Admission v7 promotion / Receipt

Capacity policy

Capacity usd min

500000

Capacity usd min previous

5000000

Reason. A 5,000,000 USD floor per sleeve implies a 70,000,000 USD book at fourteen sleeves. It was written for a fund that does not exist yet, and it did not reject candidates uniformly: the return sources most likely to be genuinely anti-correlated with a momentum-and-carry book -- catastrophe bonds, municipal basis, freight, power, sovereign dislocation -- are structurally thin. The floor therefore selected AGAINST the diversification the objective depends on, which is the opposite of what a capacity gate is for.

Review trigger. Re-raise per sleeve when deployed capital per sleeve exceeds one fifth of this floor. Capacity is a property of the strategy, but the FLOOR is a property of the book, and a floor that outruns the book is a constraint that costs edge for nothing.

What this does not relax. The capacity CURVE, its monotonicity reconciliation, and the stressed fill-ratio floor are unchanged. A candidate must still show Sharpe decaying and cost rising with capital, and must still reconcile its reported capacity to that curve.

Deflation policy

Book deflation is mandatory for maturity

yes

Book deflation is strictly additional

no

Book maturity threshold

0.950000

Book measurement key

book_deflated_sharpe_must_be_measured

Book selection unit

complete_union_hypothesis_identities

Book threshold key

—

Per sleeve is measured not gated

yes

Per sleeve measurement key

deflated_sharpe_must_be_measured

Per sleeve selection unit

family_trial_account

Per sleeve threshold key

deflated_sharpe_min

Per sleeve gate removed reason. The per-sleeve 0.95 DSR gate was already removed in v6 because it measured the wrong object and silently required about eight times the declared standalone Sharpe floor. Per-sleeve DSR remains mandatory to measure and publish. V7 additionally re-scopes the full-union book DSR threshold: it remains mandatory to measure at every admission and becomes a 0.95 gate at portfolio maturity, while incremental admission is decided by the positive bootstrap lower bound on book-Sharpe improvement and the other unchanged robustness, execution, stress, capacity, and trial-accounting gates.

Reason. DSR deflates for the trials the selection was drawn from. A pre-registered, direction-locked candidate in a new family was not selected from other families' trials. The published claim is the book's, so the book carries the full union.

Diversification evidence policy

Candidate weight

predeclared upstream and supplied explicitly

Confidence level one sided

0.950000

Confidence method

deterministic circular moving-block bootstrap

Default block size

21

Default bootstrap samples

2000

Default seed

20260816

Implementation

alphaforge.validation.diversification.diversification_report

Return type

aligned OOS simple returns with no dropped rows

Stress mask

predeclared upstream and supplied explicitly

Execution evidence policy

Dimension hash is canonical manifest hash

yes

Not applicable requires sha256

yes

Scenario hashes are canonical payload hashes

yes

Tested statuses require scenario manifest

yes

Tested statuses require scenarios

yes

Tested statuses require sha256

yes

Scenario manifest required fields

• scenario_id
• status
• assumptions
• assumptions_sha256
• result
• result_sha256

Scenario statuses

• PASS
• FAIL

Frontier arithmetic

Identity

S_book = s_bar * sqrt(N / (1 + (N - 1) * rho_bar))

Incremental book average correlation delta gate exclusive

0

Incremental candidate average correlation gate

0

Incremental gates alone establish objective ceiling

no

Incremental gates alone establish objective floor

no

Legacy global zero correlation reference

0

Psd floor at target n

-0.076923

Target sleeve count

14

Book sharpe ceiling at zero global correlation

2 further fields. Read them in the artifact.

Correlation required at measured quality

3 further fields. Read them in the artifact.

Honest reading. Tightening the correlation ceiling from 0.15 to 0.00 removes a ceiling that sat BELOW the objective at every plausible quality, which is why the v4 gate could not have produced the objective at any sleeve count. It does not on its own deliver the objective. At the gate exactly, fourteen sleeves of today's measured quality reach the figure recorded above, and the remaining distance has to be bought with per-sleeve quality, genuinely negative correlation, or both. Publishing this is the point: a target sitting beside a gate that forbids it is not a stretch goal.

