codex‑open

module 1.3 · v86 ← trader

Every other module reads a number off the venue and applies a rule. This one asks gpt-5.6-luna what a question actually resolves at, without showing it the price, and bets only when the answer and the market disagree by 20% or more. Most of the time they agree, and agreement is not a trade — so most of this page is opinions the module formed and did not act on.

What the model thought, against what the market said

No market has been priced by the model yet. Every new binary market gets an estimate; this fills as they arrive.

acted on (0 of the 0 priced here) agreed, no trade (0) the diagonal is agreement; the shaded band is the 20% it will not cross

What it thought, market by market

Every market the model priced, with the argument it gave — including the ones it argued itself out of. A multi-choice market is priced one answer at a time and each answer is its own row, named after the question. On this module the reasoning is the strategy, so a decline without it is the one decline nobody can check.

marketpricemodelgapoutcome
Nothing priced yet.

Where candidates died · 0 markets judged, 0 rows

Nothing judged yet.

The book · v86

utcversionmarketbetresultwhy
09-09 15:32v86Will a Portion or All of Going-To-The-Sun Rd close this week?YES 49settled -49model 0.88; edge +0.38; luna-medium price-blind says 0.88 vs market 0.50 (edge +0.38 >= 0.2): NPS already reported an alpine closure this week, has not published a reopening, and additional rain/snow is forecast before September 13, though the rules are ambiguous about whether an ongoing pre-market closure counts.
09-09 12:28v86Will Chatgpt-type Chatbots become permanent part of our daily lives before 2100? :: NoNO 30open +14model 0.06; edge -0.44; luna-medium price-blind says 0.06 vs answer 0.50 (edge -0.44 >= 0.3): Base rate: Pew measured 24% of U.S. adults using AI chatbots daily in February 2026, leaving a 26-point gap; the rules require only one qualifying Pew/Gallup/comparable publication before 2100, not permanence, and no specific scheduled event changes the outlook, so I estimate a 94% chance the threshold is eventually documented and 6% that the No option resolves YES.
09-09 12:28v86Will Chatgpt-type Chatbots become permanent part of our daily lives before 2100? :: Yes, by 2100YES 0unfilledmodel 0.92; edge +0.42; luna-medium price-blind says 0.92 vs answer 0.50 (edge +0.42 >= 0.3): Base rate: 0/1 reputable surveys has yet documented ≥50% daily use, but daily chatbot use is already 24% and ever-use rose 18%→44% from 2023–2026; the rules clearly require only one qualifying survey, with no specific scheduled catalyst before 2100, and Pew’s recurring U.S. adult surveys are the exact resolver (no related-market witness used), so despite adoption uncertainty I estimate 92% YES. [Pew](https://www.pewresearch.org/internet/2026/06/17/americans-and-ai-2026-chatbots-smart-devices-and-views-on-impact/)
09-09 11:32v86Next Millenium Prize Problem Solved? :: Birch and Swinnerton-Dyer ConjectureNO 100open +2model 0.02; edge -0.18; luna-medium price-blind says 0.02 vs answer 0.20 (edge -0.18 >= 0.1): Base rate is 1 officially accepted Millennium solution since 2000 (~3.8% per year); Clay requires publication, two years, and broad community acceptance, no scheduled event before 2030 mechanically resolves it, the resolver is mathematical-community consensus rather than a fixed page, and the weak proxy forecast places BSD around 2047 rather than first, so I estimate only 2%.
09-09 11:32v86Next Millenium Prize Problem Solved? :: P vs NP ProblemNO 0unfilledmodel 0.04; edge -0.16; luna-medium price-blind says 0.04 vs answer 0.20 (edge -0.16 >= 0.1): Historical base rate is 1 Millennium problem solved since 2000 (Poincaré), about 3.8% annually for the set; the rules require consensus recognition of P vs NP as the next solution, with mild ambiguity if none is solved by close; no scheduled event mechanically resolves it (ICM 2030 is only a visibility milestone); the resolver is the Clay Mathematics Institute’s official Millennium Problems status page plus expert consensus, while Manifold witnesses of 6% for P vs NP solved by 2030 and 68% for any problem use materially different rules, so I discount them and the hype narrative.
09-09 11:32v86Next Millenium Prize Problem Solved? :: Yang-Mills Existence and Mass GapNO 100open +24model 0.04; edge -0.16; luna-medium price-blind says 0.04 vs answer 0.20 (edge -0.16 >= 0.1): Base rate is 1 of 7 Millennium problems solved in roughly 26 years (~4% annual for any problem), implying about 15% for any successor by 2030; Yang–Mills has no decisive current breakthrough, no scheduled resolving event, related markets use materially different rules, and the subjective consensus resolver adds ambiguity, so I estimate only 4%.
09-09 11:32v86Next Millenium Prize Problem Solved? :: Hodge ConjectureNO 100open -43model 0.07; edge -0.13; luna-medium price-blind says 0.07 vs answer 0.20 (edge -0.13 >= 0.1): Base rate is 1 Millennium problem solved in 26 years (about 3.8% annually, roughly 15% for any problem by 2030), and although AI raises that baseline, Hodge remains officially unsolved, no scheduled event mechanically resolves it, the resolver is broad mathematical-community consensus anchored by Clay, and no matching market witness was found; ambiguity over “considered solved” modestly pulls the estimate toward 0.5.
09-09 11:05v86Will the Republic of Venice be playable in CK3 by the end of 2028?YES 0unfilledmodel 0.94; edge +0.44; luna-medium price-blind says 0.94 vs market 0.50 (edge +0.44 >= 0.2): Base rate is 6/6 previously announced CK3 major expansions releasing, and Paradox’s official Silk & Silver dev diary says it adds playable republics specifically including Venice, with Q4 2026 scheduled before the 2028 close; the resolver is the official CK3 Steam DLC/announcement page, with modest discount for delay or scope risk.
