Weak after spread on /backtest: ~-2.6pp expected 24h move, ~40% win rate across the top-15 sample. Treat as research, not a profitable signal.
SampleHIGHFull backtest
Research Queue
Research Queue · ranked by Research Score
A daily worklist of markets where multiple factors line up. Each row carries a transparent score, confidence context, and a verification next step.
Research candidates
7
Resolution jobs
15
Top score
60
High-confidence
5
Score blends live signal evidence, multi-kind agreement, liquidity, spread quality, and resolution-risk penalty. The action column stays verification-first: watch, alert, verify, compare.
Methodology explanation
Review-only opportunity
No paper intent is emitted from the public opportunity row. Paper action appears only after the paper governor evaluates the row.
read_only_explanationpaper-onlysend false
Why this market is in review
signalMomentum / follow-through
89% source confidence on this opportunity row.
Paper-only action
paper-onlyReview-only opportunity
read_only_explanation; no live order, no network send, no raw secrets.
Risk / veto readback
reviewPaper governor not run here
This card explains the opportunity row only; a paper scan is still required before a paper intent or veto exists.
Source evidence
source4 mapped surfaces
283/283 sources runtime-backed; all registered sources are runtime-backed; live authority remains false.
Signals
- Momentum / follow-throughclear
89% source confidence on this opportunity row.
Veto / blockers
- Paper governor not run heresource
This card explains the opportunity row only; a paper scan is still required before a paper intent or veto exists.
Costs / sizing
- Research score
- Composite opportunity score before paper-governor costs and vetoes.
- Capacity
- Estimated research capacity, not an approved size.
- Liquidity
- Market liquidity visible on the opportunity row.
Explanation packet2026-06-16.1
DecisionSignalsVetoes and capsCostsDriftShadow policyLearning statusAuthority
What can be learned
Source-to-runtime coverage283/283 runtime-backed0 unmapped
all registered sources are runtime-backed; live authority remains false; this card cites mapped surfaces only.
P32-Source-to-Runtime-Coverage-Matrix-v1microstructure_quality_kernelexecution_markout_kernelpaper_governor_policy+2
live falsesecrets falsetraining falsenot advice
Aggregate after-spread verdict per signal kind, mirroring the /backtest page. The inspector on the Research tab also shows the live slice verdict per signal so users can see when the live observation carries higher evidence than the historical aggregate would support.
Weak after spread on /backtest: ~-1.8pp expected 24h move, ~49% win rate across the small sample available. Inspect, not actionable.
SampleLOWFull backtest
Forward-only — flow signals need persistent trade-feed history we don't yet store.
SampleFORWARD-ONLYFull backtest
Forward-only — news ingestion isn't persisted yet, so there's no backtest to aggregate.
SampleFORWARD-ONLYFull backtest
Forward-only — event-time risks can't be replayed from price history alone.
SampleFORWARD-ONLYFull backtest
Verdicts are a static snapshot from the latest /backtest run on the top-15 markets by volume; they drift as upstream data changes. A persistent backtest history will replace this snapshot once the signal-firing ledger lands.
Honest caveats
- Not a trade signal. The Research Score is a multi-factor ranking for daily research attention, not a forecast. Use the suggested next step — watch, verify, alert, compare — to validate independently.
- Capacity is an estimate. ~8% of visible liquidity. Real capacity depends on the depth of the order book at the moment you place a trade, which this view doesn't see.
- Horizon is a heuristic. Flow + divergence patterns play out in hours; momentum in days. Expiry dominates if closer.
- No execution costs modeled. Spread + Polymarket fees would eat into any nominal advantage if you traded on the score directly — which is why every row carries a verification next-step instead of a trade recommendation.