Tools chosen: polymarket_direct polymarket_related kalshi_related claude_news gdelt_news wikipedia code_execution

Sub-questions (Fermi decomposition)

Planner reasoning

This is a Polymarket question on whether the KMT wins the most of Taiwan's 22 mayor/magistrate races in November 2026. The KMT won 14 of 22 in 2022, so the base rate strongly favors them; key uncertainties are candidate nominations, KMT-TPP cooperation/splits, DPP recovery after the 2024/2025 recall battles, and race-by-race polling in the six special municipalities. I will anchor on the Polymarket price, triangulate with Kalshi and related Polymarket markets, and use news search plus Wikipedia for nomination and polling details.

# Current state KMT enters the November 28, 2026 elections as the incumbent plurality-holder from 2022 (holding roughly 14 of 22 relevant seats per the Fermi model), buoyed by having survived all 2025 recall attempts and by an emerging (though strained) KMT–TPP electoral coordination pact. Nominations for most of the 22 races are now finalized (as of mid-2026), and Polymarket prices KMT retaining its plurality at 92.5%. No credible source projects a DPP seat-count overtake; commentary instead focuses on whether KMT-TPP cooperation holds together through election day. # Timeline of key events - 2022-11-26/12-18 (confirmed): KMT won a large plurality of local head-of-government seats (~14/22); DPP suffered major losses, Tsai Ing-wen resigned as DPP chair. - 2025-07-26 & 2025-08-23 (confirmed): Mass recall campaigns targeting KMT legislators fail entirely in both waves; scholars view this as affirming KMT's legislative strength. Lai Ching-te's approval fell to ~35% post-recall. - 2025-10 (confirmed): DPP begins nominating candidates (Wang Mei-hui/Chiayi, Chen Ying/Taitung, Ho Hsin-chun/Taichung); primaries organized in Tainan. - 2025-11 (confirmed): DPP formally nominates Su Chiao-hui (New Taipei) and Chen Pin-an (Miaoli). - 2025-12 (confirmed): KMT announces southern-Taiwan slate (Ko Chih-en/Kaohsiung, Hsieh Lung-chieh/Tainan, plus Pingtung/Taitung nominees). - 2026-01 (reported): Poll shows KMT's Chiang Wan-an with a 7-point lead in Taipei. - 2026-03-31 (confirmed): KMT names Johnny Chiang as Taichung mayoral candidate. - 2026-04 (confirmed): Huang Kuo-chang (TPP) suspends New Taipei bid; Lee Shu-chuan becomes joint KMT-TPP candidate there. - 2026-05 (reported): DPP nominates Puma Shen to challenge Chiang Wan-an in Taipei. - 2026-08 (reported): Taipei Times/Diplomat coverage shows KMT-TPP alliance "under strain" over the Jhubei (Hsinchu County) mayoral contest, though referendum cooperation continues (three opposition-backed referendums approved by Legislature, 2026-08-14). # Event Will KMT's official candidates win the most head-of-government (mayor/magistrate) races among Taiwan's 22 special municipalities/cities/counties in the November 28, 2026 local elections? # Outcomes to forecast - Yes (KMT wins plurality of the 22 seats) - No (another party, e.g., DPP, wins plurality or ties resolved alphabetically against KMT) # Kalshi market anchor No Kalshi-direct price was returned for this ticker in the raw research; only tangential Taiwan markets exist (travel advisory, Trump recognition — irrelevant). Use Polymarket as the primary cross-market anchor. # Sub-question answers 1. **Polymarket price/trend** — Currently 92.5% YES; up from 80% low, +5% over 30 days, +0.5% over 7 days; steadily rising trend, $61.4K volume over 90 days. [polymarket_direct] 2. **Current KMT seat count/term limits** — KMT held ~14 of 22 relevant seats after 2022 (Fermi model baseline); specific term-limit data not detailed in research, but incumbents like Chiang Wan-an (Taipei) are seeking reelection. [code_execution, claude_news] 3. **Polling in six special municipalities** — Only Taipei polling reported: KMT's Chiang Wan-an led DPP's Puma Shen by ~7 points (Jan 2026) and by wider margins (50%:25%) in late 2025. No comparable polling found for New Taipei, Taoyuan, Taichung, Tainan, or Kaohsiung. [claude_news] 4. **KMT-TPP coordination** — Yes, formal coordination underway: joint New Taipei candidate (Lee Shu-chuan) after Huang Kuo-chang withdrew; cross-party primary mechanism agreed by KMT chair Cheng Li-wun; but alliance reported "under strain" over the Jhubei (Hsinchu County) race as of Aug 2026. [claude_news, gdelt_news] 5. **Candidates nominated** — KMT: Chiang Wan-an (Taipei, incumbent), Johnny Chiang (Taichung), Ko Chih-en (Kaohsiung), Hsieh Lung-chieh (Tainan), Lee Shu-chuan (New Taipei, joint w/ TPP). DPP: Puma Shen (Taipei), Ho Hsin-chun (Taichung), Su Chiao-hui (New Taipei), Wang Mei-hui (Chiayi), Chen Ying (Taitung), Chen Pin-an (Miaoli), Chen Ting-fei (Tainan, reported). No defections/independent runs by KMT incumbents noted. [claude_news] 6. **National approval trends** — Lai/DPP approval fell to ~35% (historic low) after failed 2025 recalls; KMT seen as strengthened/legitimized as largest Legislature party; TPP viewed by analyst Templeman as having "peaked," risking decline. [claude_news] 7. **Historical base rate** — KMT won plurality in 3 of 4 cycles (2010, 2018, 2022), losing only in 2014 (DPP wave) → ~75% base rate. [code_execution] # Key facts (high-confidence, factual) 1. [Wikipedia] 2022 election: DPP suffered major losses in the north; Tsai resigned as DPP chair. 2. [claude_news] All 2025 recall attempts against KMT legislators failed (July 26 & Aug 23). 3. [claude_news] Lai's approval fell to ~35% post-recall. 4. [claude_news] KMT-TPP joint candidate secured in New Taipei (Lee Shu-chuan) after Huang Kuo-chang withdrew. 5. [gdelt_news, Aug 2026] KMT-TPP alliance strained over Jhubei mayoral contest. 6. [Polymarket] YES price 92.5%, uptrending over 30/90 days. # Cross-market signals - Kalshi related: No direct Taiwan-election markets found; unrelated Taiwan geopolitical markets (travel advisory 9.9%, Trump recognition 12%) offer no pricing signal. - Polymarket: 92.5% YES, rising steadily (80%→92.5% over 90 days), moderate volume ($61K). - Sportsbook implied: None available. # Analyst opinions and speculation - Kharis Templeman: 2026 results will determine KMT chair Cheng Li-wun's influence over 2028 nominations; TPP may be in structural decline. [claude_news] - East Asia Forum: 2026 is a test of whether KMT-TPP cooperation becomes a durable alliance; DPP sees it as a rebuilding opportunity ahead of 2028, not necessarily an opportunity to overtake KMT's seat plurality. [claude_news] - No analyst source predicts a DPP seat-plurality outcome. # Directional lean per outcome - **Yes (KMT plurality)**: Supported by incumbency (~14/22 seats), successful recall defense, weak DPP presidential approval, KMT-TPP coordination reducing vote-splitting, strong Taipei polling, high base rate (75% historically), and dominant Polymarket pricing (92.5%). Opposing: binomial retention modeling shows KMT needs to retain a high share of its existing seats, and moderate retention probabilities (60-70%) imply plausible risk of losing plurality; KMT-TPP strain (Jhubei) could cause vote-splitting in select races. - **No (DPP or other plurality)**: Supported only by structural seat-retention math under moderate erosion scenarios and TPP-KMT friction risk; no direct polling or nomination evidence supports a DPP overtake in marquee races. # Gaps / unknowns - No polling data for New Taipei, Taoyuan, Taichung, Tainan, or Kaohsiung — the pivotal swing municipalities. - Exact current KMT seat count and which seats are open/term-limited not explicitly detailed. - Kalshi-direct price for this specific ticker was not returned; anchor relies solely on Polymarket. - Extent/resolution of KMT-TPP Jhubei dispute and its downstream effect on other close races unclear. # Calibration anchors - Polymarket YES price (anchor): 92.5%, uptrending. - Historical base rate: KMT won plurality in 3/4 cycles since 2010 (~75%). - Binomial seat-retention model: implies wide range (4%–70%) depending on assumed incumbent retention probability (60–80%) and tie-threshold assumptions — median plausible range ~20–50%, well below Polymarket's 92.5%, indicating potential overconfidence in the market price absent race-level polling confirmation.