09-09 10:35v86Who will be IWGP Heavyweight Champion after Destruction In Kobe? :: Yota TsujiYES 100open +7model 0.86; edge +0.53; luna-medium price-blind says 0.86 vs answer 0.33 (edge +0.53 >= 0.1): Base rate: incumbent IWGP heavyweight champions retain roughly 70% of recent singles defenses; rules are underspecified, so I interpret “after” as the official result of the September 27 event, where Yota Tsuji defends against Hirooki Goto; the scheduled October 12 match has Ryohei Oiwa facing the winner, a strong booking witness favoring Tsuji, though plans can change; the resolver is NJPW’s official event result/title-history record, with the announced card corroborated by [Sports Navi](https://sports.yahoo.co.jp/official/detail/2026090800059-spnaviow) and NJPW’s official guidebook listing Oiwa against the Tsuji-Goto winner.
09-09 10:35v86Who will be IWGP Heavyweight Champion after Destruction In Kobe? :: NeitherNO 100open +10model 0.01; edge -0.32; luna-medium price-blind says 0.01 vs answer 0.33 (edge -0.32 >= 0.1): Base rate is about 0/20 recent IWGP Heavyweight title matches yielding neither wrestler; the rules lack a description but the scheduled September 27 match is Yota Tsuji vs Hirooki Goto, and NJPW’s official title history will settle it, so only an unlikely vacancy/cancellation implies “Neither” [NJPW](https://www.njpw.co.jp/champions/iwgp) [TV Asahi](https://www.tv-asahi.co.jp/ch/contents/sports/0493/).
09-09 10:27v86Will the 2028 Democratic nominee have addressed the Aipac policy conference before?YES 33open -0model 0.82; edge +0.62; luna-medium price-blind says 0.82 vs market 0.20 (edge +0.62 >= 0.2): Historical base rate is 6/7 recent Democratic nominees (86%) having addressed AIPAC; the rule is broad but could ambiguously exclude messages or non-policy-conference events, so I adjust down for the post-Gaza boycott trend, while the likely spring 2027/2028 conferences occur before close and the resolver is AIPAC’s official conference archive cross-checked against the official Democratic nominee announcement; the related market concerns endorsement, not speaking, so it is not comparable. [historical record](https://www.washingtoninstitute.org/policy-analysis/presidential-candidates-middle-east) [Harris at AIPAC](https://www.youtube.com/watch?v=McK8bPR8pzU) [endorsement-market mismatch](https://manifold.markets/JackP/2028-democratic-nominee-previously)
09-09 10:11v86Will a Manifolder win the Metaculus Fall Cup.NO 29open +4model 0.24; edge -0.26; luna-medium price-blind says 0.24 vs market 0.50 (edge -0.26 >= 0.2): Base rate: 0/1 directly comparable settled Manifold market resolved YES, but its 100-trade threshold and bot exclusion differ; the Fall rules include bots and ambiguously define “Manifolder,” recent evidence shows the Summer Cup leader was bot Laertes, the tournament continues past the Dec 21 market close to Jan 1, and the exact resolver is Metaculus’s final Fall 2026 leaderboard, so I estimate 24%.
09-09 09:30v86Will ChatGPT Pro be available for new subscribers by: :: Oct 31.YES 0unfilledmodel 1.00; edge +0.57; luna-medium price-blind says 1.00 vs answer 0.43 (edge +0.57 >= 0.3): Base rate: Pro has remained available for roughly 21 months since its December 2024 launch; rules are clear, the official resolver currently says users can upgrade anytime, scheduled October OpenAI events provide no announced withdrawal signal, and no matching witness market was needed.
09-09 09:30v86Will ChatGPT Pro be available for new subscribers by: :: Sep 20.YES 0unfilledmodel 0.99; edge +0.56; luna-medium price-blind says 0.99 vs answer 0.43 (edge +0.57 >= 0.3): ChatGPT Pro is currently directly purchasable, has remained available since its December 2024 launch, and no scheduled event or official signal indicates withdrawal before Sep 20.
09-09 09:30v86Will ChatGPT Pro be available for new subscribers by: :: Sep 30.YES 0unfilledmodel 0.99; edge +0.56; luna-medium price-blind says 0.99 vs answer 0.43 (edge +0.57 >= 0.3): ChatGPT Pro is already directly offered to subscribers, with official pricing and Help Center pages confirming a live signup/upgrade path; only a sudden discontinuation or access restriction before Sep 30 creates meaningful downside.
09-09 07:27v86Will Barcelona beat Feyenoord?YES 47settled +32model 0.87; edge +0.37; luna-medium price-blind says 0.87 vs market 0.50 (edge +0.37 >= 0.2): Base rate: market-implied home-win probability is about 87% after removing bookmaker margin; the rules are clear, Barcelona’s strong form is secondary to that rate, the scheduled event is Barcelona–Feyenoord on September 9 at 18:45 CEST, and UEFA’s official match report is the resolver, with sportsbook prices only a proxy.
09-09 06:11v86Will aging be solved by EOY 2035?NO 28open +11model 0.03; edge -0.39; luna-medium price-blind says 0.03 vs market 0.42 (edge -0.39 >= 0.2): Base rate is 0% confirmed cases of reversible aging plus human biological immortality; current geroscience targets healthspan, no scheduled event mechanically settles this, and the resolver is the creator’s subjective purchase test, so ambiguity modestly raises the estimate toward 0.5. [JAMA review](https://pubmed.ncbi.nlm.nih.gov/40773213/)
09-09 02:58v86So uhhhNO 0unfilledmodel 0.03; edge -0.47; luna-medium price-blind says 0.03 vs market 0.50 (edge -0.47 >= 0.2): Base rate: 0% for markets lacking a defined YES condition; the rules specify only a one-year NO fallback, with no calendar event, exact resolver, or matching witness, so the unresolved wording leaves only a small chance of discretionary YES resolution.
09-09 02:10v86Will there be >250k tech layoffs in 2027 according to Layoffs.fyi?NO 0unfilledmodel 0.30; edge -0.20; luna-medium price-blind says 0.30 vs market 0.50 (edge -0.20 >= 0.2): Only 1 of 4 completed years exceeded 250,000, while 2026 is elevated and AI restructuring raises the chance of another unusually severe year; I estimate a 30% probability.
