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

1 pick, 61 logged rejects, $171 max deployed (6.8% of capital), 93.2% cash.

The 45-day window over the allowlist produced 62 open markets. The mechanical cuts were brutal and correct: 50 markets are penny books or dead volume (most Tier-1 departure/legal series trade at 0–3¢ YES — the NO side is a ≥95¢ deep favorite with no edge left), the ECB cluster is inside 7 days, and the Banxico/Ebola books fail spread or volume floors. Exactly two markets survived to Stage 2, both strikes on the same event: KXIPOANTHROPIC-DATE (Oct 17 and Oct 24). The Oct 17 leg becomes the run's single pick — NO at ~69¢ against a mechanical resolution bar the calendar can't reach on the reported schedule. The Oct 24 leg is a below-band reject. Zero-to-three picks is the expected Anchor outcome; this is a one-pick week.

“When will Anthropic officially announce an IPO?” — Before Oct 17, 2026 · closes 2026-10-17 03:59Z · archetype: corporate-event · cluster: anthropic-ipo-confirm

YES condition sentence

For YES to pay, the SEC must declare Anthropic's Form S-1 effective, OR the IPO must be priced, OR a securities exchange must assign it a ticker, by Oct 17, 2026, resolved by Kalshi against those three mechanical triggers (rules_secondary, verbatim). A public filing does not count; a confidential filing does not count; “sources say October” does not count.

The calendar case

• No vehicle inside the strike. No public S-1 exists on EDGAR as of Sep 8, 2026 (verified directly against the SEC company search this run). The only filed document is the confidential draft S-1, submitted Jun 1, 2026 (Anthropic press release).
• The freshest primary-adjacent reporting puts every step after the strike. Reuters (via CNBC, Sep 5, 2026): the public flip has slipped to late September, the investor roadshow to mid-October at the earliest, and the listing to late October — “just days before the November midterm elections,” contingent on a pending $15B credit facility.
• A statutory clock sits between flip and pricing. Under the JOBS Act, a confidential submission must be publicly filed at least 15 days before the roadshow. Flip in late September ⇒ roadshow no earlier than ~Oct 10–15 ⇒ pricing (where effectiveness and ticker assignment historically land) in the Oct 20+ window. For YES before Oct 17, Anthropic must flip within ~10 days of today and compress the roadshow — two accelerations against the reported schedule.

Precedents consulted

KXIPOANTHROPIC-DATE-26JUN01 and -26JUL01 both resolved NO: “only S-1 effectiveness, a priced offering, or an assigned ticker count — private-company IPO gossip, banker mandates, and funding rounds carry zero resolution weight,” and the confidential-filing headline in June actually made YES cheaper (130k contracts traded it down to 3¢). KXIPO-26-SPACEX resolved YES with all three triggers landing together in pricing week (Jun 11–12), confirming the triggers fire late and together, not at filing. KXIPOANTHROPIC-DATE-26SEP01 finalized NO on Sep 1 (verified via API this run; new precedent row written).

Resolver profile & risk pricing

Resolver: Kalshi against three objective, publicly observable triggers (EDGAR effectiveness notice, pricing press release, exchange ticker assignment). Precedent shows a strict mechanical reading across four settled instances. Priced risks: −3 pts for the ticker-assignment clause firing early (an exchange certifying the listing during the roadshow, before pricing — did not happen early in the SpaceX precedent); −2 pts for schedule-reporting error (Reuters could be wrong about the slip). Total deductions 5 pts, inside the 15-pt gate, already reflected in the 18% final.

The tail

Cleanest way NO loses: Anthropic files its public S-1 this week or next (a new discrete event — checkable daily on EDGAR) and runs a compressed roadshow that prices by Oct 16. That is an acceleration chain contradicting the Sep 5 reporting, not a continuation — the current trend (each report has pushed the schedule later) helps this pick. Priced in at 18%: the market itself opened this strike at 77¢ on Sep 4 and traded it down to 28–33¢ after the Reuters slip report, so the residual 31–33¢ is mostly early-flip hope, not information we lack.

Entry context & 48h check

Book at signal: YES 31 bid (53) / 33 ask (150); NO crossable at 69¢×53, 70¢×250, 71¢×422 — 725 contracts at or below the 71¢ limit. Vol24h 659 (above the 500 floor, not specialist-dense), OI 2,157. History: opened 77¢ Sep 4 → 23–35¢ after the Sep 5 Reuters report → 28–33¢ since; today ticked 28→33 on ~550 contracts (post-Labor-Day flip anticipation). No adverse move we can't explain; slippage from 69¢ signal to 71¢ limit is 2¢ (≤3¢ gate).

• Concentration: the single new leg shares an event with one existing live pick (Oct 1 NO), but that leg is 98–99¢ won with 22 days of pure publication-lag risk left; joint downside requires an IPO confirmation before Oct 1, which is off the table on any reported schedule.
• Max loss: $171.11 (6.8% of capital) if the IPO prices / goes effective / gets a ticker before Oct 17.
• Kill signal: a public Anthropic S-1 appearing on EDGAR before ~Sep 21 reopens a pre-Oct-17 pricing path — that is the exit trigger, watchable daily and free.
• Cash: 93.2% reserve. An efficient week; the record says this is a position, not a failure.

• Rest NO bids at 69–70¢ first (the 24–27¢ YES-side resting stack suggests sellers appear intraday); cross to 71¢ only for remainder. Do not chase past 71¢ — at 72¢ the edge is 10¢ against a 7,656-contract wall of 28¢ YES bids that will slow any further NO fill improvement.
• Market can close early if the event occurs — a fill is final; no exit liquidity is assumed after a surprise flip + accelerated calendar. Size is set to hold to resolution.
• Re-check EDGAR (company search: Anthropic, Form S-1) each trading day; the thesis dies loudly, not quietly.