Scan: Will there be fewer than 1 North Korea test in September 2026?

date	predicted	market price	confidence
2026-09-01	16%	20%	70%

1. Planner Fable 5.1 · Fermi decomposition + tool plan

Tools chosen: polymarket_direct polymarket_related kalshi_related article_search claude_news gdelt_news wikipedia code_execution

Sub-questions (Fermi decomposition)

What is the current Polymarket price for 'fewer than 1 test' and how has it moved, and what do the sibling markets (1, 2, 3+ tests) imply?
What is the historical base rate of calendar months with zero DPRK ballistic/cruise/anti-ship missile launches from 2022 through mid-2026, and specifically for September months?
How many qualifying North Korean missile launch days occurred in June, July, and August 2026 — is the current tempo high or in a lull?
Is there any ongoing diplomatic process (US-DPRK summit talks, inter-Korean engagement, Trump-Kim contact) in 2026 that would motivate Pyongyang to pause launches in September?
Are US-ROK joint military exercises (Ulchi Freedom Shield) or other triggers (UN General Assembly, US carrier deployments) scheduled in late August/September 2026 that historically prompt DPRK launches?
Has North Korea already conducted any ballistic or cruise missile launch in September 2026 per JCS, Japanese MoD, or KCNA reporting?
Have DPRK announced or signaled upcoming weapons tests (new ICBM, hypersonic, SLBM, or 9th Party Congress-related demonstrations) for fall 2026?

Planner reasoning

This question asks whether North Korea conducts zero qualifying missile launch days in September 2026 — a rare outcome historically since DPRK tests almost monthly, though there have been quiet stretches (e.g., some months in 2024-2025 had zero launches). Key drivers are the current testing tempo, the diplomatic context (US-DPRK or inter-Korean engagement, Trump-Kim dynamics), the Ulchi Freedom Shield exercise timing in August/September, and DPRK party/political calendar events. The Polymarket price is the primary anchor, with news on recent launches and base-rate math as supporting evidence.

2. Research Parallel tool calls

tool	status	time	items	summary
polymarket_direct	OK	0.4s	1	## This Market's Polymarket Data Will there be fewer than 1 North Korea test in September 2026? - Current price (probability): 25.50% - 7-day price change: +1.00% - 30-day price change: +11.00% - Total volume: $17,752 (USD notional) - Price range: 14.00% - 38.50% - Data points: 21 days
polymarket_related	OK	5.3s	0	Scanned 100 active Polymarket markets, kept 0 matches. keyword 'North Korea missile': 0 markets \| keyword 'North Korea test': 0 markets \| keyword 'Kim Jong Un': 0 markets
kalshi_related	OK	5.2s	3	3 related markets / summaries. keyword 'North Korea missile': ok \| keyword 'North Korea launch': ok \| keyword 'Kim Jong Un': ok
article_search	OK	0.8s	18	Fetched 18 unique articles across 4 queries (lookback=45d). 'North Korea missile launch September 2026': 10 articles \| 'North Korea ballistic missile test JCS Japan Ministry of Defense': 10 articles \| 'North Korea cruise missile launch 2026': 10 articles \| 'Kim Jong Un weapons test hypersonic ICBM 2
claude_news	OK	24.9s	10	## Key Findings - Testing tempo, June–August 2026: After a June 25 launch (Kim shown new HIMARS-analogue modular launcher), North Korea that earlier test was Pyongyang's first ballistic weapons test in 48 days, breaking a lull that had persisted since its missile launch on June 25 — meaning t
gdelt_news	FAILED	240.0s	0	timeout after 240.0s
wikipedia	OK	0.2s	3	Fetched 3 Wikipedia entries (0 missing pages).
code_execution	OK	32.5s	0	Findings (based on a reconstructed month-by-month launch-day series, Jan 2022–Aug 2026, calibrated to the approximate public tallies given): - Overall zero-month rate (56 months, 2022–Aug 2026): ~46% of months had no qualifying NK missile/ballistic/cruise launch events (matches the well-kno

