Kalshi Economics — Mispricing Audit

Report date 2026-06-10 · Category Economics · Horizon ≤45 days · Portfolio capital $1,000

The Economics book is a hard place to find edge: most of it is efficient macro markets and not-yet-released data prints I have no public-information advantage on. The honest result of this screen is thin — two defensible picks, both on the AAA retail-gasoline complex, and one headline-looking "mispricing" (the June FOMC having zero dissents priced at ~70%) that turns out to be a trap once you read the news. I'd rather ship two picks I can defend than pad the list.

1 · How this was researched

Universe. Pulled every active market in trading_events.category = 'economics' closing before 2026-07-25 (≤45 days) via the read-only DB mirror — ~400 markets across ~100 series (CPI, unemployment, Treasury yields, gas, central-bank decisions, payrolls, PCE, …).
Liquidity screen. Joined each market to its most-recent market_snapshots row and cut anything with 24h volume < 1,500, bid-ask > 5¢, or already pinned (YES bid ≥95¢ / ask ≤5¢). That left ~11 genuinely tradeable contracts.
Diligence per candidate. Read the exact rules text (resolution is often subtler than the title); reconstructed each underlying's recent trajectory from the snapshot ladder; hit the live Kalshi orderbook for executable depth; and grounded priors in primary-source news (AAA, BEA/BLS calendars, Federal Reserve, wire coverage of the Israel–Iran situation).
Reality check. The mirror is the live 2026 world: the AAA daily national average I reconstructed from the gas ladder matches the published figures to the cent (Jun 4 ≈ $4.24, Jun 6 = $4.19, Jun 9 = $4.161). That means real-world news is directly usable for pricing.
Reject hard. Anything that's an un-released future print with no public edge, definitionally ambiguous, already priced, or un-executable without ≥3¢ slippage was cut to the rejects/watchlist.

2 · Markets reviewed

Ticker	Question	YES bid/ask	Vol 24h	Verdict
KXAAAGASW-26JUN15-4.120	AAA US avg gas > $4.120 on Jun 15	51 / 52	3,504	PICK 1 — BUY YES
KXAAAGASM-26JUN30-4.40	AAA US avg gas > $4.40 on Jun 30	22 / 24	1,587	PICK 2 — BUY NO
KXAAAGASW-26JUN15-4.100	AAA US avg gas > $4.100 on Jun 15	27 / 70*	1,914	WATCHLIST (spread)
KXFOMCDISSENTCOUNT-26JUN-0	Exactly 0 dissents at Jun 17 FOMC	69 / 71	4,846	REJECT — trap
KXAAAGASW-26JUN15-4.140	AAA US avg gas > $4.140 on Jun 15	24 / 25	6,975	REJECT — fair
KXAAAGASW-26JUN15-4.160	AAA US avg gas > $4.160 on Jun 15	8 / 9	6,325	REJECT — fair tail
KXAAAGASM-26JUN30-4.20	AAA US avg gas > $4.20 on Jun 30	42 / 44	2,791	REJECT — too near spot
KXCPI-26JUN-T0.2	June CPI m/m > 0.2%	19 / 20	5,494	REJECT — unreleased
KXPCECORE-26MAY-T0.3	May core PCE m/m > 0.3%	4 / 7	2,267	REJECT — unreleased
KXJOBLESSCLAIMS-26JUN11-220000	Initial claims ≥220k (wk Jun 6)	51 / 53	1,741	REJECT — unreleased
KXPAYROLLS-26JUN-T-25000	June NFP > −25,000	96 / 97	2,058	REJECT — priced
KXHOUSINGSTART-26JUN16-T1.400	May housing starts > 1.400M	86 / 92	2,008	REJECT — spread/unreleased

*Snapshot caught 4.100 at a momentary 70/72; the live book is 0.27/0.70 — buying YES costs ≥70¢ against a ~62¢ fair mid, so it fails the slippage test and moves to the watchlist.

3 · The picks

Ordered by how defensible the edge is (rules clarity + source quality + executable depth), not by raw EV. Two picks, both on the AAA gas complex — see the portfolio section for how they partially hedge each other.

