The Oracle

Ask the Oracle.

Ask any resolvable question about a future event and get back a calibrated probability — not a guess, a number you can hold to account. Every answer is reasoned by an ensemble of eight frontier models, weighed by a supervising model, then published publicly and stamped to the Bitcoin blockchain so it can never be quietly changed after the fact.

$5 per question · every answer public · every answer timestamped

The Oracle reads your question first and helps sharpen it before answering.

How to ask a good question

The Oracle answers questions that reality can settle. Three things make one resolvable:

  • A clear event. One specific thing that either happens or doesn’t — no vibes, no “will AI change everything.”
  • Explicit resolution criteria. Say exactly what counts as yes. Name the source that decides it if there’s any doubt.
  • A date. Give the question a deadline, so the answer can be scored when the day comes.

Good: “Will SpaceX land humans on the Moon before 1 Jan 2028, per an official NASA or SpaceX announcement?”

Proof of accuracy

The Oracle has a track record you can check. These numbers come straight from the resolved record below — they move honestly as questions settle.

33/41
Resolved correct
80.5%
Accuracy
0.1343
Brier score
<0.001
p-value

How it works · Full track record

The public record

Every question the Oracle has answered on Polymarket markets — open bets first, then everything that has since resolved. Nothing is ever taken down.

2026-07-14

US recognizes Russian sovereignty over Ukraine before 2027?

Trimmed mean (8%) and market (11%) differ by only 3 points — below the 5% trade threshold, so protocol dictates SKIP. The consensus factual picture is solid and undisputed: no qualifying act has occurred, negotiations are stalled, the latest US proposal offered only non-qualifying 'de facto' recognition, and the historical base rate is near zero. However, a genuine tail path...

Oracle: 8.0% Market: 11.0% Polymarket ↗

Read analysis

2026-07-14

Will “DTF St. Louis” win Emmys 2026: Outstanding limited or anthology series?

Factual consensus is airtight and freshly verified: DTF is the clear #2 at ~34% per Gold Derby's combined odds (post-nominations, July 9), behind Beef S2 at ~60%, with all major pundits agreeing. I adjust the ensemble's 26% UP to 35% because agents double-discounted below the best public aggregator and the fresh-breakout-beats-sequel pattern has some validity (consistent with our known bias...

Oracle: 35.0% Market: 55.0% Polymarket ↗

Read analysis

2026-07-14

Will Republicans hold 50–52 Senate seats and 223+ House seats after the 2026 midterms?

This is an elections-category market (our 71% hit-rate segment) with genuine agent consensus: trimmed mean 6.3%, full range 2–19%, all coherent. The required outcome is a demanding conjunction — GOP must gain House seats in a D+6 midterm environment (defying one of the most robust patterns in US politics) while simultaneously losing 1–3 Senate seats, two conditions that pull in...

Oracle: 6.3% Market: 37.6% Polymarket ↗

Read analysis

2026-07-14

Will Republicans hold 53+ Senate seats and ≤207 House seats after the 2026 midterms?

*Note: only 5/8 ensemble models completed on this run.* This is a conjunction market: R Senate ≥53 (zero net losses from exactly 53) AND R House ≤207 (losing 11+ seats). The House leg is likely (~55-65%) given D+6 generic ballot and near-universal midterm penalty, though mid-decade redistricting (TX +5R) compresses Dem gains. The Senate leg is unlikely (~15-25%): NC is...

Oracle: 9.7% Market: 37.9% Polymarket ↗

Read analysis

2026-07-14

Will Rick Brattin be the MO-05 Republican nominee?

After discounting three polarity-error agents (2% submissions contradicting their own pro-Brattin evidence), the real ensemble signal is ~61%, not 40.2%. My independent assessment lands around 68-72%: Brattin is the clear endorsement/institutional frontrunner in a fragmented six-way, low-turnout primary, with CFG air cover offsetting Burks's cash; residual risk comes from no public polling, the new district geography, Burks's late-money blitz capacity,...