Psd floor note. Average pairwise correlation cannot fall below -1/(N-1) for any real correlation matrix. At fourteen sleeves that floor is far from binding, which is why a negative average is arithmetically available here and is not at 250 sleeves.

Quality precondition at the gate

5 further fields. Read them in the artifact.

Gate power audit

Candidate return artifacts read

0

Content hash

sha256:743c803a4a65869be14957cec1ce0d7737568d26fb3d89278f827dfedd0186fb

Hypothesis identities consumed

0

Path

artifacts/analysis/admission_gate_power_audit/result.json

Sha256

0efb66ef4b7f5e497da56fcb2e178a69ef522f69ac4ba7d9b09297c5ae057f42

Incremental gate scope policy

Book deflated sharpe

5 further fields. Read them in the artifact.

Correlation

7 further fields. Read them in the artifact.

Lineage types

Code commit field

code_commit

Boolean fields

• point_in_time_data
• survivorship_control
• corporate_action_control
• direction_locked
• parameters_locked

Identifier fields

• family_trial_account
• return_identity_id

Sha256 fields

• preregistration_sha256
• data_manifest_sha256
• diversification_report_sha256

Objective

Average pairwise correlation objective

-0.030000

Honest forward sharpe target

1.500000

Minimum new sleeves

10

Portfolio max drawdown statistic

expected_maximum_drawdown

Portfolio max drawdown target

0.110000

Portfolio p95 max drawdown is gated

no

Portfolio p95 max drawdown must be published

yes

Target total sleeves

14

Targets are admission evidence

no

Average pairwise correlation objective note. An objective, not a gate. -0.03 is the average pairwise correlation at which fourteen sleeves of measured quality reach 2.5. Thresholds are gates and live in 'thresholds'; anything that is not enforced must not sit there wearing a gate's costume.

Backtest to forward haircut

3 further fields. Read them in the artifact.

Implied in sample target

3 further fields. Read them in the artifact.

Portfolio sharpe target

• 2.250000
• 3

Portfolio sharpe target unit. IN-SAMPLE, being the band that an honest forward 1.5 implies at the 1.5x and 2.0x ends of this book's own measured haircut. The headline target is honest_forward_sharpe_target.

Superseded in sample sharpe target

• 2
• 2.500000

Superseded reason. Stated in-sample with no haircut named, and unreachable as a forward result: a verified forward 2.0 at measured per-sleeve quality needs an average pairwise correlation of -0.062 against a PSD floor of -0.077. Withdrawn 2026-08-21 and replaced with an honest FORWARD target of 1.5, which is inside the floor at every plausible quality.

What reaching it requires. Both halves, not either: an average pairwise correlation meaningfully BELOW zero, and per-sleeve quality above roughly 0.60. At today's measured 0.469 the forward target is reachable only at a correlation near -0.030 and only on the optimistic haircut.

Why this is not a retreat. The rarer thing was never the number. A forward Sharpe near 1.5 that is genuinely verifiable -- signed forward record, every killed candidate published in full, gates that provably bite, and the arithmetic showing where the gate falls short of the target printed on the same page as the target -- is harder to find than a 2.5 nobody can check.

Overlay policy

Annualized vol target

0.100000

Both halves shipped

no

Covariance halflife days max

21

Halflife pending execution costing

yes

Implementation

alphaforge.portfolio.overlay.vol_target

Mandatory vol target overlay

yes

Production cov halflife today

720

Production realized vol halflife today

240

Realized vol halflife days max

240

Realized vol leg must be unlevered

yes

Sigma hat rule

max(ex_ante, realized)

Cost of the halflife change

6 further fields. Read them in the artifact.

Covariance halflife note. 21 is the only tested covariance halflife that holds expected max drawdown at or under the objective at the permitted stressed correlation. Provisional pending execution costing.

Halflife unit defect

4 further fields. Read them in the artifact.

Known defect

6 further fields. Read them in the artifact.

Production cov halflife unit. BARS, and therefore a different duration per sleeve: 720 bars is 30 days on the crypto H1 sleeve and 720 sessions, about 2.86 years, on equity D1. Incoherent by construction and unchanged, because changing it changes the live book.

