Kalshi Politics — Mispricing Audit (Keel v4, archetype-targeted)

Prob: 72% pre-market → 79% final (blended toward a liquid, specialist-dense book: vol24h 29,263, OI ~30,000). Path priced: committee vote Tue Aug 4 (~88%) × floor cloture + 30h Cabinet post-cloture before senators leave (~80%) ≈ 72%; the market's 86¢ implies near-certainty on both legs of a week that also has to fit a CR and a budget resolution. Resolver risks priced: withdrawal-resolves-NO clause (Trump floated a pull on Jul 30, −3pts) and "first person confirmed" wording (−1pt) are inside the deduction cap but real.

Why rejected: the 48h price-move screen. This book crashed 57¢→5¢ on the Jul 29 postponement and ripped 20¢→87¢ this morning after the Aug 2 fund rescission — the news is fully captured, and at 86¢ ask my blended 79% is negative edge. The NO side (~14¢ effective) sits below the 40¢ band (tails 0-for-8 lifetime).
Entry context: 85.0 / 86.0, depth 35×46, vol24h 29,263, OI 29,999.
Precedents consulted: KXAGCONF-26-JUN01, -JUN15, -JUL01 all NO ("price the clock from formal transmittal, 5–6 weeks is the floor"); KXAGCONF-26-AUG01 resolved NO on Aug 1 (verified this run — committee postponement inside 10 days of a rung deadline is mechanically fatal).

Prob: 60% pre-market → 50% final. For: 4 confirmable judges sit on the calendar, three are district seats (2h post-cloture each), and leaving lifetime judges hanging five weeks before a recess is historically rare — the pre-recess judge stack is the norm. Against: the same "rush" was predicted for the week of Jul 27 and fizzled (floor went to Russia sanctions + a 74-nominee en-bloc of executive nominees), no cloture is on file as of the Aug 3 calendar, and this week's queue (CR, budget resolution, Blanche's 30-hour clock) outranks judges. RECNCH precedent lessons cut against schedule optimism: "count actual session days," "the rung after upstream progress is the classic trap."

Why rejected: the entry band. YES at 31¢ is below the 40¢ floor — a hard v4 rule (sub-40¢ tails are 0-for-8 lifetime) — regardless of the apparent +19pt edge; NO at 69–71¢ is negative edge at a blended 50%. Note the "slide" from 71¢→31¢ happened on ~3 traded contracts with 5 contracts at the ask: this is a two-day-old book being quote-walked, not informed flow, which is exactly the phantom-edge shape v4 exists to avoid.
Entry context: 29.0 / 31.0, depth 127×5, vol24h 1,776, OI 2,470.
Precedents consulted: none for KXJUDGECOUNT (cold derivation); KXRECNCH ladder lessons applied.

Prob: ~30% final. Requires the full stack — all four pending nominees including a circuit seat carrying a 30-hour post-cloture clock — inside the same jammed four-day window.

Why rejected: both sides sit in banned bands (YES 6–10¢ < 40¢; NO 90–94¢ ≥ 90¢ deep-favorite memorial band), and it is the cluster runner-up to KXJUDGECOUNT-26AUG-1 — one leg per cluster; this leg shadow-tracks.
Entry context: 6.0 / 10.0, depth 4×5, vol24h 762, OI 816.