09-09 01:52v86Will a persistent “chem trail” appear on at least one UW–Madison rooftop camera from 9 AM–1 PM CT on SeptemberNO 0unfilledmodel 0.28; edge -0.22; luna-medium price-blind says 0.28 vs market 0.50 (edge -0.22 >= 0.2): Base rate: persistent contrails occur in roughly 12–15% of unobscured hourly observations; the four-hour, five-camera opportunity raises this to about 25–35%, but September is near the seasonal low. Forecasts call for sunny skies, light NNW wind, and 0–7% precipitation, favoring usable footage but not especially persistent trails; no scheduled event materially changes this. The exact resolver is the [UW–Madison AOSS archive](https://www.aos.wisc.edu/weather/wx_obs/WebCam.html), and no matching witness market was found. [EPA summary](https://hero.epa.gov/reference/1609131/)
09-09 01:43v86Guess my 3 favourite male music artists! :: PARTYNEXTDOORNO 42cancelledmodel 0.15; edge -0.35; luna-medium price-blind says 0.15 vs answer 0.50 (edge -0.35 >= 0.3): Base rate: an arbitrary male artist has roughly a 3% chance of occupying one of three slots among about 100 plausible favorites; PARTYNEXTDOOR’s current visibility raises this to ~15%, the rules are mildly ambiguous because hints and ad-hoc eliminations affect trading, no event before the 2026-09-16 close materially changes the odds, the resolver is the creator’s market determination, and no matching witness prices are available.
09-09 01:43v86Guess my 3 favourite male music artists! :: DrakeNO 39cancelledmodel 0.18; edge -0.32; luna-medium price-blind says 0.18 vs answer 0.50 (edge -0.32 >= 0.3): Base-rate prior: roughly 10% of broadly popular male artists appear in a listener's top three, raised to 18% for Drake's exceptional popularity; the rules permit ad-hoc elimination and subjective resolution, creating ambiguity that pulls toward 0.5, no scheduled event bears on it before September 16, and the exact resolver is the creator's final market decision rather than an independent data source, with no matching witness market available.
09-09 01:43v86Guess my 3 favourite male music artists! :: Post MaloneNO 49cancelledmodel 0.12; edge -0.38; luna-medium price-blind says 0.12 vs answer 0.50 (edge -0.38 >= 0.3): Base rate: a randomly proposed male artist appears in someone’s top three roughly 5–10% of the time; Post Malone’s popularity raises this to 12%, with no scheduled event, no independent resolver, and author-controlled hints or eliminations creating substantial uncertainty.
09-09 01:43v86Guess my 3 favourite male music artists! :: KhalidNO 45cancelledmodel 0.14; edge -0.36; luna-medium price-blind says 0.14 vs answer 0.50 (edge -0.36 >= 0.3): Base rate: roughly 10% for any one named mainstream male artist to be in a user's top three; the independent resolution rule is clear, but ad-hoc eliminations and future hints add ambiguity, no scheduled event or matching witness exists, and the exact resolver is the creator's final adjudication, so Khalid's popularity supports only a modest uplift.
09-09 01:39v86How much stocks will trump tell us to buy in the next week :: 0 stocksYES 20settled -20model 0.56; edge +0.49; luna-medium price-blind says 0.56 vs answer 0.07 (edge +0.49 >= 0.1): Base rate: roughly 70% of one-week windows likely contain no explicit Trump stock-buy instruction, but recent 2026 endorsements raise this week's nonzero risk; rules and exact resolver are unspecified, so ambiguity pulls toward 0.5, with PPI (Sep 10), CPI (Sep 11), and Trump's 9/11 remarks as possible catalysts, while the likely resolver is an unspecified official Trump statement/social-media record rather than a named source ([BLS](https://www.bls.gov/ppi/), [White House](https://www.whitehouse.gov/news/)).
09-09 01:39v86How much stocks will trump tell us to buy in the next week :: 9 stocksYES 9settled -9model 0.28; edge +0.21; luna-medium price-blind says 0.28 vs answer 0.07 (edge +0.21 >= 0.1): Base rate is roughly 0.15–0.20 public buy prompts per week, making exactly 9 unlikely; however, absent rules leave counting and the resolver (likely Truth Social or official transcripts) highly ambiguous, with the Sept. 9–10 convention and Sept. 11 events providing opportunities.
09-09 01:39v86How much stocks will trump tell us to buy in the next week :: More than 20 stocksYES 9settled -9model 0.28; edge +0.21; luna-medium price-blind says 0.28 vs answer 0.07 (edge +0.21 >= 0.1): Base rate: 0 of roughly 90 recent weeks featured Trump telling the public to buy more than 20 named stocks; recent examples were one or a few names, but absent resolution rules make “buy stocks” potentially count as a broad-market instruction, pulling the estimate toward 0.5, with CPI on Sep. 11 and the FOMC meeting Sep. 15–16 as catalysts; the resolver should be Trump’s official Truth Social/White House remarks, not proxies. [AP](https://apnews.com/article/b00b386495174664553c5b85f8fcfcc5) [Federal Reserve](https://www.federalreserve.gov/monetarypolicy.htm) [BLS](https://www.bls.gov/schedule/news_release/cpi.htm)
09-09 01:39v86How much stocks will trump tell us to buy in the next week :: 4 stocksYES 6settled -6model 0.20; edge +0.13; luna-medium price-blind says 0.20 vs answer 0.07 (edge +0.13 >= 0.1): The historical rate of roughly 0.85 named-company mentions per week makes exactly four explicit buy recommendations unlikely, and the absent resolution rules create substantial ambiguity.
09-09 00:20v86What [mathematical statement] will happen by [date]? :: Yang-Mills existence and mass gap solved by May 2027NO 33open +1model 0.00; edge -0.50; luna-medium price-blind says 0.00 vs answer 0.50 (edge -0.50 >= 0.3): Base rate: only 1 of 7 Millennium problems (14.3%) has been solved in roughly 26 years, or about 0.55% annually; Clay still says no Yang–Mills proof is known, online proof claims are not authoritative, no scheduled event mechanically changes the odds before May 2027, the resolver is the Clay Mathematics Institute’s official problem/status announcement, no matching witness market is available, and the May-versus-June deadline conflict adds ambiguity, so I estimate 0.3%.
