← back to pipeline
mom
Will Oh Se-hoon win the 2026 Seoul Mayoral Election ↗
will-oh-se-hoon-win-the-2026-seoul-mayoral-election
· 2026 Seoul Mayoral Election Winner · category: Elections, Global Elections, Politics, South Korea, World, World Elections
· resolves
run_id: a47396ff-3434-4ad0-a7a9-4761310d85f3
· started
· status
skipped
· planner_skipped
· ✕ planner rejected
action_not_tradeable: STALL
signal detected
● executed
detect_momentum · UR
direction
UP
current price
10.5%
anchor → trigger
6.5% → 10.5%
cusum @ trigger
5.18
trigger hours
1h
duration
1h
72h range
5.5%—10.5%
(Δ5.0pp)
tick density (6h)
0.67
Price rose 4.0pp from 6.5% to 10.5% over 1h (steady). Now at 10.5%.
news gathered
● executed
news retrieval (tomographic)
raw→filtered: 169 → 7
quality: HIGH
tokens: 6077
gnews_indirect_evidence
relevant
정원오·오세훈 지지율 격차, 두자리에서 한 자릿수로 - v.daum.net
article_thesis
relevant
與 조작기소 특검법 역풍 '보수 결집'...서울·영남 지지율 격차 확 줄었다
article_thesis
relevant
[여론조사 '경향'] 대구·경남은 접전, 서울·부산은 민주당 우위... 국정 지지율은 소폭 하락
article_precursor
direct
오세훈, 여론조사 격차 "서울시민 정책 효과 체감 필요해" | 아주경제
article_precursor
direct
오세훈 "여론조사 격차, 빠른 속도로 줄고 있어"
gnews_indirect_evidence
direct
오세훈 "서울 여론 격차 빠르게 줄어…이 추세라면 할 수 있다" - 뉴스1
gnews_indirect_evidence
relevant
두손 들어 보이는 오세훈 후보 - 연합뉴스
AI verdict
● executed
Prompt-UR · continuation analysis
STALL
conviction: LOW
opportunity: SMALL
action: OMIT
gpt-5 · prompt_2_trend_ur_v6.4.0 · 14768 tokens
Days to resolution: 20 days. Election/leadership market — Theta: NEUTRAL. YES-WORLD: Oh Se-hoon is officially declared the winner of the 2026 Seoul mayoral election on/after June 3; NO-WORLD: any other outcome, including the vote occurring but he is not the winner. For the three ★DIRECT items, two are Oh’s remarks about a narrowing gap and one is a Gallup figure (Jung 46%, Oh 38%); NEAR-MISS detected — all are EVIDENCE-tier-via-NEAR-MISS (related but do not satisfy “wins”), with the remarks being RHETORIC and the poll a credible pre-election indicator. The ★DIRECT news is MIXED ASYMMETRIC: a fresh credible poll still shows Oh trailing (counter), while multiple pieces boost his visibility and frame momentum; the narrative is plateauing after the poll release rather than intensifying. Fresh vs stale: trend_fresh=1 (E9 visibility), trend_stale=0, counter_fresh=1 (D9 credible poll showing trailing), counter_stale=0; Momentum signals: trend=1, counter=1. With equal counts but the higher-quality signal being the poll that shows him behind, the steady +4pp looks fragile, and without a stronger catalyst there’s limited follow-through despite ample headroom (Remaining space: 89.5pp). Hard rules: Election market skip theta; ASYMMETRIC MIXED tilted to counter plus “plateauing” triggers Rule 4 → prefer STALL.
rule audit
● executed
audit_gate
✓
all
trade plan
○ skipped
position_planner
✕ action_not_tradeable: STALL
the deterministic planner rejected this trade. typical causes: poor R:R, spread exceeds gap, settlement danger zone, market too close to resolution. LLM assessment is above (Stage 6); planner is a pure-math gate layered on top.
step post-omit
● executed
news watch (A.2 / A.3)
watch: 3b8b7d9d
status:
⏳ expired
armed: 2026-05-13T09:58Z
closed: 2026-05-14T10:34Z
polls: 6
signals seen: 4180
decision price: 10.5pct