Oracle: 40.2% Market: 76.5% Polymarket ↗

Read analysis

2026-07-14

Will South Korea GDP growth in Q2 2026 be between 3.5% and 3.9%?

*Note: only 6/8 ensemble models completed on this run.* The 60-point gap between ensemble (26.8%) and market (87.1%) is driven by two low-side agents anchored to stale annual forecasts and a mismatched q/q base-rate brief. My verification killed their case: Korea's finance ministry just raised its 2026 forecast to 3.0%, ING forecasts Q2 'well above 3%' YoY, CEIC nowcasts 3.885%,...

Oracle: 26.8% Market: 87.1% Polymarket ↗

Read analysis

2026-07-14

Will the Democratic Party win the IN-01 House seat?

Trimmed mean (85.5%) and market (82%) differ by only 3.5% — below the 5% trade threshold. All coherent agents favor a Democratic hold on undisputed facts: century-long Democratic control, incumbent with a growing margin, weak/fractured GOP opposition, dead redistricting threat, and a favorable midterm environment. The counter-case (Rust Belt rightward drift, self-funding challenger, no polling) justifies the market's ~18% NO...

Oracle: 85.5% Market: 82.0% Polymarket ↗

Read analysis

2026-07-06

US recognizes Russian sovereignty over Ukraine before 2027?

All 8 agents agree on direction (well below coin-flip, NO-favored) and on the core facts: (1) the Nov 2025 28-point plan proposed recognition but was never implemented and explicitly downgraded to 'de facto' language, (2) Russia rejected the revised plan in Jan 2026 and the June 2026 deadline passed without a deal, (3) Congress has erected legal barriers, and (4)...

Oracle: 2.8% Market: 8.0% Polymarket ↗

Read analysis

2026-07-06

Will Darijana Filipović be the Croat Presidency Member of Bosnia and Herzegovina following the 2026 general election?

The ensemble mean of 38.5% is a pure averaging artifact. The distribution is bimodal: five agents (2,3,4,5,8) cluster tightly at 59-63%, and three agents (1,6,7) output 1.00%. Critically, the three 1% agents present evidence that CONTRADICTS their own numbers — Agent 1 lists 'HDZ nominated her, opposition fragmented, markets price her as frontrunner' as UP evidence and its confidence is...

Oracle: 66.7% Market: 62.0% Polymarket ↗

Read analysis

2026-07-06

Will Sara Rodriguez win the 2026 Wisconsin Governor Democratic primary election?

The ensemble mean of 25.4% is an averaging artifact, not a genuine signal. Three agents (1, 2, 4) output 1.00% probabilities that flatly contradict their own evidence summaries — each describes Rodriguez as a top-two contender, straw poll winner, and co-frontrunner in a fluid race, which is consistent with 30-45%, not 1%. These appear to be output/calibration errors and should...

Oracle: 33.3% Market: 41.5% Polymarket ↗

Read analysis

2026-07-06

Will Slaven Kovačević be the Croat Presidency Member of Bosnia and Herzegovina following the 2026 general election?

The split is bimodal: five agents cluster tightly at 32-36% (matching the market at 35%), while three agents (1, 4, 7) output 1%. Examining the 1% agents, their evidence and research are nearly identical to the consensus agents — Agent 1 explicitly notes the market prices Kovačević at 35% and Agent 7's evidence-UP list is arguably stronger than its DOWN...

Oracle: 24.7% Market: 35.0% Polymarket ↗

Read analysis

2026-07-06

Will South Korea GDP growth in Q2 2026 be between 3.0% and 3.4%?

*Note: only 6/8 ensemble models completed on this run.* AGREEMENT: All six agents agree on the core facts — Q1 2026 YoY was 3.8% (strongest since 2020), Q2 2026 export data is exceptionally strong (June +70.9% YoY, semiconductors ~+200%), full-year institutional forecasts (BOK/KDI/OECD 2.5-2.6%) imply deceleration, and Trading Economics points to 3.5% for Q2. All six also land BELOW the...