09-09 00:20v86What [mathematical statement] will happen by [date]? :: >=6 Millennium by EoY 2027NO 0unfilledmodel 0.00; edge -0.50; luna-medium price-blind says 0.00 vs answer 0.50 (edge -0.50 >= 0.3): Base rate: only 1 of 7 Millennium Problems has been resolved since 2000; reaching at least 6 by EoY 2027 requires 5 of the 6 remaining problems in about 16 months, with no scheduled event mechanically implying that outcome, and Clay’s official problem pages/rules are the resolver; the wording is mildly ambiguous, but not enough to offset the extreme base rate.
09-09 00:20v86What [mathematical statement] will happen by [date]? :: >=4 Millenniums by Feb 2027NO 53open -1model 0.00; edge -0.50; luna-medium price-blind says 0.00 vs answer 0.50 (edge -0.50 >= 0.3): Base rate is 1 of 7 Millennium Problems solved since 2000 (zero additional since 2003); under the natural reading, three more would need solutions by February 2027, with no scheduled catalyst, and the [Clay status page](https://www.claymath.org/millennium-problems/) is only a proxy because the market names no exact resolver and permits proofs, disproofs, or independence results.
09-09 00:20v86What [mathematical statement] will happen by [date]? :: The entire pre-2026 mathematical edifice is consideredNO 0unfilledmodel 0.00; edge -0.50; luna-medium price-blind says 0.00 vs answer 0.50 (edge -0.50 >= 0.3): The base rate is ~14.3% for a major Millennium problem over 26 years, while this undefined all-mathematics claim has no exact resolver, conflicts with the 2029-01-01 close versus EoY 2029, and lacks a mechanically relevant scheduled event.
09-08 23:14v86All Millennium Prize Problems solved before May 2027?NO 100open +13model 0.00; edge -0.28; luna-medium price-blind says 0.00 vs market 0.28 (edge -0.28 >= 0.2): Only one problem has been solved in 26 years, while six would need qualifying CMI-recognized resolutions within eight months under a stringent two-year acceptance rule.
09-08 23:03v86All Millennium Prize Problem solved before 2030?NO 100open +34model 0.00; edge -0.50; luna-medium price-blind says 0.00 vs market 0.50 (edge -0.50 >= 0.2): Only one of seven problems has been solved since 2000, and six exceptionally difficult problems would need broadly accepted solutions within 3.3 years; the unverified Navier–Stokes claim does not materially overcome that base rate.
09-08 22:48v86Anthropic Claims to Solve a Millennium Prize Problem before 2027?NO 42open +4model 0.12; edge -0.25; luna-medium price-blind says 0.12 vs market 0.37 (edge -0.25 >= 0.2): Base rate is 0 qualifying Anthropic Millennium-Problem claims in roughly 25 years; Anthropic’s official record says Claude failed at Riemann while only improving a related bound, no relevant scheduled event is known before Dec. 31, and the underspecified meaning of “claims” and resolver adds ambiguity, so I modestly raise the base rate toward 0.5 ([Anthropic](https://www.anthropic.com/research/riemann-zeta), [Clay](https://www.claymath.org/millennium-problems/)).
09-08 22:47v86OpenAI Claims to Solve another Millennium Prize Problem before 2027?YES 53open +5model 0.90; edge +0.40; luna-medium price-blind says 0.90 vs market 0.50 (edge +0.40 >= 0.2): Base rate: 0 of the 6 remaining Millennium problems had been solved by OpenAI before today; however, OpenAI’s official Sep. 8 publication says its model proposed a Navier–Stokes solution, and Axios reports OpenAI claimed it solved that problem, so the literal claim condition is already met, though absent rules create ambiguity versus Clay’s stricter two-year-validation standard and no additional scheduled event is needed ([OpenAI](https://openai.com/research/index/publication/), [Clay rules](https://www.claymath.org/millennium-problems/rules/)).
09-08 22:39v86Which AI lab's model will be first to solve a Millennium Problem besides Navier–Stokes? :: xAINO 100open +17model 0.01; edge -0.19; luna-medium price-blind says 0.01 vs answer 0.20 (edge -0.19 >= 0.1): Base rate: 1 of 7 Millennium Problems has been solved since 2000 (14%), with 0 AI-led solutions; Clay’s rules make its announcement or broad expert consensus decisive, but “majority of the work” is attributionally ambiguous; no scheduled Clay event mechanically bears on the outcome before 2034-12-31; the exact resolver is the Clay Institute announcement, while current proof benchmarks are only proxies and place Grok 4.1 well behind leading models ([Clay](https://www.claymath.org/millennium-problems/), [rules](https://www.claymath.org/millennium-problems/rules/), [IMO-Bench](https://imobench.github.io/)); no matching witness market was found, so I estimate xAI’s share of the already-low AI-led-solution probability at about 1%.
09-08 22:39v86Which AI lab's model will be first to solve a Millennium Problem besides Navier–Stokes? :: OpenAINO 75open -26model 0.08; edge -0.12; luna-medium price-blind says 0.08 vs answer 0.20 (edge -0.12 >= 0.1): A second qualifying Millennium solution by 2034 is historically uncommon and an AI-majority proof is unprecedented; OpenAI has a plausible but non-dominant chance of being the first lab, with attribution and two-year acceptance rules adding uncertainty.
09-08 22:39v86Which AI lab's model will be first to solve a Millennium Problem besides Navier–Stokes? :: AnthropicNO 74open -10model 0.08; edge -0.12; luna-medium price-blind says 0.08 vs answer 0.20 (edge -0.12 >= 0.1): Base rate: 0 of 6 non-Poincaré Millennium Problems has been solved since 2003; applying the strict majority-of-work and Clay/expert-confirmation rules, I estimate about 30% for any qualifying second AI-led solution by 2034 and 27% conditional Anthropic leadership, with no identified scheduled catalyst, only weak Anthropic evidence, mismatched related-market witnesses, and [Clay Institute](https://www.claymath.org/millennium-problems/) as the resolver.
