Methodology and governance PUBLIC SUMMARY
A documented path from raw onchain evidence to an underwriting conclusion.
OpenStrategy separates discovery, official evidence verification, quantitative screening, deeper underwriting, Index construction, and ongoing monitoring. Each layer answers a different question and has different limitations.
Methodology map
Different stages answer different questions.
A leaderboard rank is not underwriting. Underwriting is not Index fit. Index fit is not actionability. The system preserves these boundaries.
- 01
Discovery
Which observable profiles warrant official verification?
- 02
Official evidence verification
Can identity, history, activity, and data completeness be established?
- 03
Quantitative screening
Does the record pass defined quality and evidence gates?
- 04
Underwriting
What drives returns, risk, robustness, access, and monitorability?
- 05
Index construction
Is the strategy eligible under the public benchmark framework?
- 06
Monitoring and governance
Has evidence changed enough to trigger review, suspension, or removal?
O.P.E.N. Framework
A public lens over the full methodology.
O.P.E.N. organises the characteristics OpenStrategy communicates publicly: Onchain Evidence, Performance Quality, Exposure Discipline and Non-Fragility. It is a communication structure over the existing evidence, screening and underwriting system, not a replacement calculation or a standalone investment rating.
Agent Intelligence helps retrieve, compare and interpret approved evidence. Quantitative facts remain tied to deterministic calculations. Material conclusions and exceptions require human review.
Onchain Evidence
Identity, provenance, history, completeness, recency and whether material change can be monitored.
Performance Quality
Returns after considering flows, leverage, concentration, one-off effects and persistence across market conditions.
Exposure Discipline
Drawdown, leverage, concentration, liquidation sensitivity and current portfolio risk.
Non-Fragility
Whether performance persists across time, positions and regimes, or depends on one trade, event or market condition.
Calculation notes
Definitions behind the public record.
What counts as official evidence?
Official evidence is activity or portfolio data obtained from the strategy’s identified primary record and checked for identity, timestamps, coverage, continuity, asset pricing, and material gaps before analytical use.
Why does OSIQ 30 use equal weight?
Equal weighting makes the public headline less dependent on a single constituent and easier to interpret. Score weighting may be shown as context, but it is not the current reporting headline.
What do Current, Delayed, and Unavailable mean?
Current means the latest expected verified daily observation published successfully. Delayed means the expected update has not arrived within the publication tolerance. Unavailable means the public record cannot be validated or served. The publication process fails closed rather than carrying an unlabelled stale value forward.
What triggers a methodology or constituent review?
A material evidence gap, strategy drift, eligibility breach, constituent failure, calculation error, or method change can trigger human review. Eligibility is monitored daily; reconstitution is event-driven and governed. Corrections, replacements, and version changes are dated rather than silently rewritten.
OSIQ 30 construction
A public quality benchmark, not an investable product.
OSIQ 30 is a public research benchmark observing externally operated strategies. OpenStrategy evaluates their public records; it does not operate, execute or allocate to the strategies.
It tracks 30 selected strategies at equal weight. Selection uses verified evidence and a defined eligibility framework; exact constituent identities and cloneable thresholds remain private.
Track observable strategy quality
Represent a research cohort selected for performance quality, risk control, robustness, capacity, leverage discipline, and evidence confidence.
Equal weight
The headline series applies equal weights across eligible contributing constituents. Score weighting is contextual, not the reporting headline.
Flow-adjusted and chain-linked
Positive inferred deposits enter capital at risk. Withdrawals do not reduce the denominator retroactively. Returns are chain-linked across aligned official portfolio points.
Scheduled and verified
One observation is captured after 00:05 UTC each day. It becomes Final only after the selected-strategy capture and canonical verification succeed; failed verification stops publication.
Observed close, next-interval effect
A breach is recognised at the observed close. That close remains counted; return exclusion begins with the next interval until governed reselection.
Daily monitoring, event-driven review
Eligibility is monitored daily. Material evidence change, strategy failure, or a confirmed breach can trigger human review and an announced, continuity-preserving reconstitution.
Public calculations
The published series follows explicit formulas.
These simplified formulas show how the public record is calculated. Exact selection thresholds and constituent evidence packages remain private.
Flow-adjusted return
flow = Δ account value − Δ PnLcapital at risk = max(previous value + positive flow, previous value, 1)constituent return = Δ PnL ÷ capital at riskPositive inferred deposits are treated conservatively as available during the interval. Withdrawals never shrink the denominator retroactively.
Equal-weight return
Index returnt = mean(eligible constituent returnst)The headline return is the arithmetic mean of contributing constituent returns for the interval. Weight is redistributed across eligible contributors when a return exclusion is active.
Chain linking
Index levelt = Index levelt−1 × (1 + Index returnt)A governed basket change starts from the preserved prior endpoint. Earlier published rows are not restated to make a new basket appear historical.
Published-observation drawdown
drawdownt = Index levelt ÷ running peakt − 1maximum drawdown = min(drawdownt)This measure uses published observations. Intraday peak-to-trough drawdown may be larger and is not currently measured.
Two distinct events
observed close: breach recognised and close retainednext interval: return contribution excludedThe recognition date and the return-exclusion start date are recorded separately.
Operating rules
Data exceptions are handled without silent substitution.
The public process distinguishes a delayed observation, a constituent evidence problem, a governed basket change, and a historical correction.
Index record
One stated inception and one continuous series.
OSIQ 30 began at 100.00 on 1 July 2026. The record is published once daily after the scheduled evidence capture and verification complete successfully.
| Field | Convention | Current value | Interpretation |
|---|---|---|---|
| Inception | Research benchmark base | 1 Jul 2026 · 100.00 | Starting point for the published Index series |
| Latest observation | Verified daily publication | 29 Jul 2026 · 110.95 · Snapshot 29 Jul · 00:05 UTC | Final after canonical verification succeeds |
Evidence controls
Known limitations belong in the method.
Onchain visibility improves observability but does not eliminate inference risk, implementation risk, or the limits of a short record.
Short operating record
A new benchmark has limited observations. Early return, drawdown, and volatility figures should not be treated as stable long-term estimates. Reported maximum drawdown is computed from published daily observations; a full intraday series is not yet reported and could show a larger decline.
Flow distortion
Deposits, withdrawals, and near-zero capital can distort naive wallet ROI. Flow-adjusted methods reduce but do not remove every ambiguity.
Evidence incompleteness
Official activity records can still have coverage, asset-pricing, timing, or interpretation limits. Confidence is a gate, not decorative metadata.
Capacity and copyability
Observed strategy returns do not prove that external capital can access or reproduce them after liquidity and execution costs.
Strategy drift
Operators, exposures, leverage, venues, and behaviour can change. A historical label is not permanent.
Implementation boundary
Index eligibility does not establish legal, operational, custody, counterparty, mandate, or allocation readiness.
Governance
Versioned methods and human accountability.
Quantitative facts remain tied to verified evidence and deterministic calculations. Material interpretations, corrections, assessment conclusions, and exceptions require human review.
Corrections policy: Material factual or calculation errors are corrected visibly, dated, and added to the relevant revision record. Method changes receive a new version; published history is not silently rewritten.
Public/private boundary: This page explains the method sufficiently to interpret the public record without disclosing exact constituent identities, cloneable selection thresholds, or private evidence packages.