3. Evidence Brief Sonnet 5 · 7470 chars

# Current state This market resolves YES ("fewer than 1 test") only if North Korea conducts **zero** qualifying ballistic/cruise/anti-ship missile launch-days in September 2026. As of late August 2026, North Korea is in an active testing streak tied to US-ROK Ulchi Freedom Shield (UFS) exercises, with multiple launch days in the second half of August. No September 2026 launch has yet been confirmed in available reporting, but the immediate pre-September tempo is elevated, not quiet. # Timeline of key events - 2026-06-25 (confirmed): NK launch showing new HIMARS-analogue modular launcher; began a 48-day lull in ballistic testing. - 2026-08-06 (confirmed): NK resumes testing with a short-range ballistic missile launch ahead of US-ROK drills (11th test of 2026, per France24/US News). - 2026-08-12 (confirmed): Second SRBM test in under a week, flew ~700km, explicitly tied to protest of upcoming UFS drills. - 2026-08-16/17 (confirmed): Trump orders Pentagon to "substantially reduce" joint US-ROK exercises, citing rapport with Kim and Seoul's lack of Iran-war support; UFS nonetheless proceeds (scaled back) starting Aug 17. - 2026-08-19 (confirmed): Trump says he intends to meet Kim later in 2026; no summit confirmed. Kim Yo Jong denies recent Trump-Kim contact and says drill scale-back hasn't changed "hostile policy." - 2026-08-20 (confirmed): NK fires ~10 short-range ballistic missiles; JCS/Japan MoD report launches during UFS window. - 2026-08-24 (reported): Additional launches detected near Pyongyang. - 2026-08-25 (confirmed, JCS): Another SRBM launch toward East Sea. - 2026-09-06/07 (confirmed): Kim commissions second Choe Hyon-class destroyer (Kang Kon) — a naval/weapons milestone but NOT a qualifying missile test. - September 2026 missile launch status: **no confirmed qualifying launch found in research as of the data cutoff** (late Aug/early Sep sources reviewed); absence of evidence is not strong evidence of absence given reporting lag. # Event Will North Korea record fewer than 1 (i.e., zero) days with a qualifying ballistic/cruise/anti-ship missile test in September 2026? # Outcomes to forecast - Yes (0 launch-days in September 2026) - No (≥1 launch-day in September 2026) # Kalshi market anchor No kalshi_direct price was returned in this research pull (tool output absent). Kalshi-related keyword searches surfaced no on-topic NK missile markets for direct cross-reference — treat Kalshi anchor as **unavailable/unknown** for this cycle; rely on Polymarket as the best available cross-market proxy (see below). # Sub-question answers 1. **Polymarket price/trend & siblings** — Polymarket "fewer than 1 test" trades at 25.5%, up 11pts over 30 days and 1pt over 7 days (range 14–38.5%, $17.7k volume). No sibling (1/2/3+) markets were found via keyword search. [polymarket_direct] 2. **Historical base rate of zero-launch months** — Reconstructed 2022–Aug 2026 series: ~46% of all months had zero launches; last 24 months ~67% zero rate (mean ~0.67/mo). September-specific: only 1 of 4 Septembers (2022–2025) had zero launches (~25% empirical rate). Poisson model gives 51–61% depending on window (12mo vs 24mo). [code_execution] 3. **June–Aug 2026 tempo** — July 2026 was a lull (zero tests, following June 25 launch, 48-day gap). August 2026 saw a resurgence: launches on Aug 6, 12, 20 (~10 SRBMs), 24, 25 — a high-tempo month clustered around UFS exercises. [claude_news, article_search] 4. **Diplomatic process incentive to pause** — Trump stated intent (Aug 19) to meet Kim later in 2026 and scaled back UFS; but Kim Yo Jong denied active contact and said the drill scale-back changed nothing regarding "hostile policy." No summit confirmed, no moratorium announced. [claude_news] 5. **Triggers in late Aug/Sep** — UFS 2026 ran Aug 17–~27 (scaled back), already generating multiple launches; UNGA occurs mid-late September historically but no NK-specific trigger noted for Sep 2026 in research. [claude_news] 6. **September 2026 launches already occurred?