Source of truth for prices is the live Kalshi mirror (trading_markets joined to trading_events for category, market_snapshots for intraday history) plus the Kalshi public orderbook API for current depth. News diligence is grounded in primary reporting and official statements, cited in §6.
The Politics board this fortnight is dominated by one fast-moving story — the collapse of the Director of National Intelligence (DNI) succession — and that is where two of the three picks sit. The third is a clean "fade an overpriced near-term tail" on Mitch McConnell.
| Ticker | Market | Px (¢) | Read | Verdict |
|---|---|---|---|---|
| KXGABBARDOUT-26JUN-JUN29 | Gabbard out as DNI before Jun 29 | YES 74 | Pulte takes over Jun 19 → she's out | PICK · HIGH |
| KXLEAVEHOUSE-26APR-MMCC | McConnell out as Senator before Jul 2026 | NO 90 | No resignation signal; serves to Jan '27 | PICK · MED |
| KXCLAYTONCONF-26JUN11-JUN24 | Clayton confirmed DNI before Jun 24 | NO 60 | Full-Senate confirm in <1 wk is a squeeze | PICK · LOW |
| KXFABLERESTORE-27-26JUL01 | Claude Fable 5 access restored before Jul 1 | 56 | Coin flip; fell ~12¢ today, actively repricing | REJECT |
| KXTRUMPUFC-26JUL-DJT | Trump attends UFC 329 (Jul 11, Las Vegas) | YES 17 | No announcement; ~base-rate fair | REJECT |
| KXKASHOUT-26APR-JUL01 | Kash Patel out as FBI Director before Jul 1 | YES 7 | No catalyst; tail fairly priced | REJECT |
| KXTARIFFRATEPRC-26JUL01-5 | China import tariff below 10% on Jul 1 | YES 6 | Tariffs far above 10%; ~priced | REJECT |
| KXAPRPOTUS-26JUN19-* | RealClearPolitics approval-rating buckets | — | Data market; no edge without live RCP feed | REJECT |
| KXLAGODAYS-26JUN-0 | Trump makes 0 Mar-a-Lago trips in June | YES 94 | Likely true but already priced | REJECT |
| KXHORMUZWEEKLY-26JUN14-* | Strait of Hormuz weekly transit-call ladder | — | PortWatch data I can't independently verify | REJECT |
| KXDJTATTACKWARSH-27-JUL | Trump criticizes Kevin Warsh before July | YES 7 | Warsh a Fed favorite; tail ~priced | REJECT |
Prices are mid/ask at the time of writing (2026-06-15, ~16:00 UTC). "Px" shows the side referenced.
Resolves YES if "Tulsi Gabbard leaves as Director of National Intelligence (DNI) before Jun 29, 2026." Plain, role-based language — the moment someone else is running ODNI, she has "left."
The market still carries a ~26% discount for a departure-date reversion that the news has already overtaken. Gabbard's resignation was first announced (May 22) as effective June 30 — one day after this market's cutoff, which is why it traded ~15¢ for most of early June. That date has since collapsed forward: interim DNI Bill Pulte is set to take over on June 19, ten days before the cutoff. Majority Leader John Thune has publicly referenced June 19 as the handover date. Once Pulte is sworn in, Gabbard has left → YES. At 74¢ the market prices this at coin-flip-plus; the public record supports ~88%.
The handover date has already moved twice; if the June 19 swearing-in slips back toward the original June 30 effective date — e.g. a brief Pulte standup delay — Gabbard is still DNI on June 28 and the market resolves NO. That is the one clean loss path, and it is exactly what the residual 26¢ is paying for.
Buy YES at 74¢ → 26¢ profit / contract if it resolves YES. At P≈0.88, EV ≈ 0.88·100 − 74 = +14¢ (+19% on stake). Top-of-book depth to buy YES is ~$669 within 3¢ — adequate for a few-hundred-dollar clip with <2¢ slippage.
~15¢ June 5–8 → spike on the Pulte ouster drama → 74–78¢ since June 12, stable for three sessions on ~25k–75k daily volume. The thesis is largely priced; the remaining edge is the date-reversion discount.
Resolves YES if "Mitch McConnell leaves as Senator before July 2026." Cutoff is July 1 — about two weeks out. YES needs a resignation, incapacitation, or death inside that window.
The YES side at 12–13¢ is an overpriced near-term tail. McConnell announced in Feb 2025 he would not seek re-election but has been explicit he intends to serve out his term through January 2027. As of mid-June there is no resignation signal, no succession planning, no new health event. A useful cross-check: Polymarket prices roughly a one-third chance he leaves at any point before his term ends (a seven-month-plus horizon). Compressing that into the next two weeks puts fair YES around 3–5%, not 12–13%. NO at 90¢ therefore carries ~5–6¢ of edge with a ~95% hit rate.