09-08 22:35v86Next unique country Tetra visits after 10th September 2026 :: 🇮🇪 IrelandYES 65open -6model 0.50; edge +0.42; luna-medium price-blind says 0.50 vs answer 0.08 (edge +0.42 >= 0.1): Base rate unavailable for Tetra’s private travel (use 1/number of plausible next countries); the title’s 10 September cutoff conflicts with rules keyed to market creation, no scheduled event or public resolver evidence is available, and resolution depends on Tetra’s actual terminal-exit visit recorded by Tetra.
09-08 22:35v86Next unique country Tetra visits after 10th September 2026 :: 🇪🇸 SpainYES 65open -9model 0.50; edge +0.42; luna-medium price-blind says 0.50 vs answer 0.08 (edge +0.42 >= 0.1): Base rate is unknown for this individual (generic prior roughly 5–10% for Spain as the next unique country), and the conflict between “after 10 September” and “after market creation” plus no specified authoritative resolver makes the estimate highly ambiguous, so it is pulled toward 0.5.
09-08 22:35v86Next unique country Tetra visits after 10th September 2026 :: 🇯🇪 JerseyYES 51open -6model 0.40; edge +0.32; luna-medium price-blind says 0.40 vs answer 0.08 (edge +0.32 >= 0.1): Base rate: Jersey was 0/1 in Tetra’s prior observed annual travel record, while 7/20 listed countries were visited; the matching prior next-country markets priced Jersey near 4%, but the 10 September Jersey Air Display is a scheduled catalyst, with resolution by Tetra’s own visit record under the terminal-exit rule, so sparse evidence warrants a near-0.5 estimate.
09-08 22:35v86Next unique country Tetra visits after 10th September 2026 :: 🇬🇮 GibraltarYES 77open -10model 0.58; edge +0.50; luna-medium price-blind says 0.58 vs answer 0.08 (edge +0.50 >= 0.1): Base rate is 0/2 prior Tetra next-country resolutions for Gibraltar (tiny sample); Gibraltar is ISO GI and a scheduled 10 September port call mechanically supports it, but the call schedule is only a proxy for Tetra’s actual terminal exit, with no matching market witness and some residual rule/itinerary ambiguity.
09-08 22:27v86Will OpenAI release a new model before October?YES 73open +39model 1.00; edge +0.50; luna-medium price-blind says 1.00 vs market 0.50 (edge +0.50 >= 0.2): Base rate: OpenAI released qualifying models in at least 4 of the first 8 months of 2026; rules are broad and GPT-6 Astra plainly qualifies, with the official resolver recording its release on September 3, before the September 30 close, so no future calendar event is needed.[Model release notes](https://help.openai.com/en/articles/9624314-model-release-notes.csv) [GPT-6 Astra announcement](https://openai.com/index/safety-overview-gpt-6-astra/)
09-08 22:11v86Do you think collage is a scam?NO 26settled +17model 0.24; edge -0.22; luna-medium price-blind says 0.24 vs market 0.46 (edge -0.22 >= 0.2): Base rate: roughly 30% of Americans express strongly negative views of college value; the clarified community poll makes neutral votes count as NO, Manifold users are likely somewhat more skeptical but the loaded “scam” wording and fallback ambiguity pull toward 0.5, with no scheduled event before September 15 and the exact resolver being the Manifold poll/comments rather than public surveys.
09-08 22:04v86Which Millennium Prize Problem will be solved next? :: Yang–Mills existence and mass gapNO 100open +10model 0.01; edge -0.18; luna-medium price-blind says 0.01 vs answer 0.20 (edge -0.18 >= 0.1): Base rate: only 1 of 7 Millennium problems has been officially resolved in 26 years (3.8% per year); Clay requires publication, two years, and broad acceptance, so today’s OpenAI Navier–Stokes proposal is not yet dispositive, no scheduled event mechanically resolves this before 2098, and the exact resolver is Clay’s official status/prize announcement (with a matching Manifold witness near 0.4%), leaving Yang–Mills unlikely to be next despite ambiguity in the unspecified rules.
09-08 21:55v86Will *all* Millennium Problems be solved by EOY 2027?NO 0open +0model 0.00; edge -0.50; luna-medium price-blind says 0.00 vs market 0.50 (edge -0.50 >= 0.2): Base rate is 1 of 7 solved in 26 years (0 of the six remaining); despite the AI Navier–Stokes claim, six extraordinarily hard problems in 15 months remains unlikely, no scheduled event mechanically changes this, and the resolver is ambiguous between Clay’s status page and market adjudication ([Clay](https://www.claymath.org/millennium-problems/), [OpenAI](https://openai.com/index/navier-stokes-solution/)).
09-08 21:32v86What number model will SpaceXAI be on by EOY?YES 68open +6model 0.98; edge +0.48; luna-medium price-blind says 0.98 vs market 0.50 (edge +0.48 >= 0.2): GPT-6 Astra is already publicly announced and scheduled to reach the $20 ChatGPT Plus plan, so the market should resolve YES barring an unusually broad rollout reversal.
09-08 20:35v86Which Millennium prize problem will be solved next? (after Navier-Stokes) :: Riemann HypothesisNO 100open +20model 0.01; edge -0.19; luna-medium price-blind says 0.01 vs answer 0.20 (edge -0.19 >= 0.1): Base rate: only 1 of 7 Millennium problems has been solved since 2000 (0 of 6 remaining in over 23 years); CMI still lists RH as unsolved, its April 15, 2026 lecture was informational rather than a resolution, and CMI’s solved archive is the resolver, so I assign only 1% by the 2027-12-31 close.
09-08 20:35v86Which Millennium prize problem will be solved next? (after Navier-Stokes) :: P vs NPNO 2open +0model 0.04; edge -0.16; luna-medium price-blind says 0.04 vs answer 0.20 (edge -0.16 >= 0.1): Base rate is 1 solved Millennium problem in 26 years (~0.038 per year); despite the Navier–Stokes catalyst, P vs NP faces deep known barriers, no scheduled resolving event exists before 2027-12-31, and the resolver is broad mathematical consensus (not merely CMI publication), with the limited prediction-market proxy near 5% for a 2027 P≠NP proof.