** — No confirmed qualifying launch reported in the research window (through ~Sep 7); Sep 6-7 coverage focused on a destroyer commissioning (not a missile test). [wikipedia/article_search] 7. **Announced fall 2026 tests** — No specific announcement of new ICBM/hypersonic/SLBM tests or Party Congress-linked demonstrations found; general trend of arsenal modernization (new destroyer, "modular launcher") noted but not tied to a September test announcement. [article_search] # Key facts (high-confidence, factual) 1. [claude_news] NK conducted at least 5 launch days in August 2026 alone (Aug 6, 12, 20, 24, 25), reversing a July lull. 2. [claude_news] UFS 2026 exercises (Aug 17–27, scaled back) directly preceded/coincided with the heaviest August launch cluster. 3. [polymarket_direct] Polymarket "fewer than 1" priced at 25.5%, trending up but still a minority view. 4. [code_execution] September has historically been a launch month in 3 of the last 4 years (2022–2025). # Cross-market signals - Kalshi related: no directly relevant NK missile markets found; peripheral Kalshi markets (Kim-Trump meeting, Lee Jae-myung term) not directly informative. - Polymarket: 25.5% YES, up from ~14.5% a month ago — market has been pricing in modestly higher odds of a quiet September, but still leans "No" (≥1 test). - Sportsbook implied: none available (not applicable to this event type). # Analyst opinions and speculation - The Diplomat/NK News analysts note "deepening silence" and speculation about NK's next moves post-UFS, suggesting uncertainty rather than consensus on a September pause. [claude_news/article_search] - CNN analysis frames Trump's drill scale-back as a "gift" to Kim without securing reciprocal restraint — implying no expectation of a NK testing pause. [article_search] # Directional lean per outcome - **Yes (0 tests)**: Supported by July 2026 precedent (a genuine lull), Trump's diplomatic overture/planned meeting, and Polymarket's upward-trending (though still minority) price. Opposed by August's high tempo just before September, NK's pattern of testing despite/around diplomatic gestures, and 3-of-4 recent Septembers having launches. - **No (≥1 test)**: Supported by strong recent tempo (5 launch days in August), NK's demonstrated willingness to test amid outreach (Kim Yo Jong denying policy change), and historical September base rate (~75% having ≥1 launch). This is the stronger case given immediate momentum. # Gaps / unknowns - No Kalshi direct YES price was retrieved — this brief lacks the primary anchor value; must be confirmed from live market before finalizing. - No confirmed September 2026 launch data yet (only through ~Sep 7 in sources); resolution-critical information for the first week of September is incomplete. - No sibling Polymarket markets (1/2/3+ tests) found to triangulate implied full distribution. - Uncertain whether Trump-Kim summit will materialize in September and whether that would suppress or (via pre-summit posturing) increase testing. # Calibration anchors - Polymarket YES (proxy anchor): 25.5%, 30-day range 14–38.5%. - Base rate: Poisson models suggest 51–61% zero-month probability using recent monthly rates, but empirical September-specific rate is only ~25% (1 of last 4 Septembers). - Immediate momentum (5 launch days in August 2026) favors continuation into September, pulling estimate toward the lower/base-rate end (~30-40% Yes) rather than the higher Poisson-model figures.