McConnell is 84 and has a documented history of public freezing episodes and a fall. A sudden health event causing resignation or worse inside the two-week window is the irreducible risk and the reason this is sized small, not large.
Buy NO at 90¢ → 10¢ profit / contract on a NO resolution. At P(NO)≈0.95, EV ≈ 0.95·100 − 90 = +5¢ (+5.6% on stake). Low payout but a high-probability anchor. Depth to buy NO is deep (~$1,945 within 3¢); fills at 90–91¢ are easy.
Quiet, range-bound around 10–13¢ YES on ~5.8k daily volume — no recent thesis-moving jump, so the edge has not been eaten.
Resolves YES if "Jay Clayton is the first person confirmed as Director of National Intelligence before Jun 24, 2026." Key word: confirmed — a full-Senate floor confirmation, not an acting designation. Pulte taking over as acting DNI does not trigger YES.
YES at ~40¢ overstates how fast the Senate can actually confirm. Clayton was nominated June 11; his Intelligence Committee hearing is June 17. From a June 17 hearing, a full-Senate confirmation before June 24 requires committee report, a cloture filing, the intervening-day ripening, a cloture vote, and post-cloture time — for a Level-I (cabinet-rank) nominee that post-cloture clock is up to 30 hours, not the 2-hour sub-cabinet track. Even on a "warp-speed" timeline that realistically lands the floor vote on June 23–25, straddling the cutoff. Thune himself flagged that confirmation is "unlikely before June 19" and, asked about timing, said "I don't know what 'realistic' is, but we're gonna probe the limits of it." That is the language of a squeeze, not a lock. I put YES at ~33% → NO is worth ~67¢, available at 60¢.
The §702 urgency is real and cuts against me: if leadership secures unanimous consent to collapse the clock, or holds a weekend session, Clayton could be confirmed June 22–23 and NO loses. That live possibility is exactly why this is sized small and labeled LOW conviction rather than a core position.
Buy NO at 60¢ → 40¢ profit / contract on a NO resolution. At P(NO)≈0.67, EV ≈ 0.67·100 − 60 = +7¢ (+12% on stake). Depth to buy NO is the best on the board (~$1,685 within 3¢).
Listed June 12 on the nomination; settled into a ~35–40¢ YES band over June 13–15. No fresh jump, but a single procedural headline (UC deal / vote scheduled) could move it fast in either direction — watch the §702 negotiations.
Capital $1,000. Deploy $882 across three picks; hold $118 (~12%) in cash reserve for opportunistic adds (e.g. a Clayton dip on a slipped vote schedule, or a Gabbard add if it tests 70¢).
| Pick | Side | Limit ¢ | Contracts | Stake $ | Max payout $ | True P | EV ¢/ct | EV $ | EV % | Conv. |
|---|---|---|---|---|---|---|---|---|---|---|
| KXGABBARDOUT | YES | 74 | 560 | 414.40 | 560.00 | 0.88 | +14 | +78.40 | +18.9% | HIGH |
| KXLEAVEHOUSE-MMCC | NO | 90 | 300 | 270.00 | 300.00 | 0.95 | +5 | +15.00 | +5.6% | MED |
| KXCLAYTONCONF | NO | 60 | 330 | 198.00 | 330.00 | 0.67 | +7 | +23.10 | +11.7% | LOW |
| Total | — | — | 1,190 | 882.40 | 1,190.00 | — | — | +116.50 | +13.2% | — |
"True P" is my subjective probability of the side bought resolving in the money. EV$ = TrueP·payout − stake. Blended EV% = total EV$ ÷ total stake. Reserve $117.60 held in cash.
Gabbard / DNI departure
Clayton / DNI confirmation
McConnell
Rejects (context)
Data sources. Live Kalshi mirror DB (trading_markets, trading_events, market_snapshots) for prices, volume and intraday history; Kalshi public orderbook API (api.elections.kalshi.com) for depth-of-book; cited primary news for diligence. Prices captured 2026-06-15 ~16:00 UTC and move continuously — re-check the book before trading.
Disclaimer. Not financial advice. Every probability here is a subjective estimate, not a measured frequency, and reasonable analysts will disagree. Prediction-market contracts resolve to zero when they lose — you can lose 100% of the stake on any single position. Size accordingly and never deploy capital you can't lose.