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

Anthropic confirms an IPO (S-1 effective, priced, or exchange ticker assigned) before Oct 17, 2026?

Thesis. Reuters (Sep 4) moved Anthropic's public S-1 to late September and the roadshow to mid-October at the earliest; the SEC 15-day rule plus a roadshow puts pricing after Oct 17, so YES at 29-30c is stale post-headline residue.

Prob: 90% pre-market → 86% final (NO side), blended toward a 1,150-contract/day book that carries a 7,740-contract resting YES bid at 28¢. Limit 71¢ (NO ask; 533 contracts resting at that level, 7,740 more at 72¢). Edge 15¢ · EV 21% for the record.

Rules. Primary: "If Anthropic confirms an IPO before Oct 17, 2026, then the market resolves to Yes." Secondary (live, no dated clarifications appended): confirmed if (1) the SEC declares the S-1 effective OR (2) the IPO is priced OR (3) a securities exchange has assigned a ticker; resolves immediately on any of them.

Diligence (primary sources). EDGAR full-text and company searches on Sep 9, 2026 return no Anthropic S-1 or S-1/A (only third-party SPV funds named "Anthropic"). Anthropic's own Jun 1, 2026 release described a confidential draft S-1 with no share count or price. Reuters (Sep 4, 2026, via Investing.com/CNBC): the public prospectus "had been expected … as early as next week … now that is not expected until late September"; Anthropic "is expected to begin marketing its initial public offering in mid-October at the earliest and complete the listing days before the U.S. midterm elections" (Nov 3), pending a $15bn revolving credit facility. Under the SEC's rules for confidential filers the public S-1 must be on file at least 15 days before the roadshow; the SpaceX precedent ran 22 days from public S-1 to pricing. A late-September public filing therefore puts the earliest pricing around Oct 20; the Oct 17 03:59Z cutoff sits inside that window. Cross-venue: Polymarket prices "Anthropic shares trading by Oct 15" at 7% and "by Oct 31" at 59%.

Precedents consulted. KXIPOANTHROPIC-DATE-26JUL01 resolved NO — the Jun 1 confidential-S-1 announcement was none of the three triggers (store lesson: "treat 'filed' and 'confirmed' as separated by months of process"). KXIPOANTHROPIC-DATE-26JUN01 resolved NO — "officially announce" is mechanical, not journalistic. KXIPO-26-SPACEX: Kalshi did NOT resolve on the public S-1's proposed symbol "SPCX" and settled only after effectiveness/pricing on Jun 11 (store lesson: a proposed symbol is not "a securities exchange has assigned a ticker"). Verified this run against the live rules text: the three-trigger language is unchanged.

Resolver profile and deductions. Resolver: Kalshi, mechanical three-trigger test, historically strict (3-for-3 in the store). Deductions: −4 timeline re-acceleration (credit facility closes fast, public S-1 lands mid-September, one-week roadshow); −3 ticker-assignment ambiguity (a Nasdaq pre-listing announcement could be argued as "assigned" — the SpaceX ruling cuts the other way); −3 single-outlet sourcing on the Sep 4 timeline. Total 10 pts (< 15 → keep).

Cleanest tail risk. Anthropic publishes the S-1 within days and compresses the roadshow to price before Oct 16 — a total reversal of the reported plan.

Entry context. YES 29 / 30 · NO ask 71 (533) then 72 (7,740) · vol24h 1,150 · OI 2,799 · 14-day path: YES 69¢ (Sep 5) → 35¢ (Sep 6) → 29¢ (Sep 7) → 29–30¢ since; the move is fully explained by the Sep 4 report and is outside the 48h window.

Sizing. MEDIUM cap 7% = $175. The live Keel leg KXIPOANTHROPIC-DATE-26OCT01 NO @84 (75 contracts, $63) shares this cluster, so the cluster budget for a new leg is $112 → 157 contracts @ 71¢ = $111.47. Depth cap 1/3 × 533 = 177 ≥ 157. Note the Anchor strategy also opened NO @71 on this same leg on Sep 8; that is a separate ledger, but members running both should not double up.

Through the Sep 1, 2026 03:59Z cutoff Anthropic had made no public IPO filing: EDGAR full-text and company searches on Sep 9, 2026 return no Anthropic S-1/S-1A (only third-party 'Anthropic'-named SPV funds), so none of the three triggers (S-1 effectiveness, pricing, exchange ticker) could have occurred. Kalshi settled NO at 13:50Z on Sep 1. Reuters (Sep 4) then reported the public prospectus had slipped from the week of Sep 7 to late September, with marketing in mid-October at the earliest and a listing days before the Nov 3 midterms, pending a $15bn revolving credit facility.

Lesson: Clock this ladder off the PUBLIC S-1 date: the SEC's 15-day public-filing-before-roadshow rule plus a roadshow means pricing is >= ~3 weeks after the public filing, so any rung whose cutoff sits inside 'public S-1 date + 3 weeks' is a mechanical NO, and reported slippage of the public filing pushes every near rung down together.

Edge notes: The Sep 4 Reuters exclusive repriced the OCT17 rung from 69c to 29c across Sep 5-7 (1.1k+ contracts), but the residual 29-30c YES on Sep 9 still sat far above Polymarket's 7% for 'trading by Oct 15' — the rung six days later (OCT24) traded 54-58c vs Polymarket's 59% for 'by Oct 31'. Post-headline residue on the near rung, not the far one, was the dislocation.

The July 28-29, 2026 FOMC held at 3.50-3.75% on a 9-3 vote; Cleveland's Beth Hammack, Minneapolis' Neel Kashkari and Dallas' Lorie Logan each preferred a 25bp hike. The 'exactly 3' strike settled YES at the 18:00Z Jul 29 expiration. The same three presidents had dissented (against the statement's easing bias) at the April 28-29 meeting, where Stephen Miran added a dovish cut dissent for a 4-dissent total; June was 12-0.

Lesson: Dissent count is a discrete, whip-countable institutional outcome, not a continuous number: an established dissenting bloc that has already voted together (Hammack/Kashkari/Logan) is the base case for the next hold, and the ladder's modal strike should be priced off that roster rather than off the rate decision alone.

Edge notes: Kalshi candlesticks show the 3-strike traded 13-15c for the final eleven days and 10-11c on Jul 28-29 (7.7k and 4.5k contracts) before settling at 100, while the 0-strike held 21-35c into the decision — the crowd priced a repeat of the April/June split as a tail even though the same three hawks had dissented three months earlier and hike odds were ~36%.

Settled NO on Jul 29, 2026: the FOMC's hold carried three hawkish dissents (Hammack, Kashkari, Logan), so the unanimous-vote strike failed. It was the ladder's most-traded leg (182k contracts lifetime) and closed at 17c.

Lesson: A unanimous vote is the tail, not the base case, whenever the prior meeting produced named dissenters who have not publicly changed position — treat 'exactly 0' as a NO-side carry leg and check the current voter roster (2026: Warsh, Powell, Jefferson, Barr, Bowman, Cook, Waller, Williams, Hammack, Kashkari, Logan, Paulson) before pricing it.

Edge notes: The 0-strike traded 32-35c on Jul 18-22 and still 21-27c on Jul 27-29 with 8-16k contracts a day; NO at 65-79c was the ladder's cleanest carry and paid in full.