Production realized vol halflife unit. BARS, not days, and NOT converted per calendar. On the crypto H1 sleeve 240 bars is 10 days, not the 240 days the realized_vol_halflife_days_max threshold names. Only the covariance leg gained a calendar conversion; the realized leg still has none. Stated here rather than left implied, because the earlier wording of this block asserted a per-calendar conversion for both legs and that was false for this one.

Realized vol halflife note. 240 is production's value and the value every sweep cell used. No shorter realized-leg halflife was tested, so none is required; this threshold bounds regression, not design.

Reason. Permitted stressed pairwise correlation of 0.50 implies an expected maximum drawdown near 20% with no overlay and 16.4% with the overlay as shipped today. Correcting the realized-leg scale and shortening the covariance halflife to 21 holds it at 10.2%. The overlay, not a tighter correlation ceiling, is what holds the drawdown objective.

Remaining gap to the drawdown objective. OPEN, and wider than a previous revision of this contract claimed. The measured study holds expected maximum drawdown at 10.2% only with BOTH the realized-leg scale correction (shipped) and the covariance halflife at 21 days (NOT shipped). The halflife change was briefly made the default and has been reverted, for three reasons recorded in full below: it silently re-sizes every live sleeve on the next daily tick; the sweep cell used to price it does not correspond to what the crypto sleeve actually ran; and the live estimator is windowed and seeded, so it does not reproduce the untruncated recursion the sweep simulated. The objective is also an EXPECTED maximum drawdown -- no tested configuration held the 95th percentile at or under 11%, the best being 13.8%.

Why the halflife change was reverted

7 further fields. Read them in the artifact.

Pbo policy

Measured status

MEASURED

Never report not defined as zero

yes

Single identity status

NOT_DEFINED_SINGLE_IDENTITY

Single identity value must be null

yes

Prospective scope

Applies only to identity reservations after effective hash

yes

Effective

yes

Effective contract content hash

sha256:bf175acfebead6fdfaa3d0f0f8380636797e468eb071fa20beae12f44d55ddb5

Effective on or after reservation ordinal

229

Known candidates may not be regraded

yes

Legacy identities retired and ineligible

228

Owner promotion record

config/admission_v7_promotion.json

Promotion requirements

• INDEPENDENT_TEST_SUITE_PASSES
• PUBLIC_GATE_PROJECTION_RECONCILES
• TRIAL_BUDGET_V7_IS_HASH_BOUND
• OWNER_PROMOTION_RECORDED

Retired admission gates

Average pairwise correlation max

3 further fields. Read them in the artifact.

Book deflated sharpe min

3 further fields. Read them in the artifact.

Significance policy

Minimum oos observations raised from

252

Newey west t min

0.250000

Newey west t min role

sign gate

Newey west t ratio min

0.600000

Newey west t ratio role

autocorrelation-inflation gate

Primary significance gate

book_sharpe_delta_lower_95_min_exclusive

Correlation observations decoupled. minimum_correlation_observations is separate from minimum_oos_observations because they measure different things: a candidate's own sample is the candidate's to supply, while the correlation window is its intersection with the BOOK. Tying them together made admission impossible for a reason that had nothing to do with any candidate -- v6 briefly required 756 correlation observations against a book that can supply 728 on its blessed basis and 177 on its live-forward one.

Minimum oos observations reason. Raised to three years because the binding constraint of this whole programme is an average pairwise correlation, and a correlation measured over 252 observations carries a sampling error near 0.063 -- twice the size of the -0.03 effect the objective turns on. Gating a number the sample cannot resolve is not a strict gate. Paired with average_pairwise_correlation_upper_95_max, which gates the bound rather than the point estimate.

