Kalshi Politics — Mispricing Audit (Anchor, NO-side structural)

Zero picks. Cash is the position this week. The Stage-0 allowlist (120 rows, 114 distinct series) holds 859 open markets, but only 70 of them close inside the 45-day horizon, and the Anchor mechanical cuts leave two Stage-2 candidates. Both were researched to a probability and both fail the entry band: the Tesla/SpaceX combination leg (NO ask 91¢, deep-favorite band, zero passive depth) and the Bank of Canada October hold (57¢, below the 60¢ floor on either side). No YES thesis of the two permitted shapes (already-happened, process-in-motion) exists in the horizon.

The remaining within-horizon volume sits in markets the rules exclude mechanically: sub-5¢ favourites (McConnell, Pelosi, Iran deal, Anthropic IPO rungs), sub-week expiries (Oct 1 rungs across Hegseth, Kash Patel, Iran, Mamdani, the California Jewish-ethnicity bill), Mentions-type speech markets (Trump criticising Warsh), and markets whose close_time sits more than an hour after expected_expiration_time (the November ballot measures and the impeachment ladder).

Existing pick KXCRYPTOSTRUCTURE-26JAN-OCT (NO) is not in a target series this run and was not re-examined; no pick or reject collides with it.

0 picks No market survived Stage 3.5. Under Anchor a zero-pick run is a normal, correct result; the band and depth rules are doing the work the Sep 2026 study asked them to do (no ≥90¢ favourites, no coin-flip lane).

corporate-eventno-vehicle Tesla / SpaceX definitive combination agreement before Nov 1, 2026 — KXCOMPANYACTIONMERGER-27-26NOV01

YES condition: For YES to pay, Tesla or SpaceX must officially announce (press release, 8-K, earnings call, verified account) a definitive, binding agreement to combine the two companies before Nov 1, 2026, resolved by Kalshi off official company channels; per rules_secondary a Musk statement counts only as “an unambiguous, affirmative confirmation of a definitive signed agreement — not an expression of intent”.

Thesis shape: Anchor-shaped (a new action by an institution by a deadline). Side: NO (status quo / no vehicle).

Diligence. Vehicle: none. Tesla's EDGAR 8-K list ends at 2026-07-22 (items 2.02/9.01, earnings); SpaceX's ends at 2026-08-14 (items 2.01/3.02/9.01, the Cursor merger closing). No 1.01 “entry into a material definitive agreement” from either company references the other. Calendar: the only official channel event inside the window is Tesla's Q3 earnings call (reported as Oct 28, 2026, after close). Already-happened: no — Musk's All-In Summit remark (September 2026: “who can imagine what action one might take when there's so much close collaboration”) is intent, which the rules exclude; Polymarket's Dec-31 announcement contract was 18% on Aug 5 (Yahoo Finance).

Prob: 5% pre-market → 6% final. Book: YES 9/11, last 10¢, vol24h 5,651, OI 97,046. NO ask 91¢ with 134 contracts; next 92¢ (209), 93¢ (2,467). A passive NO rest at 89¢ would join 23 contracts already bid there and has zero NO offers at or below it, so the 1/3-depth cap sizes the order at 0.

Cleanest tail: a surprise 8-K or a Q3-call confirmation of a signed agreement. That is a new discrete event, an acceptable shape, but the pick fails before shape matters: deep-favorite band (≥90¢; max win ≤ 11% of risk and the 90¢+ cohort showed no edge). Consistent with the Anchor rejects of Sep 21 and Sep 24.

Precedents consulted: no rows for KXCOMPANYACTIONMERGER; nearest archetype rows are KXIPOANTHROPIC JUN01/JUL01 (NO) and KXIPO-26-SPACEX, whose lesson — “filing-adjacent headlines move the price without touching the resolution trigger” — is the same dynamic seen here on Aug 14–20 (YES 10–16¢ on merger chatter) and Sep 13 (12–16¢), each reverting to 8–10¢.

scheduled-processstatus-quo Bank of Canada holds at the Oct 28, 2026 meeting — KXCBDECISIONCANADA-26OCT-H0

YES condition: For YES to pay, the Bank of Canada must leave the policy rate at 2.25% at its scheduled Oct 28, 2026 announcement, resolved by Kalshi off the Bank's decision.

Thesis shape: a scheduled decision by an institution (Anchor-shaped), but the natural side is a hold-YES, which is a status-quo forecast on the YES side — not one of the two permitted YES shapes. The Anchor expression would be NO on the 25bp-hike leg, which failed Stage 0 on volume (366 < 500) and prices at 55¢, below the band.