Oracle: 27.4% Market: 6.0% Polymarket ↗

Read analysis

2026-07-06

Will the 2026 United Left primary for the 2027 French presidential election be canceled?

The ensemble mean of 27% is an artifact of averaging: three agents (Grok, Kimi, GLM) output 1%, but their own evidence sections are internally contradictory — Grok lists strong evidence UP (primary 'near KO', PS withdrawing, ~55-70% historical abandonment rate) yet outputs 1%, which strongly suggests a probability-direction error or misreading of the resolution polarity. Kimi and GLM at least...

Oracle: 36.5% Market: 43.5% Polymarket ↗

Read analysis

2026-04-01

AI data center moratorium passed before 2027?

7 of 8 agents with completed research converge on 2-12% probability (mean ~5% excluding the failed agent). The evidence is overwhelming and specific: the only federal vehicle (S.4214) was introduced by minority-party progressives with zero cosponsors into a Republican Congress that is actively pushing the opposite direction. The Trump administration released a framework to preempt state AI restrictions just 5...

Oracle: 2.3% Market: 26.5% Polymarket ↗

Read analysis

2026-04-01

Any Brazil STF Justice removed by impeachment before 2027?

Seven of eight agents converge tightly at 3-12%, with six at 3% and one at 6%. Agent 1 (50%) should be completely disregarded as it explicitly failed to conduct any research (0 searches, 0 pages read). Excluding this outlier, the effective ensemble mean is ~4.7% with very low dispersion. The consensus is overwhelming and well-evidenced: 1. ZERO base rate: No...

Oracle: 2.3% Market: 17.0% Polymarket ↗

Read analysis

2026-03-31

SpaceX Starship fully reusable before 2027?

Seven of 8 agents with actual research agree on the core facts: V3 hasn't flown, tight timeline, multiple sequential successes needed. The well-researched agents cluster 20-42% with median ~32%. Agent 1 (50%) had zero research and is discarded. The critical path requires Flight 12 success → ocean landing → Flight 13 success → ocean landing → catch attempt, all in...

Oracle: 23.5% Market: 39.5% Polymarket ↗

Read analysis

2026-03-31

Will Trump create a tariff dividend by June 30?

Seven of eight agents with substantive research converge strongly on 4-12% probability (Agent 1 at 50% had zero research and should be discarded entirely). The consensus is remarkably tight: Agents 3-8 cluster at 4-8% with high confidence (0.75-0.88). Agent 2 at 12% is the highest informed estimate. Key consensus findings across all informed agents: 1. Supreme Court struck down IEEPA...

Oracle: 3.2% Market: 15.0% Polymarket ↗

Read analysis

2026-03-30

Canada recession before 2027?

All 8 agents converge tightly (38-46%, σ=3.2%) around the market price of 42%. The core analysis is unanimous: Q4 2025 contraction sets up a possible technical recession if Q1 2026 is also negative, but most forecasters expect Q1 to be slightly positive. The Q4 weakness was inventory-driven (domestic demand +2.4%), January 2026 was flat, and major banks project 1-2% growth...

Oracle: 38.1% Market: 42.0% Polymarket ↗

Read analysis

2026-03-30

Kylie Jenner and Timothée Chalamet engaged in 2026?

AGENT AGREEMENT: Six of eight agents cluster between 25-32% (Agents 2,3,4,6,8 at 25-32%, Agent 5 at 42%). Two outliers pull the mean up: Agent 1 (50%) had zero research and should be discarded entirely, and Agent 7 (48%) weighted the insider reports and Chalamet's 'Yes. Yes.' marriage intent statement very heavily. Excluding the zero-research Agent 1, the informed mean drops...

Oracle: 19.9% Market: 36.5% Polymarket ↗

Read analysis

2026-03-30

Will a new country join the Abraham Accords before 2027?