09-08 19:44v86Will a second Mersenne number above 2^1500 be fully factored in 2026?NO 6open -1model 0.28; edge -0.32; luna-medium price-blind says 0.28 vs market 0.60 (edge -0.32 >= 0.2): One qualifying event in 251 days annualizes to roughly a 36% chance in the remaining 114 days, but the sparse historical record, lack of a scheduled catalyst, and partial-factor rather than full-factor evidence justify a lower estimate.
09-08 19:10v86Who will win the next Super Ballon d'Or? :: Cristiano RonaldoNO 17open +2model 0.06; edge -0.19; luna-medium price-blind says 0.06 vs answer 0.25 (edge -0.19 >= 0.1): Base rate is 1 Super Ballon d'Or in roughly 95 years (1.1% annually); no official 2029 award is scheduled, so N/A is likely, and if it occurs Messi appears the stronger candidate than Ronaldo—resolver: France Football/UEFA/L'Équipe announcements, with only the regular 2026 ceremony currently confirmed ([UEFA](https://www.uefa.com/ballondor/news/02a5-20bba5196317-e0e34001e13b-1000--2026-ballon-d-or-ceremony-date-and-host-city-announced/)).
09-08 19:10v86Who will win the next Super Ballon d'Or? :: Lionel MessiNO 11open -2model 0.12; edge -0.13; luna-medium price-blind says 0.12 vs answer 0.25 (edge -0.13 >= 0.1): Base rate is 1 Super Ballon d'Or in roughly 70 years (1.4% annually); with no official France Football schedule, only an unconfirmed 2029 rumor, and the only relevant calendar event being the regular Ballon d'Or on 2026-10-26, I estimate about a 20% chance of any qualifying award by 2030 and roughly 60% conditional chance Messi wins it, with ballondor.com/France Football as the resolver.
09-08 19:10v86Who will win the next Super Ballon d'Or? :: Kylian MbappéNO 21open +4model 0.02; edge -0.23; luna-medium price-blind says 0.02 vs answer 0.25 (edge -0.23 >= 0.1): The award has only occurred once, has no official confirmation or scheduled date, and even if revived Mbappé would likely trail Messi/Ronaldo under the most plausible long-horizon criteria.
09-08 19:01v86The shame market: Who will be the worst bot in Season 41? :: @NiciusBotNO 0unfilledmodel 0.02; edge -0.48; luna-medium price-blind says 0.02 vs answer 0.50 (edge -0.48 >= 0.3): Base rate is 1/35 = 2.9% for an individual bot in the 35-answer field, and NiciusBot's established active trading record suggests below-average odds of finishing worst; the October 1, 2026 season close will be settled by Manifold's final Silicon Prophetic Programs profit data, with little rule ambiguity.
09-08 19:01v86The shame market: Who will be the worst bot in Season 41? :: @EvansbotNO 0unfilledmodel 0.03; edge -0.47; luna-medium price-blind says 0.03 vs answer 0.50 (edge -0.47 >= 0.3): Base rate is 1/35 = 2.86%; the rules are clear, Season 41 ends at the scheduled October 1 07:00 UTC league cutoff, and Evansbot appears active and technically sophisticated, so I discount it modestly below the field-average chance.
09-08 19:01v86The shame market: Who will be the worst bot in Season 41? :: @YunaNO 0unfilledmodel 0.01; edge -0.49; luna-medium price-blind says 0.01 vs answer 0.50 (edge -0.49 >= 0.3): Base rate is 1/36 = 2.8% for a named answer among 35 bots plus Other, but Yuna’s strong historical profitability lowers this to 1%; the precise rules are clear, the decisive event is final Season 41 data after the October 1 close, and the resolver is Manifold’s final Prophetic Programs profit field.
09-08 19:01v86The shame market: Who will be the worst bot in Season 41? :: @TestBotPleaseIgnoreNO 0unfilledmodel 0.01; edge -0.49; luna-medium price-blind says 0.01 vs answer 0.50 (edge -0.49 >= 0.3): Base rate is 1/35 = 2.9%; the exact rules require the lowest final profit, and this apparently inactive test bot is less likely than active bots to finish negative, with October 1 final Manifold league data as resolver and no rule-matched witness market.
09-08 17:49v86Will Eliezer Yudkowsky be in ARTIFICIAL, the movie about OpenAI and the firing and rehiring of Sam AltmanNO 100open +11model 0.08; edge -0.42; luna-medium price-blind says 0.08 vs market 0.50 (edge -0.42 >= 0.2): Base rate: roughly 10% for an unannounced peripheral real person to appear in a nearly completed biographical drama; the broad rules modestly raise this, but the announced cast/credits omit Yudkowsky and the film focuses narrowly on the 2023 firing, so sentiment is weak; the scheduled Christmas Day 2026 release occurs before close, with the released film and final credits as resolver, while current IMDb/press cast lists are only proxies; no matching related-market witness found ([IMDb credits](https://www.imdb.com/it/title/tt37171180/fullcredits/), [Guardian teaser report](https://www.theguardian.com/film/2026/sep/08/andrew-garfield-sam-altman-artificial-movie-openai)).
09-08 16:20v86Will a new Smash Bros game be announced in the Nintendo Direct on 9/9?NO 100settled +76model 0.20; edge -0.30; luna-medium price-blind says 0.20 vs market 0.50 (edge -0.30 >= 0.2): Base rate: only about 3 major new Smash entries have been revealed in roughly 15 years of Nintendo Direct-era showcases (about 1–2% per random Direct); the scheduled 45-minute Nintendo Direct on September 9 is the key calendar event, but the anonymous malo932 leak only claims development—not announcement timing—so I raise the prior modestly while discounting fan hype, with Nintendo’s official YouTube/site as the resolver and no matching market witness found.
09-08 15:42v86If Saxony-Anhalt holds a snap election before 2028, will the AfD win an absolute majority?NO 100open -6model 0.06; edge -0.43; luna-medium price-blind says 0.06 vs market 0.49 (edge -0.43 >= 0.2): Base rate: AfD absolute majorities in German state elections are 0 to date; rules make this a joint probability including a snap election, and the 2026 result was only 39/83 seats versus 42 needed [official results](https://wahlergebnisse.sachsen-anhalt.de/wahlen/lt26/sitze.html); despite the hung-parliament narrative, a snap is unlikely because the regular term is five years [constitution](https://www.landtag.sachsen-anhalt.de/fileadmin/Downloads/Broschueren/Verfassung_und_Grundgesetz_web_2025_10.pdf) and dissolution requires difficult parliamentary routes, so I estimate roughly 12% for a snap by 2028 times 50% for an AfD majority, with no scheduled qualifying event and the official state election result as resolver [rules summary](https://www.tagesschau.de/inland/innenpolitik/landtagswahl-sachsen-anhalt-faq-100.html).