Diligence. Vehicle: the Oct 28 fixed announcement date and Monetary Policy Report (official BoC calendar). Calendar: September jobs (~Oct 9), Business Outlook Survey (~Oct 19), September CPI (~Oct 20) all land before the decision. News: BNN Bloomberg, Sep 18, 2026 — markets see the meeting “as a coin flip, narrowly in favour of a hike”; policy rate 2.25% after the Sep 2 hold; Desjardins (Bartlett) expects no move until Q1 2027; Capital Economics (Brown): “base case is that the bank will not hike in October, though it will likely be a close call”; the Fed hiked the same week.

Prob: 55% pre-market → 55% final. Book: YES 56/57, vol24h 555, OI 29,184; H0 has slid from 90¢ (Sep 7) to 57¢ (Sep 23–25) on the hawkish Sep 2 statement and oil. Precedent store: June/July H0 legs resolved YES at consensus; the Sep 24 note records that “the hold-leg carry disappears the moment guidance turns conditional”.

below band YES 57¢ < 60¢ and NO 44¢ < 60¢. Nothing to do here before the Oct 20 CPI print; if the ladder re-forms with a hike leg ≥60¢ NO, that leg would be the Anchor candidate at the next run.

Precedents consulted: KXCBDECISIONCANADA-26JUL-H0 (YES), -26JUL-H25 (NO), -26JUN-H0 (YES), -26JUN-H25 (NO), plus the un-resolved -26SEP-H0/H25 notes of Sep 24.

• Capital at risk: $0 of $2,500. Max drawdown this cycle: $0.
• Correlation: none (no legs).
• Specialist-dense books flagged (vol24h > 5,000): Iran-deal OCT rung (71k), CA wealth tax (32.6k), Iran-deal NOV rung (22k), McConnell (6.8k), Tesla/SpaceX NOV01 (5.7k). All excluded on mechanical grounds, not on edge.

• targets.json generated 2026-09-25T16:37Z, min OI 5,000: 120 rows, 114 distinct series (six duplicated across categories). Live pull: 859 open markets; 70 close within 45 days.
• Cuts inside the horizon: 25 inside 7 days, 9 close-after-expiration, ~45 below the 500 volume floor, 3+ sub-5¢ asks, 2 Mentions-type. Two candidates reached Stage 2.
• Two API quirks worth recording for the target-scan: (a) dated “before <date>” rungs often carry the parent event's expected_expiration_time (e.g. Iran-deal OCT rung closes Oct 1 but shows Jan 2027 expiry), so the 7-day rule was applied to the earlier of close and expiration; (b) the impeachment ladder and the Nov 3 ballot measures show an expiration earlier than close, which the “close after expiration” rule catches — the ballot measures (CA wealth tax at 31/32, 32.6k vol24h) are the highest-volume market this rule removed.
• No Tier 3 rows exist in the file; no fallback scan was run.

• Nothing to execute. Do not rest a NO bid at 89¢ on the Tesla/SpaceX leg: it sits behind 23 contracts with no offers below 91¢, and a fill would only come from a YES buyer paying up into a rising market — exactly the adverse-selection profile the band rule exists to avoid.
• Watchlist for the next run: BoC Oct H25 leg (NO) if volume clears 500 and NO ≥ 60¢ after the Oct 9 jobs or Oct 20 CPI print; any Nov-dated departure/ouster rung that lifts to a 60–84¢ NO with volume.

• DB query API (/schema read first): resolution_precedents (162 rows; 12 KXCBDECISIONCANADA, 4 corporate-event), signals (prior Anchor/Keel rejects on both candidates).
• Kalshi public API: /markets?series_ticker=…&status=open for all 114 series; /markets/{ticker}, /markets/{ticker}/orderbook, daily candlesticks for the two candidates (all *_fp/*_dollars fields).
• Primary sources: SEC EDGAR 8-K lists for Tesla (CIK 1318605, latest 2026-07-22) and SpaceX (CIK 1181412, latest 2026-08-14); BNN Bloomberg, Sep 18, 2026 (BoC hike odds); Motley Fool, Sep 24, 2026 (All-In Summit quote); Yahoo Finance, Aug 5, 2026 (Polymarket 18% Dec-31 contract); earnings-calendar aggregators for the Tesla Q3 date (Oct 28, 2026).
• Kalshi rules_primary/rules_secondary quoted verbatim for both candidates.