KXAAAGASW-26JUN15-4.120 · BUY YES @ 0.52 Medium conviction

Resolves

YES if the AAA national average for regular gas is strictly > $4.120 on Jun 15, 2026.

Current

YES 0.51 / 0.52 (last 0.51) · close 2026-06-15T03:59Z

My fair

P(>$4.120) ≈ 0.60 → ~+8¢ edge vs the 0.52 ask

Thesis — the market is extrapolating a finished decline. The AAA national average has fallen from $4.393 (May 29) → $4.161 (Jun 9) ≈ $4.151 (Jun 10), mirroring crude's pullback from its Israel–Iran war-spike. But the fall has decelerated to a near-stall: daily moves over the last three sessions were −1.1¢, −1.0¢, ~0¢. Retail gas lags crude by 1–2 weeks; with Brent stabilising around $95–96, most of the downward pass-through is already done, so the natural landing zone is ~$4.13–4.15 — above $4.12. The market's implied median for Jun 15 (~$4.12, i.e. another ~3¢ drop) prices in continued decline that requires a fresh catalyst.

Evidence

Reconstructed AAA daily average from the KXAAAGASD ladder (50/50 crossover strike each day): May29 $4.393 · Jun01 $4.323 · Jun03 $4.253 · Jun05 $4.221 · Jun07 $4.172 · Jun08 $4.161 · Jun09 $4.161 · Jun10 ≈$4.151. Deceleration is unmistakable in the last three prints.
Cross-checks against published AAA figures: Jun 4 ≈ $4.24, Jun 6 = $4.19, Jun 9 = $4.1610 — the mirror is the real series (AAA Fuel Prices, MacroMicro AAA daily series).
Crude has flattened, not crashed: Brent ~$96.25 / WTI ~$94.73 as ceasefire hopes capped the move (CNBC, Jun 4 2026). A flat crude floor → a flat retail floor.
The contract was created Jun 8 and is thin/2-day-old; today it whipsawed 28→55→52 intraday, which is consistent with a not-yet-efficient price rather than a settled consensus.

Tail risk (the single cleanest way this loses): a firm Israel–Iran / Strait-of-Hormuz de-escalation deal actually lands inside the window. President Trump publicly said an agreement was reachable "over the next week" (CNN, Jun 1 2026) — that window is Jun 10–17. A signed deal would drop Brent toward the $80s and pull retail gas below $4.12 by Jun 15. That ~35–40% scenario is exactly why this is Medium, not High, and why it's sized as the larger but not dominant leg.

Numbers. Fill 0.52, fair 0.60. Edge +8¢/contract. Per-dollar return if it wins = (1−0.52)/0.52 = +92%; expected value ≈ +15% per dollar at fill. Liquidity: live YES ask 0.52 with $1,009 resting at that price and $526 more at 0.54 — my whole intended size fills at 0.52 with zero slippage.

KXAAAGASM-26JUN30-4.40 · BUY NO @ 0.78 Low conviction

Resolves

YES if the AAA national average for regular gas is strictly > $4.40 on Jun 30, 2026. I am buying NO (gas ≤ $4.40).

Current

NO 0.76 / 0.77 (YES 0.22 / 0.24) · close 2026-06-30T03:59Z

My fair

P(>$4.40) ≈ 0.15 → NO worth ~0.85 → ~+7¢ edge vs ~0.78 fill

Thesis — a +25¢ round-trip in 20 days needs a war, and the war is de-escalating. Spot is ~$4.15 and drifting down. For the average to be above $4.40 on Jun 30, gas has to reverse and add ~25¢ (+6%) in three weeks — which realistically requires crude to jump from ~$95 back toward $110–120, i.e. a serious Israel–Iran re-escalation (Hormuz closure or collapsed ceasefire). The current momentum is the opposite: a Lebanon ceasefire is in place and Washington is signalling a broader deal. The market's 22–24% on ">$4.40" looks rich against that backdrop.

Evidence

Gas was $4.39 on May 29 and has fallen every week since on easing crude (AAA news); the burden for a YES is a full reversal of a multi-week trend.
Oil is easing on diplomacy, not tightening: Brent −1.6% to $96.25, WTI −1.3% to $94.73 on the Lebanon ceasefire (CNBC, Jun 4 2026).
The market itself moved my way today: P(>$4.40) drifted 28 → 22 over the session, so some of the edge is already being recognised — a reason to size this small.

