Kalshi Politics — Mispricing Audit (Anchor, NO-side structural)
0 picks30 logged rejects$0 allocated · 100% cash2 precedent rows written
1. Summary
No deployable pick this run. The Stage-0 allowlist of 120 series contained 825 open markets, of which only 30 close inside the 45-day horizon and only 7 clear the mechanical cuts. Two of the seven are already live picks, three sit at 94–96¢ on the NO side (outside the Anchor entry band), and the remaining pair is the Bank of Canada October 28 decision ladder. That ladder is the only in-band candidate, and NO on the 25bp-hike leg fails the continuation-tail rule: the cleanest way it loses is oil staying above $100 and the Bank doing what its Governor pre-announced on Sep 21. It also carries no edge, with my blended fair value for a hike (33%) slightly above the market's 27–31¢. Cash is the position: 100% of capital held.
2. Universe and Stage-0 funnel
Source: /tmp/out/targets.json generated 2026-10-02T16:37Z (min OI 5,000; 120 series, Tier 1 = 73, Tier 2 = 47, no Tier 3). Live data from the Kalshi public API (*_fp/*_dollars fields). Cuts applied in order: close beyond horizon → expiration inside 7 days → close more than 1h after expiration → volume24h < 500 → spread > 5¢ → YES bid ≥ 95¢ → YES ask ≤ 5¢ → Mentions / continuous-number screens.
| Stage | Count | Note |
|---|---|---|
| Open markets in allowlisted series | 825 | Most Tier-1 departure/enactment series run to Dec 31, 2026 or 2027 and fall outside the horizon |
| Close within 45 days (by 2026-11-16) | 30 | None were cut for the 7-day rule, the close-after-expiration rule, or as Mentions markets |
| Cut: volume24h < 500 | 20 | Includes KXRETIREPELOSI-26, KXDEMSCHUMER-27-NOV03, KXDJTATTACKWARSH, Anthropic OCT10/OCT17 rungs |
| Cut: YES ask ≤ 5¢ | 3 | KXRETIREMM-26, KXUSAIRANAGREEMENT-27-26NOV, KXIPOANTHROPIC-DATE-26OCT24 |
| Passed to Stage 1–2 | 7 | Patel exit, Hegseth exit, Tesla/SpaceX merger, Anthropic NOV01 (active), diesel ban (active), BoC Oct H0 and H25 |
Specialist-dense flag (volume24h > 5,000): KXIPOANTHROPIC-DATE-26NOV01 only. No market in the allowlist was researched outside the universe.
3. Picks and position sizes
None. Position-size table, EV-for-the-record and cluster table are empty by construction. Reserve held: $2,500 of $2,500 (≥10% rule satisfied trivially).
| Ticker | Side | Limit | Contracts | Cost | Conviction | Cluster |
|---|---|---|---|---|---|---|
| No positions. Cash: $2,500 (100%). | ||||||
4. Candidate deep-dive — the one in-band market
KXCBDECISIONCANADA-26OCT-H25 — Bank of Canada hikes 25bps on Oct 28, 2026?
archetype: scheduled-processmechanism considered: status-quoside considered: NOverdict: REJECT — continuation tail
YES condition: For YES to pay, the Bank of Canada's Governing Council must announce a 25bp increase in the target for the overnight rate (2.25% → 2.50%) at the Oct 28, 2026 fixed announcement date, resolved by Kalshi off the Bank's press release. This is a new action by an institution, so the market is Anchor-shaped and was carried to full diligence.
Rules: primary "If the Bank of Canada does a Hike of 25bps at its Oct 28, 2026 meeting, then the market resolves to Yes." No rules_secondary clarification. Event is mutually exclusive across C26/C25/H0/H25/H26. Note the series naming trap flagged in the precedent store: the hold leg is titled "Hike rates by 0bps".
Precedents consulted: KXCBDECISIONCANADA-26JUL-H0/H25/H26/C25/C26 (all July 15, 2026; hold paid, every other rung NO: "two-sided guidance keeps both middle rungs alive in the narrative while neither gets priced"); KXFEDDECISION-26JUL-H25/H0 (July 29, 2026: a 25bp-hike leg repriced from ~11% to ~36–38% in nine days still resolved NO, "a hawkish repricing into a live FOMC is a fade candidate"); KXCBDECISIONJAPAN-26JUN15-H25 (YES: "moves arrive telegraphed through speeches before they are priced"). The store had no Bank of Canada row later than July; the September ladder is written back this run.