Agent 8 found the strongest specific evidence: Somaliland counts per market clarification and has pledged to join. This creates a clear pathway to YES resolution. However, the edge over market price (69%) is only ~3%, which is below the 5% threshold for trading. The agents are genuinely split (48-88% range, 12.3% std dev) reflecting real uncertainty about whether pledges/announcements will...

Oracle: 82.3% Market: 69.0% Polymarket ↗

Read analysis

2026-03-29

Will the Iranian regime survive U.S. military strikes?

All 8 agents converge on 78-88% with strong consensus on key facts: strikes occurred, Khamenei killed but succession completed, regime still functioning after 4 weeks, no ground invasion planned, historical base rate ~80-85%. The strongest agents (3, 5, 6) found specific US intelligence assessments confirming regime intact. The unprecedented leadership decapitation is the main downside risk but the regime has...

Oracle: 91.3% Market: 80.0% Polymarket ↗

Read analysis

2026-03-29

Will the Republican Party win the VA-06 House seat?

Agents fundamentally disagree on the referendum's likelihood of passing, creating a 50-point range (42%-92%). The high-R agents weight the Roanoke College poll and early voting; the low-R agents weight prediction market prices and Cook's 'Favored to Pass' rating. My calculation using blended referendum probabilities yields 55-66% for R, with the median around 60%. However, the sibling D market at 63%...

Oracle: 58.6% Market: 51.0% Polymarket ↗

Read analysis

2026-03-28

Will the Democratic Party win the FL-14 House seat?

Seven of eight agents converge at 88-92% based on strong fundamentals: Solid D Cook rating, 15-point incumbent win margin, weak GOP field, favorable national environment. Agent 7 (55%) identified a legitimate redistricting threat from DeSantis's April 2026 special session. This risk is real but constrained: FL-14 is a secondary target, the timeline is extremely compressed (maps must be drawn, passed,...

Oracle: 90.6% Market: 56.5% Polymarket ↗

Read analysis

2026-03-26

Delcy Rodríguez out as leader of Venezuela by December 31, 2026?

7/8 agents cluster between 20-26%, with strong consensus that Rodríguez is consolidating power effectively in the near term. The strongest evidence (Agents 1-3) shows military loyalty, US cooperation, PSUV institutional control, and no credible threats. However, the 9-month window to December 2026 includes the constitutional deadline, potential US policy shifts, and ongoing internal balancing challenges. Agent 6 at 12% underweights...

Oracle: 13.1% Market: 25.5% Polymarket ↗

Read analysis

2026-03-24

Will the Democrats win the Arizona governor race in 2026?

All 7 substantive agents (excluding Agent 1 which failed to research) converge on 62-74% probability for Democrats, with remarkable agreement on the key evidence: Hobbs leads all polls, incumbency advantage, midterm dynamics favor Dems, Biggs is a weak general election candidate, but Arizona leans R structurally and Hobbs has mediocre approval. The agents differ mainly in how they weight these...

Oracle: 79.0% Market: 71.0% Polymarket ↗

Read analysis

2026-03-23

Israel strike on Yemen by June 30, 2026?

Discarding Agent 8 (incomplete research), the remaining 7 agents range from 62-95%. The three most thoroughly researched agents (3, 4, 6) all converge around 93-95%, finding specific evidence of imminent Houthi war entry and active regional conflict. Agents 2 and 5 are lower (62-72%) primarily because they weight current Houthi restraint more heavily. However, the question asks about a 3+...

Oracle: 92.5% Market: 75.0% Polymarket ↗

Read analysis

2026-03-22

Will the Democrats win the 2028 US Presidential Election?

All substantive agents agree on the directional case for Democrats (historical patterns, low Trump approval, unpopular war) but also agree the 2.5-year time horizon severely limits confidence. The ensemble mean of 56.5% exactly matches the market price, and removing the failed agent and outlier barely changes it (56.2%). Agent 7 explicitly would bet Republican at this price, while Agents 2-6...

Oracle: 61.9% Market: 56.5% Polymarket ↗

Read analysis

2026-03-21

Will the Republicans win the Iowa Senate race in 2026?