Tail risk (the single cleanest way this loses): the Israel–Iran ceasefire collapses and crude spikes on a Strait-of-Hormuz threat — a genuinely fat-tailed, hard-to-handicap geopolitical event. Iran has already suspended talks and accused Israel of violating the Lebanon ceasefire, so this is not negligible. That fat left tail, the 20-day horizon, and the partially-eaten edge are why this is Low conviction and the smallest leg.

Numbers. Fill ≈ 0.78 (NO), fair ≈ 0.85. Edge +7¢/contract; EV ≈ +8–9% per dollar. Liquidity: thinner — ~$170 of NO available at 0.77–0.79 and ~$60 more at 0.80; fine for a small, ≤$200 leg but not scalable. Use a limit and accept partial fills.

4 · Recommended $1,000 portfolio

Deploy ~$826 across the two picks, hold ~$174 (17%) in cash for opportunistic adds (chiefly the 4.100 watchlist if it tightens). The two legs partially hedge on the geopolitical axis: a crude spike (escalation) wins Pick 1 and loses Pick 2; a crude collapse (a signed deal) loses Pick 1 and wins Pick 2; the most-likely "status-quo" path (crude parked ~$95, gas ~$4.10–4.20) wins both. They cannot both lose except in a violent down-then-up whipsaw.

Pick	Side	Limit	Contracts	Cost	Max payout	Fair	EV ¢	EV %	Conv.
KXAAAGASW-26JUN15-4.120	YES	0.53	1,200	$630	$1,200	0.60	+7.5	+14.3%	Med
KXAAAGASM-26JUN30-4.40	NO	0.80	250	$196	$250	0.85	+6.5	+8.3%	Low
Total deployed				$826	$1,450			+12.8%
Cash reserve				$174	$174

Risk profile

Worst case (both lose): a sharp gas crash below $4.12 by Jun 15 and then a spike above $4.40 by Jun 30 — a whipsaw that's nearly mutually exclusive. Realistic worst case is one leg losing: lose Pick 1 only (gas slides on a deal) = −$630 on that leg but Pick 2 wins (+$54 net there); lose Pick 2 only (escalation) = −$196 there but Pick 1 wins big. Maximum capital at risk is the $826 deployed.
Best case: status-quo crude → both resolve YES-for-me → $1,450 returned on $826, a +$624 gain (+76% on deployed, +62% on the full $1,000).
Most-likely outcome: gas parks in the $4.10–4.20 band; Pick 1 is a true coin-flip-plus that I expect to win ~60% of the time, Pick 2 wins ~85%. Blended expected dollar edge ≈ +$106.
Concentration: 100% of deployed capital is the same underlying (US retail gasoline / crude). This is a single-factor book. The internal hedge softens directional risk but a data-definition or AAA-methodology surprise would hit both — sized accordingly (well under the full bankroll, large cash reserve).

Execution notes

Pick 1: rest a limit at 0.52, willing to 0.53. ~$1,000 is available at 0.52 — do not chase above 0.54 (edge thins fast). If it trades back toward 0.45 on a quiet crude tape, add from the reserve.
Pick 2: limit 0.78 on the NO, accept up to 0.80, accept partial fills — the book is thin. Do not market-buy.
Watchlist — KXAAAGASW-26JUN15-4.100 YES: attractive on fundamentals (fair ~0.78–0.80) but the live spread is 0.27/0.70. Rest a 0.62 limit; only a fill at ≤0.65 clears the slippage bar. Do not pay the 0.70 ask.
What invalidates each thesis: Pick 1 — a confirmed Israel–Iran/Hormuz agreement, or two consecutive AAA prints resuming a >1.5¢/day decline before Jun 13. Pick 2 — any Brent move back above ~$105, or the Lebanon ceasefire formally collapsing. Either headline should trigger an immediate re-evaluation (and the legs hedge, so one breaking is partly offset by the other).
Hedge already embedded: no separate hedge needed; the two legs are the hedge. If only running one leg, prefer Pick 1 (cleaner liquidity, shorter horizon).