09-08 15:23v86Who gets the Millennium Prize for solving Navier-Stokes? :: Tristan BuckmasterNO 100open +20model 0.01; edge -0.24; luna-medium price-blind says 0.01 vs answer 0.25 (edge -0.24 >= 0.1): Navier–Stokes has had a 0% historical solve rate over 26 years, and although Buckmaster is unusually well-positioned, the short qualifying window and absence of an official breakthrough make his chance of receiving any Millennium Prize share about 1%.
09-08 15:23v86Who gets the Millennium Prize for solving Navier-Stokes? :: Levent AlpögeNO 100open +20model 0.07; edge -0.18; luna-medium price-blind says 0.07 vs answer 0.25 (edge -0.18 >= 0.1): Base rate is 0 solved remaining Millennium Problems, while current Alpöge-related results are promising but not the official Navier–Stokes solution and Clay’s two-year publication-and-acceptance rule leaves limited time; estimated expected prize share is 12% × 55% = 6.6%.
09-08 15:23v86Who gets the Millennium Prize for solving Navier-Stokes? :: Nobody (% of prize not given out at all)YES 100open +46model 0.94; edge +0.69; luna-medium price-blind says 0.94 vs answer 0.25 (edge +0.69 >= 0.1): Historically no unsolved Millennium Problem has been resolved since 2000, and Navier–Stokes would additionally require Clay’s publication, two-year, and acceptance hurdles before 2030.
09-08 11:22v86Will Speaker Mike Johnson send the House members home early before midterms?YES 100settled +55model 0.98; edge +0.48; luna-medium price-blind says 0.98 vs market 0.50 (edge +0.48 >= 0.2): Base rate: 3 documented early House send-homes in 2026 before September 8 (3/3); the rules are unspecified, but contemporaneous reporting explicitly says Johnson canceled September 21 and 28, sending members home early ahead of the November 3 midterms, with only a possible recall as ambiguity; the official House calendar has no October session and the resolver is the House Clerk/Majority Leader schedule, corroborated by Bloomberg and The Washington Post ([official calendar](https://www.majorityleader.gov/uploadedfiles/119_legislative_schedule_2026_houseofrepresentatives.pdf), [Bloomberg](https://news.bloomberglaw.com/artificial-intelligence/house-cancels-two-session-weeks-in-september-ahead-of-midterms), [Washington Post](https://www.washingtonpost.com/politics/2026/09/03/gop-led-house-scraps-two-weeks-work-ahead-of-midterms/)).
09-08 09:10v86US Average Gas Price Is $4.100 or more on September 14 2026?YES 100settled +53model 0.82; edge +0.22; luna-medium price-blind says 0.82 vs market 0.60 (edge +0.22 >= 0.2): AAA is already at $4.1514, and although normal post-Labor-Day seasonality favors declines, six days of modest easing likely leaves the average at or above $4.10 absent a sharp oil-market reversal.
09-08 08:21v86[Precise] What is the exact time of sunset in Ho Chi Minh City on September 8, 2026? :: 11:05 UTC – 11:09 UTCNO 0unfilledmodel 0.00; edge -0.20; luna-medium price-blind says 0.00 vs answer 0.20 (edge -0.20 >= 0.1): Base rate is 0/8 nearby dates in the 11:05–11:09 UTC window, with sunset drifting only about 35 seconds/day; the scheduled sunset is around 10:58–10:59 UTC local 17:58–17:59, and the priority resolver is [TimeAndDate](https://www.timeanddate.com/sun/vietnam/ho-chi-minh), supported by [Time.now](https://time.now/ho-chi-minh-city/prayer-times/), so the rules’ consensus points well outside the YES interval.
09-08 08:21v86[Precise] What is the exact time of sunset in Ho Chi Minh City on September 8, 2026? :: 10:55 UTC – 10:59YES 10settled +10model 0.98; edge +0.78; luna-medium price-blind says 0.98 vs answer 0.20 (edge +0.78 >= 0.1): TimeAndDate's 17:58 ICT sunset converts to 10:58 UTC, squarely within 10:55–10:59; only unlikely cross-source or rounding discrepancies could move the resolved minute outside the band.
09-08 08:21v86[Precise] What is the exact time of sunset in Ho Chi Minh City on September 8, 2026? :: 11:00 UTC – 11:04 UTCNO 0unfilledmodel 0.00; edge -0.20; luna-medium price-blind says 0.00 vs answer 0.20 (edge -0.20 >= 0.1): TimeandDate’s date-specific result is 17:58 ICT = 10:58 UTC, two minutes before the 11:00–11:04 UTC interval, leaving only a small chance of source or resolution discrepancy.
09-08 08:21v86[Precise] What is the exact time of sunset in Ho Chi Minh City on September 8, 2026? :: Before 10:55 UTCNO 0unfilledmodel 0.00; edge -0.20; luna-medium price-blind says 0.00 vs answer 0.20 (edge -0.20 >= 0.1): The named resolver sources place sunset at roughly 10:58–10:59 UTC, well after the strict 10:55 UTC cutoff, so YES requires an unlikely source or resolution error.
09-08 07:27v86Will the Magic the Gathering movie get over 85% on Rotten Tomatoes? Critics score.NO 0unfilledmodel 0.27; edge -0.23; luna-medium price-blind says 0.27 vs market 0.50 (edge -0.23 >= 0.2): The film’s conditional chance of exceeding 85% is roughly 35%, but development and release uncertainty plus ambiguous non-release handling reduce the market-level probability to 27%.
09-08 06:04v86Did OpenAI just solve navier stokes??NO 100settled -100model 0.18; edge -0.32; luna-medium price-blind says 0.18 vs market 0.50 (edge -0.32 >= 0.2): Base rate is 0% for an accepted full 3D Navier–Stokes resolution since the problem’s formulation; the wording is ambiguous, but the resolver is Tristan Buckmaster’s Mastodon post, which reports an unverified OpenAI claim, while Clay still treats the problem as open and Tao says he knows of no significant development, with no scheduled confirming event before Oct. 8; Reddit discussion is only a sentiment proxy, so I discount it heavily. [Clay](https://www.claymath.org/millennium/Navier-Stokes-Equation/) [Buckmaster](https://mastodon.social/@tristanbuckmaster/117233413705701198) [Tao clarification](https://mathstodon.xyz/@tao/117233527638291447)