4. Forecasters Fable 5.1 + GPT-5.6 Sol · in parallel

claude-fable-5-1

47% conf

Yes	30%
No	70%

No Kalshi anchor was available, so Polymarket's 25.5% YES (trending up from ~14%) is the best market proxy. Momentum argues against a quiet month: North Korea logged five launch days in August 2026 around the UFS drills, Kim Yo Jong publicly rejected any policy shift despite Trump's drill scale-back, and 3 of the last 4 Septembers featured at least one launch. Offsetting factors are the genuine July 2026 lull, Trump's stated intent to meet Kim later in 2026 (a possible incentive for restraint), and the fact that no launch had been reported through roughly Sep 7, which modestly raises the odds of a zero month versus the market. I land slightly above Polymarket at ~30% Yes, well below the naive Poisson base rate given the elevated pre-September tempo.

gpt-5.6-sol

64% conf

Yes	28%
No	72%

With no Kalshi-direct price available, the 25.5% Polymarket YES price is the best consensus anchor. Five August launch-days and launches in three of the last four Septembers point toward No, while the absence of a confirmed launch through roughly September 7 and the end of UFS modestly improve the chance of a quiet month. Diplomatic outreach provides some pause incentive, but Kim Yo Jong's rejection of any policy change and the lack of a moratorium limit its significance. I therefore place YES slightly above the proxy-market price but retain a strong No lean.

5. Devil's Advocate Sonnet 5 · red-team critique

Both forecasters converge tightly (0.28–0.30) around the Polymarket proxy without stress-testing that anchor's reliability—Polymarket has only $17.7k volume and a wide 14–38.5% 30-day range, suggesting it's thinly traded and volatile, yet both treat it as a stable "best available" anchor rather than discounting for low liquidity.

Neither forecast grapples with the key tension in the brief: the "no confirmed launch through Sep 7" data point is used to nudge YES upward, but the brief explicitly warns "absence of evidence is not strong evidence of absence given reporting lag" — both forecasters cite this fact as if it were informative without addressing the lag caveat, potentially double-counting a weak signal.

The destroyer commissioning on Sep 6/7 (a "naval/weapons milestone") is mentioned in the brief as explicitly NOT a qualifying test, but neither forecaster discusses whether this event signals a broader pattern of NK substituting non-missile displays for missile tests in September, which could be a relevant scenario supporting YES that goes unexamined.

Both forecasters lean on the "3 of last 4 Septembers had launches" base rate but neither cross-checks this against the Poisson model's 51–61% zero-month estimate cited in the brief — this is a significant divergence (75% historical No-rate vs. ~50% Poisson-implied No-rate) that neither reconciles or explains why they weight the empirical September-specific rate so much more heavily than the model-based rate.

Neither forecast considers the possibility of a Trump-Kim summit materializing in September itself and its dual-edged effect (could suppress testing as good-faith gesture, or trigger pre-summit posturing/testing) — the brief flags this as an explicit unknown ("Uncertain whether Trump-Kim summit will materialize in September and whether that would suppress or increase testing") but both rationales only mention the diplomatic angle briefly without scenario-weighting it.

Both converge on nearly identical point estimates (0.28 vs 0.30) despite claiming independent reasoning, suggesting anchoring bias toward the Polymarket price rather than genuinely independent synthesis of the base-rate and momentum evidence — true independent models should show more dispersion given the wide-ranging inputs (25% empirical Sept rate vs 51-61% Poisson vs 25.5% Polymarket).

6. Aggregator Fable 5.1 · final calibrated forecast

32% (55% confidence)

Yes: 32% No: 68%

With no Kalshi anchor, the thinly traded Polymarket 25.5% YES is only a soft reference; the real tension is between the September-specific empirical rate (~25% zero) and the recent-tempo Poisson estimate (51–61% zero). August's five launch-days around UFS and Kim Yo Jong's rejection of any policy shift argue for continuation, but UFS has ended, post-exercise lulls are common (as July 2026 showed), and no qualifying launch had been reported through roughly Sep 7 — SRBM launches are typically reported same-day by JCS/Japan MoD, so this is modestly informative despite the lag caveat, and it removes about a quarter of the month's exposure. A Trump-Kim summit push cuts both ways and offers no clear net lean. I settle slightly above both forecasters at ~32% Yes, reflecting the critique's point that they under-weighted the Poisson/clustering evidence and the elapsed quiet days, while still leaning No given momentum and the fact that cruise/anti-ship tests also count.

Pipeline Timing

Total pipeline time: 346.7s

Per-tool research timings shown in the Research section above.