All 8 agents agree on the fundamental direction (Republican favored) with disagreement only on magnitude. The strongest evidence comes from structural factors (R+13 lean, 4/4 recent wins, expert ratings of 'Likely R') which multiple agents independently verified. The market at 61.5% appears to underweight these structural advantages. The main counterargument - midterm headwinds under unpopular Republican president - is real...

Oracle: 80.1% Market: 61.5% Polymarket ↗

Read analysis

2026-03-20

Will Gavin Newsom announce a Presidential run by December 31, 2026?

All 8 agents converge on NO being heavily favored (range 10-25%, 6 of 8 at 10-18%). The structural case is strong: Newsom is still governor until January 4, 2027, has explicitly said he'll decide after midterms (leaving only ~6 weeks), and no modern presidential candidate has announced this early for a race 2+ years away while still serving as governor....

Oracle: 7.5% Market: 22.5% Polymarket ↗

Read analysis

2026-03-19

Miguel Díaz-Canel out as leader of Cuba by June 30?

Agents split dramatically (12-68%) reflecting genuine uncertainty about an unprecedented situation. The two lowest agents (12%) anchor too heavily on historical base rates without adequately accounting for the 2026 US aggression pattern. The highest agent (68%) overweights the Maduro/Khamenei precedent without accounting for Cuba's different circumstances (island geography, no arrest warrant, stronger institutional resilience). The middle cluster (35-48%) is most...

Oracle: 26.1% Market: 50.5% Polymarket ↗

Read analysis

2026-03-19

Ukraine agrees to limit size of armed forces before 2027?

All 8 agents agree on core facts but differ on weighting. Agent 7 (20%) found the strongest specific evidence - the limitation clause was removed from the updated plan and language shifted from 'limits' to 'remains/maintains.' The ensemble mean of 30% is inflated by Agents 4 and 5 (35-38%) who overweight private draft agreements without sufficiently discounting the public commitment...

Oracle: 13.9% Market: 26.0% Polymarket ↗

Read analysis

2026-03-18

Will the Republicans win the North Carolina Senate race in 2026?

All 8 agents agree Cooper is heavily favored with strong polling leads, fundraising dominance, and candidate quality advantage. The disagreement is only on degree (17-33% for Republicans). The median estimate of 21% aligns closely with the market price of 18.5%, and the Democrat sibling at 78% implies ~22% for all non-Democrat outcomes. With only ~1.5% edge above market, this falls...

Oracle: 10.7% Market: 18.5% Polymarket ↗

Read analysis

2026-03-18

Will Trump launch a coin by December 31?

All 8 agents agree on the core facts: existing tokens don't count, DJT token is non-transferable, no new tradable token is announced. The disagreement is purely about weighting Trump's historical pattern vs. current evidence of restraint. The deliberate non-transferability of the DJT token is strong specific evidence that Trump's legal team is steering away from tradable tokens while he's president....

Oracle: 9.4% Market: 18.0% Polymarket ↗

Read analysis

2026-03-17

Will Mamdani freeze NYC rents before 2027?

Seven of eight informed agents converge on the same critical finding: the market requires 0% on BOTH lease types, which has never happened in 55+ years of RGB history. Six agents cluster at 32-42%, with strong agreement on the key constraint. Agent 1 at 62% overweights board control; Agent 5 at 15% underweights Mamdani's unique position. The community discussion explicitly...

Oracle: 30.3% Market: 54.5% Polymarket ↗

Read analysis

2026-03-17

Will Tesla launch robotaxis in California by June 30?

All 8 agents converge on 3-6% probability with remarkable agreement (std dev only 1.1%). This is one of the strongest consensus cases I've seen. The evidence is overwhelming and multi-layered: 1. REGULATORY IMPOSSIBILITY: California requires 50,000 supervised autonomous miles → driverless testing permit → commercial deployment permit. Tesla has logged 562 total miles since 2016 and ZERO since 2019. This...