5 · What I rejected and why

The headline trap: "June FOMC has 0 dissents" priced at ~70%

KXFOMCDISSENTCOUNT-26JUN-0 screams mispriced on a naive read: the April 29, 2026 FOMC produced four dissents (Stephen Miran for a cut; Hammack, Kashkari and Logan over the easing-bias language) amid what reporters called a "deepening divide" (FOMC minutes, Apr 28–29 2026). A jump from four dissents to zero looked too cheap, so "buy NO @ ~0.31 (bet on ≥1 dissent)" looked like free money.

It's a trap. The reason April had four dissents has largely evaporated for June:

The lone cut-advocate resigned. Miran — who dissented at all six meetings he attended — submitted his resignation in mid-May 2026, endorsing incoming Chair Kevin Warsh, and is gone for the June 16–17 meeting (CNBC, May 14 2026; Fortune). The single most reliable source of a dovish dissent is no longer on the committee.
The easing-bias fight is resolvable. The other three April dissents were about keeping easing-bias language, not the rate itself. With inflation re-accelerating (Apr CPI 3.8% y/y) and forecasters expecting the June statement to remove the easing bias, those three are likely satisfied rather than dissenting.
New-chair honeymoon. A first meeting under a new Chair (with Miran's public blessing) tends toward a show-of-unity vote. ~99% of markets price a hold; the rate decision itself is not in doubt.

Net: P(0 dissents) of ~70% is fair to slightly low, not a buy on either side. Already an active pick? No — but it's the cleanest example this week of a title that lies until you read the news. no defensible edge

Other rejects

Un-released data prints I have no edge on: June CPI (KXCPI-26JUN-*, releases Jul 14), May core PCE (KXPCECORE-26MAY-*, ~Jun 26), initial claims for the week ending Jun 6 (KXJOBLESSCLAIMS-26JUN11-*, Jun 11), May housing starts (KXHOUSINGSTART-*). These are single future releases priced around consensus; I have no public-information advantage, so any "edge" would be a guess. Reject.
Priced-in near-certainties: KXPAYROLLS-26JUN-T-25000 ("June payrolls > −25,000") at 0.96/0.97. Almost certainly true, but ~2¢ of edge against a real left-tail (a shock recessionary print) is not worth the capital. Reject.
Same underlying, no incremental edge: KXAAAGASW-26JUN15-4.140 / -4.160 look fairly priced for their distance from spot; KXAAAGASM-26JUN30-4.20 sits right on spot (max uncertainty). Adding them would just stack correlated gas risk without improving the edge. Reject.
Un-executable: KXAAAGASW-26JUN15-4.100 — good fundamentals, but a 0.27/0.70 live spread means ≥3¢ slippage to enter. Moved to the watchlist with a resting limit. Not deployed.

6 · Sources

AAA Fuel Prices — daily national average (settlement source) — gasprices.aaa.com · news
AAA daily national average series (cross-check) — MacroMicro
Oil eases on Lebanon–Israel ceasefire; Brent $96.25 / WTI $94.73 — CNBC, Jun 4 2026
Trump: Iran/Hormuz deal "reachable over the next week" — CNN live, Jun 1 2026
Fed Governor Miran resigns, backs Warsh — CNBC, May 14 2026 · Fortune · CNBC, May 15 2026
April 28–29 2026 FOMC minutes (four dissents) — Federal Reserve · FOMC calendar
Live market data & orderbooks — Kalshi public API; price/volume history — Kalshi DB mirror (market_snapshots).

Data sources: Kalshi read-only DB mirror (trading_markets, trading_events, market_snapshots) for universe, rules and price history; Kalshi public API for live top-of-book and depth; AAA / BEA / BLS / Federal Reserve and wire coverage for primary-source grounding. Prices captured 2026-06-10.

Disclaimer: Not financial advice. Probabilities here are subjective estimates, not guarantees; the "edge" is my judgement against the market's. Event contracts resolve to $0 or $1 — a losing contract is a total loss of the premium paid. Sizes assume the quoted liquidity is still there at execution; verify the live book before trading. This report is published to a public archive for educational purposes.