Primary-source diligence:
- Vehicle: the Oct 28, 2026 fixed announcement date with a Monetary Policy Report, on the Bank's published 2026 schedule. Sep 2 press release: rate held at 2.25% for a seventh straight decision; "upside risks to the Bank's inflation forecast have increased"; "prepared to adjust monetary policy as needed." (bankofcanada.ca, Sep 2, 2026)
- Deliberations (published Sep 16): members did not discuss raising the rate, but agreed that "if higher energy prices did spill over into other components of the CPI ... it could require a monetary policy response." Core inflation "remained around 2%", CPI ex-gasoline 2.2%.
- Governor's speech, Halifax, Sep 21: "We don't want to raise our policy rate and restrain growth if inflationary pressures are contained. But nor do we want to be too slow to respond if inflationary pressures are becoming more persistent." and "If oil prices stay near $100 per barrel, we would expect inflation to edge up in the coming months." (bankofcanada.ca, "Navigating uncertainty and adapting to change")
- Data: August CPI 3.0% y/y, gasoline +22.8%, ex-gasoline 2.4%, CPI-median 2.0%, CPI-common 2.6% (Statistics Canada, Sep 14). GDP flat in July, +0.2% in August. September CPI lands around Oct 20 and the Labour Force Survey Oct 9, both before the meeting.
- External: the Federal Reserve hiked 25bp to 3.75–4.00% on Sep 16, 2026, 12–0, with 16 of 18 participants projecting one more hike this year (CNBC). Brent closed $102.31 on Oct 1, +4.4%, as the US reportedly sent a third carrier to the Middle East (CNBC). National Bank and Scotiabank forecast an October hike to 2.50%; BMO, CIBC, RBC and TD forecast a hold through December (TechTimes, Aug 31). OIS pricing ranged from 26% to 54% hike across sources Oct 1–2.
- Already happened? No. Nothing to print.
Probability: pre-market 35% hike → final 33% after seeing the book (market 27–31¢, OIS ~26%). Fair value for NO ≈ 67¢ versus a NO ask of 73¢. The hold leg (H0) is the mirror: 66–67% fair versus a 69¢ ask.
Entry context (recorded, not traded): YES bid 27¢ / ask 31¢, spread 4¢; NO bids 69¢ ×172, 67¢ ×85, 66¢ ×4,800; YES asks 31¢ ×~50 then 24¢ ×750 below; volume24h 1,729; OI 31,363. Price history: 22¢ Sep 11 → 32–36¢ Sep 14–16 → 41–46¢ Sep 21–24 → 35¢ Oct 1 → 27–31¢ Oct 2. 48h move: −8 points on the hike leg with no Canadian data release behind it; that alone would trigger the v4 48h check.
Resolver profile: Kalshi settles off the Bank of Canada press release; a 25bp move is unambiguous; no announce-vs-effective gap; no inter-meeting risk the ladder would mis-handle (the >25bp rung covers an outsized move). Resolver deductions: 0–1 pt.
Cleanest tail for NO: oil holds above $100 as the Middle East escalation continues, the Sep 21 "too slow" framing is operationalised, September CPI shows spillover into core, and the Bank hikes on Oct 28. That is the current trend continuing, not a new discrete event — the Governor has already stated the reaction function and two of six big banks already forecast the move. Under Anchor this is a continuation tail reject regardless of the apparent fade opportunity. The July 2026 FOMC fade precedent is not a clean analogue: there the Fed had not moved and the Chair had not pre-committed language; here the Fed has hiked and the Governor has.
Side rule on the H0 leg: YES on "maintains rate" is not an allowed YES shape (a hold is the status quo, not an already-happened event nor a process step this resolver pays on), so it is logged as yes-side thesis and as the cluster runner-up.
5. Logged rejects — Stage 2 survivors
Every market that reached diligence, with the side Anchor would have taken, the effective price for that side, and the final YES probability where one was formed.