Oracle: 2.3% Market: 15.0% Polymarket ↗

Read analysis

2026-03-16

Billionaire one-time wealth tax on California ballot?

Agents split into two camps: those emphasizing SEIU's organizational capacity and opposition's revealed preference (62-72%) vs those emphasizing historical base rates and funding disparity (48-56%). The strongest specific evidence comes from Agent 2 (opposition behavior signals expected qualification) and Agent 6 (leverage/withdrawal risk). My estimate of 62% is marginally above market (59.5%) but the edge of ~2.5% is far below...

Oracle: 69.7% Market: 59.5% Polymarket ↗

Read analysis

2026-03-16

Will the Democrats win the Maine Senate race in 2026?

All 8 agents converge on 43-63% range, with 7 of 8 between 55-63%. The consensus is that this is genuinely competitive but NOT a clear Democratic lean. The strongest evidence comes from: (1) Expert forecaster consensus at Toss Up, (2) Collins' extraordinary 2020 overperformance which represents a systematic Maine polling bias, (3) Platner's severe scandal baggage. The market at 75%...

Oracle: 66.7% Market: 75.0% Polymarket ↗

Read analysis

2026-03-16

Will the Democrats win the Michigan governor race in 2026?

All 8 agents agree on the fundamental dynamics: Duggan is the decisive variable, he disproportionately hurts Democrats, and current three-way polls show James leading. The disagreement is purely about whether Duggan fades (historical norm) or persists (his unusual funding/backing suggests). Agents with the strongest specific evidence (3, 6, 8) cluster around 42-47%, weighting current polling. Agents weighting historical third-party fade...

Oracle: 50.0% Market: 61.0% Polymarket ↗

Read analysis

2026-03-16

Will the Democrats win the Ohio Senate race in 2026?

Six of eight agents cluster at 46-51% with two outliers at 56-58%. The median (48%) better represents informed opinion than the mean (50.1%). Key tension: Ohio's hard structural R+11 lean and Brown's recent 2024 loss vs. favorable midterm dynamics and current polling leads. The structural lean is more predictive than polls 8 months out. Critically, the Republican sibling market at...

Oracle: 46.5% Market: 52.0% Polymarket ↗

Read analysis

2026-03-15

Will the Democrats win the Texas Senate race in 2026?

Agents split into three camps: historical-base-rate-heavy (24-25%, Agents 7-8), balanced (30-35%, Agents 2-4,6), and environment-heavy (45-50%, Agents 1,5). Agent 1 had no research and should be discarded. The strongest-evidence agents (2,3,4,6) converge around 30-35%. The market at 43.5% appears to be driven by Democratic enthusiasm and hype rather than fundamentals. Key facts: both Sabato and Cook rate this Likely Republican,...

Oracle: 21.0% Market: 43.5% Polymarket ↗

Read analysis

2026-03-15

Will the Republicans win the Georgia governor race in 2026?

Seven of eight agents with actual research converge on 40-52%, with strong consensus on key factors: Republican structural advantage in GA gubernatorial races offset by hostile midterm environment, open seat, and divisive GOP primary. Professional forecasters (Cook Toss-Up, Sabato Lean R) confirm genuine competitiveness. The market at 42.5% is very close to the informed agent median of ~44-45%. Agent 8's...

Oracle: 39.8% Market: 42.5% Polymarket ↗

Read analysis

2026-03-15

Will the Republicans win the Maine Senate race in 2026?

Discarding Agent 1 (no research, 50% default), the remaining 7 agents cluster tightly between 26-36% with adjusted mean ~31%. All agents identified the same core dynamics: Collins trails Platner in polls but has a proven track record of outperformance, the national environment is hostile to Republicans, and the Democratic primary is the key variable. The ~4% edge over market price...

Oracle: 19.9% Market: 26.0% Polymarket ↗

Read analysis

2026-03-15

Will the Republicans win the Wisconsin governor race in 2026?