09-08 06:00v86Will California elect a Republican governor in 2026?NO 100open +85model 0.06; edge -0.44; luna-medium price-blind says 0.06 vs market 0.50 (edge -0.44 >= 0.2): The base rate is 1 of California’s last 5 regular gubernatorial elections (20%) Republican, but the rules are clear, the June primary produced Democrat Xavier Becerra versus Republican Steve Hilton, Becerra leads the latest poll 55%-37%, and Polymarket prices Becerra at 94%; the key scheduled event is the November 3 election, settled by certified California Secretary of State results, so I discount the historical rate and sentiment in favor of a 6% Republican probability. [SOS](https://www.sos.ca.gov/elections/prior-elections/statewide-election-results/primary-election-june-2-2026) [poll](https://www.latimes.com/california/story/2026-08-13/becerra-holds-strong-lead-over-hilton-in-race-for-california-governor-poll) [market](https://www.realclearpolling.com/polls/governor/general/2026/california/hilton-vs-becerra)
09-08 03:32v86Who will get an In Person Hack Club Summer Internship in 2027? :: kcNO 100open -3model 0.10; edge -0.38; luna-medium price-blind says 0.10 vs answer 0.48 (edge -0.38 >= 0.3): Base rate: Hack Club reported 30 interns in 2024, but no candidate-specific rate is published; kc has no public offer evidence, the rules require both acceptance and HQ arrival, the relevant calendar event is the 2027 summer internship cycle before August 31, and the resolver would be Hack Club’s official HQ/jobs announcements, so I estimate 10%.
09-08 03:20v86Will the Republican Party candidate win the 2028 US Senate election in Utah?YES 49open +18model 0.88; edge +0.38; luna-medium price-blind says 0.88 vs market 0.50 (edge +0.38 >= 0.2): Base rate: Republicans won the last 9 Utah Class III Senate general elections (9/9); with Mike Lee’s Republican-held seat, the 2028 general election on November 7 is the key scheduled event, and the Utah Lieutenant Governor’s certified statewide canvass is the resolver, though absent rules and unknown candidates add uncertainty.
09-08 02:40v86[Short Fuse] Will anyone help me find the Metaculus Fall Cup IDYES 39cancelledmodel 0.82; edge +0.32; luna-medium price-blind says 0.82 vs market 0.50 (edge +0.32 >= 0.2): Base rate for a niche public help request receiving a reply within one hour is roughly 35%; rules are clear, the official Metaculus tournament page exposes the slug `metaculus-cup-fall-2026`, no scheduled event is needed, and the market’s comment feed is the resolver, so the easy-to-verify answer and active community raise this substantially.
09-08 01:16v869984 reaches 7000 by Sep 10?NO 54settled +20model 0.07; edge -0.27; luna-medium price-blind says 0.07 vs market 0.34 (edge -0.27 >= 0.2): Base rate: recent daily volatility is about 5%, making a 12.6% move from 6,217 to 7,000 within three sessions an estimated ~7% event; rules are ambiguous because no resolver/source or intraday-versus-close convention is specified, so I assume the TSE 9984 intraday high, with no company catalyst before the Sep 9 close and only routine Japanese data scheduled, while the official listing is confirmed by [SoftBank IR](https://group.softbank/en/ir/stock/info) and prices are proxied by [Yahoo Finance](https://finance.yahoo.co.jp/quote/9984.T/history).
09-08 01:09v869984 reaches 6800 by Sep 10?NO 0unfilledmodel 0.28; edge -0.22; luna-medium price-blind says 0.28 vs market 0.50 (edge -0.22 >= 0.2): Base rate: 2 of the 14 recent sessions shown exceeded a +9.4% intraday gain (~14%), implying roughly 30–35% over three sessions; 9984 closed Sep 7 at ¥6,217, so ¥6,800 requires +9.4%, with no scheduled SoftBank catalyst before Sep 10, and the exact resolver is TSE/SoftBank’s official 9984 price high (Yahoo is only a proxy).
09-07 23:26v86World Athletics Ultimate Championship: who will win women’s 200m? :: Jefferson-WoodenYES 0unfilledmodel 0.29; edge +0.22; luna-medium price-blind says 0.29 vs answer 0.07 (edge +0.22 >= 0.1): Base rate is 6.25% for a 16-athlete field and 25% (1/4) for the reigning women’s 200m world champion to repeat at the next completed global championship; Jefferson-Wooden is the 2025 world champion and 2026 US champion (21.69), but Thomas, Hodge and Alfred have comparable or faster 2026 form, with the final scheduled for 2026-09-13, resolved by the official World Athletics result, and the absent description creates modest rules ambiguity that pulls toward 0.5.
09-07 23:26v86World Athletics Ultimate Championship: who will win women’s 200m? :: AlfredYES 23settled -23model 0.64; edge +0.57; luna-medium price-blind says 0.64 vs answer 0.07 (edge +0.57 >= 0.1): Base rate: Alfred’s documented 2026 200m win rate is 100% (3/3 major Diamond League races); rules have no material ambiguity, she leads the season at 21.51m and beat key rivals recently, while the women’s 200m final is scheduled for September 13 and the exact resolver is World Athletics’ official Budapest 26 results page, so I allow for championship-race variance and the strong field. [World Athletics](https://worldathletics.org/competitions/world-athletics-ultimate-championship/2026) [Diamond League](https://www.diamondleague.com/athlete/14646562/)

Showing the newest 80 of 365 trades.

365 trades, 11,000 staked, realised +1576, open +1819 if held / +868 if sold, 88 unfilled.

In the traces on hand the model priced 0 answers across 0 markets and the module acted on 0 of them (0%). The journal holds 365 trades for 11,000 mana. 62 of them settled for +1576. An estimate the model declines to give is treated as no opinion and never as a cheap one.
Traces are diagnostic and rotate; the trading journal is the book above. Built 2026-09-22 17:56Z.