| Ticker | Archetype | Side | Price | P(YES) | Reject reason |
|---|---|---|---|---|---|
KXCBDECISIONCANADA-26OCT-H25 | scheduled-process | NO | 73¢ | 33% | continuation tail: the cleanest loss is oil staying >$100, escalation continuing and the Bank hiking on Oct 28 — the current trend continuing, not a new discrete event; also no edge (NO ask 73¢ vs 67¢ fair: pre-market 35% hike → 33% final; market 27–31) |
KXCBDECISIONCANADA-26OCT-H0 | scheduled-process | YES | 69¢ | 66% | yes-side thesis: a hold is the status quo, not an already-happened event or a dated process step the resolver pays on; same cluster as the H25 leg (cluster runner-up); no edge at 69¢ vs 66–67% fair |
KXKASHOUT-26JUL-26NOV01 | departure/ouster | NO | 96¢ | 7% | deep-favorite band: NO effective entry 96¢ (YES 4/6). Status quo lean (precedent KXKASHOUT-26APR-JUN01/JUL01: resignation calls and anonymous sourcing never resolved YES) but max win 4% and the 90¢+ cohort showed no edge |
KXHEGSETHOUT-26AUG-26NOV01 | multi-news-other | NO | 95¢ | 4% | deep-favorite band: NO effective entry 95¢ (YES 5/6). Precedent KXHEGSETHOUT-26APR-JUL01: sitting defence chief in an active war is a structural NO, and he gave a 'State of the Force' address Sep 30 — but ≥90¢ is outside the Anchor band |
KXCOMPANYACTIONMERGER-27-26NOV01 | corporate-event | NO | 94¢ | 6% | deep-favorite band: NO effective entry 94¢ (YES 6/7). No vehicle: no definitive agreement, no 2026 annual meeting scheduled (Electrek, Sep 29), rules require a signed agreement not Musk hints — but ≥90¢ is outside the Anchor band |
KXIPOANTHROPIC-DATE-26NOV01 | corporate-event | NO | 96¢ | — | already an active pick |
KXDIESELEXPORTBAN-26 | multi-news-other | NO | 87¢ | — | already an active pick |
Notes on the deep-favorite rejects
- Patel (KXKASHOUT-26JUL-26NOV01). Status-quo lean. Co-Deputy Director Andrew Bailey resigned Sep 28 (CNN, Forbes) and Rep. Massie and Sen. Schumer have called for Patel's resignation; precedent rows KXKASHOUT-26APR-JUN01 and -JUL01 record that calls and anonymous sourcing never resolved YES and the far rungs were systematically overpriced. P(YES) 7%. NO at 96¢ is a ≥90¢ band reject.
- Hegseth (KXHEGSETHOUT-26AUG-26NOV01). Status-quo lean; he delivered a "State of the Force" address at Quantico Sep 30 announcing a drone command and general-officer cuts (NBC, Foreign Policy). Precedent: sitting defence chief in an active war is a structural NO. P(YES) 4%. NO at 95¢ is a band reject.
- Tesla/SpaceX (KXCOMPANYACTIONMERGER-27-26NOV01). No vehicle: no definitive agreement, Tesla has not scheduled a 2026 annual meeting, and the rules require "an unambiguous, affirmative confirmation of a definitive signed agreement" (Electrek, Sep 29; rules_secondary). P(YES) 6%. NO at 94¢ is a band reject.
6. Logged rejects — Stage-0 mechanical cuts
| Ticker | Archetype | NO price | Close | Reason |
|---|---|---|---|---|
KXRETIREMM-26 | departure/ouster | 98¢ (NO) | 2026-11-03 | Stage-0 cut: YES ask 3.2¢ ≤ 5¢ (NO side would be deep-favorite band) |
KXRETIREPELOSI-26 | departure/ouster | 98¢ (NO) | 2026-11-03 | Stage-0 cut: volume24h 250 < 500 |
KXLEAVEPOWELLGOV-26NOV01 | departure/ouster | 98¢ (NO) | 2026-11-01 | Stage-0 cut: volume24h 0 < 500 |
KXPROMISEACT-26AUG-26OCT03 | confirmation/enact | 100¢ (NO) | 2026-10-03 | Stage-0 cut: volume24h 0 < 500 |
KXARRESTNETANYAHU-28JAN01-26NOV01 | legal-action | 98¢ (NO) | 2026-11-01 | Stage-0 cut: volume24h 56 < 500 |
KXDEMSCHUMER-27-NOV03 | departure/ouster | 89¢ (NO) | 2026-11-03 | Stage-0 cut: volume24h 60 < 500 |
KXGALLEGOOUT-26APR-NOV03 | departure/ouster | 97¢ (NO) | 2026-11-03 | Stage-0 cut: volume24h 0 < 500 |
KXLEAVEDUNN | departure/ouster | 98¢ (NO) | 2026-11-03 | Stage-0 cut: volume24h 10 < 500 |
KXNETANYAHUPARDON-26-NOV01 | legal-action | 99¢ (NO) | 2026-11-01 | Stage-0 cut: volume24h 0 < 500 |