Reason. v4 declared net_sharpe_min 0.40, newey_west_t_min 2.0 and minimum_oos_observations 252 together. Because t is approximately Sharpe * sqrt(years), a candidate at the declared Sharpe floor needed 25 years of out-of-sample data to clear the t floor, and at the declared 252-observation minimum the pair demanded a standalone Sharpe of 2.0 -- five times the floor a reader of the config would see. v5 lowered the Sharpe floor to 0.15 and left the t floor at 2.0, raising the hidden requirement to 178 years. The t floor, not the Sharpe floor, had been doing the rejecting all along. It is now set where it does what it is for -- excluding wrong-signed and indistinguishable-from-zero results -- and load_admission_contract refuses any contract whose three significance floors cannot all be met at once, so this class of error cannot recur.

Self consistency is enforced at load. load_admission_contract refuses any contract whose significance floors cannot all be satisfied at once. It checks two pairs: net_sharpe_min against newey_west_t_min at minimum_oos_observations, and net_sharpe_min against deflated_sharpe_min solved at n_trials=2. Both pairs were violated before v6, in the same way and undetected, which is why the check is structural rather than a corrected number.

Why a standalone bar is the wrong instrument. When correlation binds, standalone Sharpe does not rank a candidate's portfolio value: a Sharpe 0.25 sleeve at correlation -0.05 beats a Sharpe 0.60 sleeve at +0.30. The gate that decides admission is therefore the bootstrap LOWER bound on the book-Sharpe improvement, which prices edge and correlation together, and it is strictly harder to game than either input alone.

Thresholds

Absolute beta max

0.100000

Average pairwise correlation upper 95 max

0.100000

Book average pairwise correlation delta max exclusive

0

Book deflated sharpe must be measured

yes

Book expected max drawdown max

0.110000

Book expected shortfall delta max

0

Book max drawdown delta max

0.010000

Book sharpe delta lower 95 min exclusive

0

Book sharpe delta min exclusive

0

Candidate average correlation to existing book max

0

Capacity curve min points

3

Capacity minimum stressed fill ratio

0.950000

Capacity usd min

500000

Covariance halflife days max

21

Deflated sharpe must be measured

yes

Minimum correlation observations

504

Minimum execution scenarios per dimension

3

Minimum leave one period out book sharpe delta exclusive

0

Minimum oos observations

756

Minimum stressed oos observations

63

Net sharpe min

0.150000

Newey west t min

0.250000

Newey west t ratio min

0.600000

Ordinary pairwise correlation max

0.350000

Pairwise correlation upper 95 max

0.350000

Pbo max

0.200000

Realized vol halflife days max

240

Realized vol leg must be unlevered

yes

Stressed pairwise correlation max

0.500000

Stressed pairwise correlation upper 95 max

0.500000

Stressed sharpe min

0.150000

Budget review

Authorized by

owner, 2026-08-21

Headroom

158

Hypothesis identities at review

162

New budget

320

Prior budget

160

Authorization. Explicit owner decision to fund the fourteen-sleeve programme, taken with the evidence cost on the table rather than inferred afterwards.

Cost of this decision

7 further fields. Read them in the artifact.

Next tripwire. At 320 recorded hypothesis identities, registration pauses again and requires a fresh owner review. Any single family exceeding 40 identities also pauses for review, so the budget cannot be consumed by one line of enquiry that is not working.

Rationale. Ten new sleeves at the measured 6.5% hit rate (3 survivors from 46 tested candidates) implies about 154 trials. The budget is set at 320 identities, which funds that with a small margin and stops well short of an open-ended search. It is a CEILING, not a plan: the campaign closes at whatever count the evidence stops justifying.

Status token note. ACTIVE is the canonical token the trial tripwire recognises. The prospective nature of this budget is recorded here rather than encoded in the status, so no guard is left unable to read the state it governs.

What did not change. A target still never creates a sleeve. Fourteen is reached only if ten or more new identities independently clear every admission gate in config/sleeve_admission_contract.json; otherwise the book remains smaller. The budget buys permission to look, not permission to admit.

Definitions

Hypothesis identity. A distinct economic or parameter identity after removing only start and end measurement-window keys.

Immutable execution record. One append-only, config-hash-unique measurement row. Rolling-window remeasurements remain records.

Window only remeasurement. A later record whose config differs from an existing identity only in start or end. It remains public but does not inflate DSR selection N.