Seven of eight informed agents agree the national environment strongly favors Democrats. Discarding Agent 1 (no research) and down-weighting Agent 8 (underweighted clear evidence), the informed consensus clusters at 18-22%. The market at 18.5% is already efficiently priced near the bottom of this range. My estimate of 20% reflects slight upward adjustment for open-seat uncertainty and Democratic primary risk, but...

Oracle: 10.7% Market: 18.5% Polymarket ↗

Read analysis

2026-03-14

European country agrees to give Ukraine security guarantee by June 30?

Discarding Agent 1 (no research, 50% default), all 7 informed agents agree on the fundamental finding: no existing agreement meets the market's strict Article 5-equivalent standard, and all planned binding guarantees are conditional on a ceasefire that doesn't exist. The informed agent mean is ~14.9% with median ~12%. Agents 2 (7%), 3 (10%), 4 (12%), 5 (12%) had the strongest...

Oracle: 5.0% Market: 21.5% Polymarket ↗

Read analysis

2026-03-14

Maduro guilty of all counts?

Informed agents (2-5) cluster at 27-35%, while less-researched agents (1, 6, 8) push higher. Agent 1 had zero research and should be discarded. The core tension is between high federal conviction rates and the 'all counts' + timeline constraints. Hernandez all-counts SDNY conviction supports YES; Noriega partial conviction and tight timeline support NO. My estimate of 33% is slightly below...

Oracle: 23.5% Market: 35.5% Polymarket ↗

Read analysis

2026-03-14

US congress stock trading ban before 2027?

All 8 agents converge on a low probability (10-17%), with strong consensus on the key dynamics. The evidence is remarkably consistent across agents: **Consensus findings (high reliability):** 1. H.R. 7008 passed House committee Jan 14, placed on Union Calendar Feb 3, but NO floor vote scheduled as of mid-March 2026 2. Trump endorsed at SOTU but action 'remains stuck' 3....

Oracle: 6.4% Market: 18.0% Polymarket ↗

Read analysis

2026-03-14

Will a Chinese AI model become #1 by June 30?

Seven of eight agents (excluding Agent 3 which failed with 50%/0% confidence) show strong consensus in the 10-22% range, with most clustering around 12-17%. The consensus is built on robust, mutually reinforcing evidence: (1) 0% historical base rate - no Chinese model has ever held #1 on Chatbot Arena in 2+ years; (2) current Elo gap of 38-50+ points between...

Oracle: 7.0% Market: 18.5% Polymarket ↗

Read analysis

2026-03-14

Will a province schedule a referendum to leave Canada before 2027?

Five agents defaulted to 50% with zero confidence, polluting the ensemble. Among agents with actual estimates, the evidence-grounded ones (Agents 4, 8) correctly note the extreme rarity of secession referendums and absence of current campaigns. Higher estimates appear speculative without identifying a specific province or triggering mechanism. The market at 49% seems inflated - possibly driven by general anxiety about...

Oracle: 14.8% Market: 49.0% Polymarket ↗

Read analysis

2026-03-14

Will BRICS add a new member in 2026?

All 8 agents agree the probability is below the market price of 40%, ranging from 22-38% with a mean of 29.4%. The strongest evidence comes from Agents 2, 4, and 6 who found specific, recent (February 2026) official statements from India explicitly deprioritizing expansion. Agent 4's quote from India's Secretary (East) is particularly compelling as direct evidence against near-term expansion....

Oracle: 16.7% Market: 40.0% Polymarket ↗

Read analysis

2026-03-14

Will the Republican Party win the FL-14 House seat?

7 of 8 agents with completed research agree this is a strong Democratic seat (range 8-42%, but 6 of 7 are 8-25%). The outlier at 42% (Agent 6) appears to anchor to market price. The consensus evidence is overwhelming: Solid D rating from all three major forecasters, 15-point incumbent margin in 2024, no retirement signals, weak GOP challenger, favorable national...

Oracle: 12.2% Market: 41.5% Polymarket ↗

Read analysis

Page 1 of 4 Older →