KXLEAVEMERZ-29JAN01-26NOV01 | departure/ouster | 98¢ (NO) | 2026-11-01 | Stage-0 cut: volume24h 0 < 500 |
KXMARTINDNCOUT-26MAY-26NOV01 | departure/ouster | 98¢ (NO) | 2026-11-01 | Stage-0 cut: volume24h 3 < 500 |
KXUSAIRANAGREEMENT-27-26NOV | scheduled-process | 98¢ (NO) | 2026-11-01 | Stage-0 cut: YES ask 4.0¢ ≤ 5¢ (NO side would be deep-favorite band) |
KXIPOANTHROPIC-DATE-26OCT24 | corporate-event | 100¢ (NO) | 2026-10-24 | Stage-0 cut: YES ask 1.0¢ ≤ 5¢ (NO side would be deep-favorite band) |
KXIPOANTHROPIC-DATE-26OCT17 | corporate-event | 100¢ (NO) | 2026-10-17 | Stage-0 cut: volume24h 200 < 500 |
KXIPOANTHROPIC-DATE-26OCT10 | corporate-event | 100¢ (NO) | 2026-10-10 | Stage-0 cut: volume24h 103 < 500 |
KXDJTATTACKWARSH-26SEP-26NOV01 | multi-news-other | 92¢ (NO) | 2026-11-01 | Stage-0 cut: volume24h 147 < 500 |
KXIPOOPENAI-26NOV01 | corporate-event | 100¢ (NO) | 2026-11-01 | Stage-0 cut: volume24h 0 < 500 |
KXIPOSTARLINK-26NOV01 | corporate-event | 100¢ (NO) | 2026-11-01 | Stage-0 cut: volume24h 0 < 500 |
KXCBDECISIONCANADA-26OCT-H26 | scheduled-process | 100¢ (NO) | 2026-10-28 | Stage-0 cut: volume24h 100 < 500 |
KXCBDECISIONCANADA-26OCT-C26 | scheduled-process | 100¢ (NO) | 2026-10-28 | Stage-0 cut: volume24h 0 < 500 |
KXCBDECISIONCANADA-26OCT-C25 | scheduled-process | 100¢ (NO) | 2026-10-28 | Stage-0 cut: volume24h 0 < 500 |
KXMILLEROUT-26JUL-26NOV03 | multi-news-other | 99¢ (NO) | 2026-11-03 | Stage-0 cut: volume24h 3 < 500 |
KXIPOANDURIL-26NOV01 | corporate-event | 100¢ (NO) | 2026-11-01 | Stage-0 cut: volume24h 0 < 500 |
7. Risk profile and execution notes
- Portfolio: 100% cash. No cluster exposure, no correlated legs, no fills to manage.
- Why zero is the right answer here: the Anchor record lives in 60–89¢ NO entries on event markets with new-event tails. This week's horizon contained exactly one market in that band and its tail is a trend. Deploying into 94–96¢ NO legs would put 4–6% max win against the 90¢+ cohort's zero historical edge.
- What would change the picture: (a) the Bank of Canada hike leg falling to ≤20¢ with the Governor's language unchanged would make NO-on-hike a fair-value trade but would still be a continuation-tail reject; (b) new Tier-1 rungs opening with Dec 2026 or Jan 2027 closes inside a longer horizon; (c) the Patel or Hegseth ladders repricing YES into the 15–40¢ range on a news spike, which the precedent store says is the fade.
- Public-info check: all facts above are from official publications or major outlets dated Sep 2–Oct 2, 2026; nothing here is non-public.
8. Precedent writeback
Two rows written to /tmp/out/precedents.json for the settled September 2 ladder: KXCBDECISIONCANADA-26SEP-H0 (YES) and -H25 (NO), verified against the Bank of Canada press release, the Sep 16 summary of deliberations and Statistics Canada's August CPI. Lesson recorded: a hold delivered with explicit upside-risk language is where the following meeting's ladder goes live, and the hike leg must be priced off core and ex-gasoline prints, not headline CPI.
9. Sources
- Bank of Canada, press release Sep 2, 2026 — bankofcanada.ca/2026/09/fad-press-release-2026-09-02/
- Bank of Canada, Summary of Governing Council deliberations, published Sep 16, 2026
- Bank of Canada, Governor Macklem, "Navigating uncertainty and adapting to change", Halifax, Sep 21, 2026
- Statistics Canada, The Daily, Consumer Price Index August 2026, Sep 14, 2026
- CNBC, Fed rate decision Sep 16, 2026; CNBC, oil prices Oct 1, 2026
- TechTimes Aug 31, 2026 (bank forecasts); Canadian Mortgage Trends Sep 21, 2026; money.ca / bluegamma OIS pricing Oct 1–2, 2026
- CNN and Forbes, Sep 28, 2026 (Bailey resignation); NBC News and Foreign Policy, Sep 30–Oct 1, 2026 (Hegseth address); Electrek, Sep 29, 2026 (Tesla retail voting program)
- Kalshi public API: markets, orderbooks and candlesticks fetched 2026-10-02 ~17:00Z; resolution_precedents store (162 rows)