) and the Devil's Advocate raised multiple flags (overconfidence, missing_info, reasoning_flaw, asymmetric_evidence) suggesting -0.10 adjustment. The 555-day horizon is very long with real tail risks (Frankel age 77 retirement/health). Portfolio already has heavy election exposure including a similar VA-06 Dem House bet on NO side, but FL-22 is uncorrelated with most positions.

Allocation Logic

Sized below baseline at $750 because the moderate edge (9%) is partially offset by low forecaster confidence (0.55), multiple critic flags, and the very long 555-day resolution window which amplifies tail risks. Not skipping because the structural edge (D+4 incumbent in midterm vs president's party) is real.

Entry price: $0.76

Current: $0.46

Status: OPEN

P&L: -$296.05