Ledger scope correction

Authorized budget unchanged

320

Budget remaining

146

Current hypothesis identities

174

New experiments run

0

Prior reported hypothesis identities

162

Recorded at

2026-08-22

Recovered legacy hypothesis identities

12

Status

CORRECTED

Reason. Canonical union discovery scanned only top-level var* profiles and omitted four durable experiment ledgers filed with legacy result artifacts. The ledgers are selection evidence and have been restored to N. Withdrawn archive_broken_prices ledgers remain excluded.

Recovered ledger paths

• artifacts/exp1/20260625T075446Z/experiments.jsonl
• artifacts/exp4/20260625T103226Z/experiments.jsonl
• artifacts/exp4/smoke_20260625T102142Z/experiments.jsonl
• artifacts/grand_backtest/20260616T143620Z/experiments.jsonl

Prospective v7 review

Admission v7 effective contract hash

sha256:bf175acfebead6fdfaa3d0f0f8380636797e468eb071fa20beae12f44d55ddb5

All new identities enter complete union deflation

yes

Authorized by

Arhan Canli, owner, 2026-08-23

Book dsr required at 400 and 756 rows

2.292043

Current probability at least ten

0.076966

Current remaining identities

92

Effective on or after reservation ordinal

229

Identities are permission to test not permission to admit

yes

No legacy identity reopened

yes

Owner promotion record

config/admission_v7_promotion.json

Planning probability is not a forecast

yes

Promotion requires admission v7 content hash

yes

Prospective probability at least ten

0.690428

Prospective remaining identities

172

Single family tripwire

40

Status

IN_FORCE

Reason. Ledger reconciliation raised the observed union from 162 to 228 without running new experiments, leaving only 92 identities under the 320 ceiling. At the previously declared planning rate of 3/46, that headroom has an exact binomial probability of only 7.70% of producing ten successes. A ceiling of 400 restores 172 prospective identities and a 69.04% planning probability under the same explicitly uncertain assumption.

Staged hard reviews

• 320
• 360
• 400

Summary trial debt correction

Already reconciled before this correction

24

Authorized budget unchanged

320

Budget remaining

92

Current hypothesis identities

228

Evidence

artifacts/audit/trial_debt_reconciliation.json

New experiments run

0

Newly reconciled in this correction

54

Prior reported hypothesis identities

174

Recorded at

2026-08-22

Recovered historical hypothesis identities

54

Status

CORRECTED

Reason. Seven persisted historical studies reported named return configurations but bypassed the durable experiment ledger. The source-bound v2 reconciliation charges every measured parameter identity, including 32 AlphaMax robustness-grid configurations and AlphaTrend's post-hoc selector configurations; rounded summaries are never used to assert duplicate economic identity.

Source identity counts

7 further fields. Read them in the artifact.

V7 power audit binding

Content hash

sha256:743c803a4a65869be14957cec1ce0d7737568d26fb3d89278f827dfedd0186fb

Path

artifacts/analysis/admission_gate_power_audit/result.json

Sha256

0efb66ef4b7f5e497da56fcb2e178a69ef522f69ac4ba7d9b09297c5ae057f42

Contract proposal

Path

config/sleeve_admission_contract_v7_proposed.json

Sha256

ec929dba5beeca64644e5ecc257bc883d6598c077a718072148d2692020ff98c

Power audit

Path

artifacts/analysis/admission_gate_power_audit/result.json

Sha256

0efb66ef4b7f5e497da56fcb2e178a69ef522f69ac4ba7d9b09297c5ae057f42

Trial policy proposal

Path

config/trial_accounting_v7_proposed.json

Sha256

91df93fd2efe119a3d1e7a865fd16e7513ec4ffe7339c44f4a548422479dec67

Trial policy v1 archive

Path

config/archive/trial_accounting_v1_superseded.json

Sha256

e7c4c3372e3482acb6f0b852a962f91e335a7655c4bd33f8c02d74e9862195fd

V6 archive

Path

config/archive/sleeve_admission_contract_v6_superseded.json

Sha256

5f628a44ae73ed9c412b225ac9dc56c4e59c1bc0645e2ebde251501e938b9932