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

Resolver and rules

Resolver: Kalshi, off the Bank of Canada's 09:45 ET announcement. Rules (verbatim, both legs): "If the Bank of Canada does a Hike of 0bps at its Oct 28, 2026 meeting, then the market resolves to Yes" / "… a Hike of 25bps …". rules_secondary is empty on both legs. Settlement timer 1,800 s. Close and expected expiration both 2026-10-28T13:29Z (21 days out).

YES condition sentence

H0: For YES to pay, the Bank of Canada must announce an unchanged 2.25% overnight target at its scheduled Oct 28, 2026 fixed announcement date, resolved by the Bank's press release. H25: For YES to pay, the Bank must announce a 25bp increase to 2.50% at that same date. The blank is a new action by an institution at a dated step, so this is an Anchor-shaped market (not a continuous-number thesis).

Precedents consulted (Stage 1)

• KXCBDECISIONCANADA-26JUL-H0 → yes; -26JUL-H25 → no (sixth consecutive hold, July 15). Store lesson: "Two-sided guidance … keeps both middle rungs alive in the narrative while neither gets priced — narrative symmetry is not distributional symmetry, and the BoC ladder pays only if you resist trading the guidance text."
• KXCBDECISIONCANADA-26SEP-H0 → yes; -26SEP-H25 → no (verified this run via public API, finalized 2026-09-02T19:07Z; written back to precedents.json).
• KXFEDDECISION-26JUL-H25 → no: a 25bp hike rung "repriced from ~11% to ~36–38% in nine days still resolved NO, so a hawkish repricing into a live FOMC is a fade candidate rather than a signal — dissent, not a rate move, is how the Committee vents that pressure."
• KXCBDECISIONJAPAN-26JUN15-HOLD → no (the counter-case): when a bank has actually telegraphed a move, "hold legs must be priced off the signalling, not off the bank's history."

Diligence (Stage 2) — vehicle, calendar, already-happened

• Vehicle: the Oct 28 fixed announcement date with a Monetary Policy Report, on the Bank's published schedule (press release, Sep 2: "The next scheduled date for announcing the overnight rate target is October 28, 2026").
• Current stance: 2.25% held for the seventh consecutive time on Sep 2, 2026, "with the economy and inflation evolving broadly as forecast in the July MPR." Headline CPI "hovering around 3% … mainly because of persistently higher gasoline prices"; ex-gasoline 2.2% and core "close to 2%" in July. Guidance: "prepared to adjust monetary policy as needed." (Bank of Canada press release, Sep 2.)
• Deliberations (Sep 2 summary): no hike was discussed; the stated trigger is spillover — "If higher energy prices did spill over into other components of the CPI, members agreed that it could require a monetary policy response."
• Data since: StatCan CPI Aug (Sep 14): 3.0% y/y, unchanged from July; ex-gasoline 2.4% (from 2.2%); gasoline +22.8% y/y; core measure 2.4% from 2.3%. LFS Aug (Sep 4): employment −42k, unemployment 6.4%. GDP July (Sep 29): flat.
• Rhetoric: Macklem, Halifax Partnership, 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"; "if oil prices stay near $100 per barrel, we would expect inflation to edge up in the coming months." Rogers (Oct 1, Victoria) spoke on housing, not on the October call.
• Calendar before the decision: LFS Sep (Oct 9); CPI Sep + Business Outlook Survey + Consumer Expectations survey (Oct 19). Brent fell to $97.97 on Oct 6 from ~$110 weeks earlier as Hormuz flows recovered (CNBC, oilprice.com).
• Already happened? No. Nothing is printed; the decision is three weeks out.

Probability (committed before opening the book)

Pre-market: 75% hold / 23% hike / 2% other. Reasoning: the Bank moves by consensus, did not discuss a hike in September, keys off core (~2%) rather than gasoline-driven headline, faces flat growth, falling employment and escalating US tariffs, and the energy driver has eased since the hawkish speech. Against that, the Halifax speech is the most hawkish BoC language of the cycle, ex-gasoline and core are drifting up, and October is an MPR meeting where a first move is natural.

Entry context (opened after committing)

The hold leg fell from 87¢ (Sep 8) to 57¢ (Sep 23–25) on the Sep 14 CPI and Sep 21 speech, then recovered to 79¢ by Oct 7 as oil slid. Overnight index swaps on Oct 7 imply 72% hold / 28% hike (+7bp priced; bluegamma); a CORRA-spread-adjusted reading puts hike odds lower still. Nanos/Bloomberg (Oct 6): 54% of Canadians expect inflation above 3% within 12 months.

Blend: 75% → 77% final for hold. The book is liquid and specialist-dense (volume24h > 5,000 on H0) and sits within 3 pts of my number; nothing in the primary sources is specific enough to override it.

Resolver risk (Stage 2.5)

Deduction 1 pt: the ladder is mechanically settled off a single primary-source number with two verified identical precedents (Jul, Sep) and no clarifications. No ambiguity on "hike of 0bps" = maintain.

Stage 3 verdict

• H25, NO @ 80¢ (core band). Rejected: continuation tail. The cleanest loss is the Oct 19 CPI showing energy spillover into core — ex-gasoline and core are already drifting up — and the Bank acting on exactly the trigger it named. That is the current trend continuing, not a new discrete event. Secondary: inside spread (78% model vs 80¢ ask; the leg has already moved 10¢ in my direction in 72 hours on the oil pullback, so I would be chasing).
• H0, YES @ 81¢. Side rule: allowed in principle as process-in-motion (dated step on the Bank's official calendar; store shows the resolver pays on the hold). Rejected: inside spread — 77% model vs 81¢ ask is negative edge, and the same continuation tail applies.

What would change this: a hold leg back near 65–70¢ on a hawkish Oct 19 CPI with core still near 2% (the Fed-July-2026 fade shape), or a BoC communication that explicitly tables a hike (the BoJ-June-2026 shape, which would make the hold leg a sell, not a buy).