1. Both forecasts anchor almost entirely on Polymarket's 92.5% and barely engage with the brief's own binomial seat-retention model, which explicitly states median plausible range is ~20–50%, "well below Polymarket's 92.5%, indicating potential overconfidence in the market price absent race-level polling confirmation" — neither forecaster explains why they dismiss this internal red flag rather than just noting it as a modest risk. 2. Neither forecast grapples with the ties-resolution rule embedded in the outcome definition ("ties resolved alphabetically against KMT"), which is a structural asymmetry that should slightly depress YES probability in close-plurality scenarios, yet both treat this as negligible. 3. The lack of polling in Taoyuan, Taichung, Tainan, and Kaohsiung — the "pivotal swing municipalities" per the brief — is acknowledged by both but not meaningfully priced in; a plurality outcome could hinge on 2-3 close races in exactly these unpolled areas, and DPP has contested candidates in several (Ho Hsin-chun/Taichung, Chen Ting-fei/Tainan) that neither forecaster weighs individually. 4. Both forecasters treat KMT-TPP strain (Jhubei) as a minor "modest discount" factor, but the brief flags this as a source of "opposition vote-splitting failures in close races" that could realistically flip 2-4 marginal seats — given KMT's plurality margin might be thin, this mechanism deserves more explicit seat-count modeling rather than a vague haircut. 5. Neither forecast reconciles the tension between the historical base rate (75%) and the market price (92.5%) — simply citing both as corroborating evidence overlooks that 92.5% implies far higher confidence than the unconditional historical frequency, and no forecaster explains what specific 2026 evidence justifies pricing above the historical ceiling. 6. Confidence scores (0.60 and 0.68) seem low relative to the near-identical point probabilities (0.92, 0.90) each forecaster assigns — if they're only ~60-68% confident in their own estimate, this internal inconsistency itself signals unacknowledged uncertainty that isn't reflected in the final probability spread.

90% (64% confidence)

Yes: 90% No: 10%

KMT begins with ~14 of 22 seats, meaning DPP would need to flip roughly 4+ races to overtake — an outcome no analyst, poll, or nomination signal supports, and the fundamentals (failed recalls, Lai at ~35% approval, KMT-TPP joint candidates, Chiang Wan-an's Taipei lead) all point toward retention. The Polymarket anchor at 92.5% and rising is the best available price; the brief's binomial retention model that suggests 20-50% is unreliable because it assumes independent moderate retention rates unsupported by any race-level evidence, whereas actual polling and nominations show no erosion. However, the critique's valid points — no polling in Taichung/Tainan/Kaohsiung/Taoyuan, KMT-TPP Jhubei friction risking vote-splitting in marginal races, and the tie rule resolving against KMT — justify a modest discount below the market rather than tracking it exactly. Final call: 0.90 YES, slightly below the market but well above the naive historical base rate given the unusually favorable 2026 conditions for KMT.

Total pipeline time: 168.7s