Point-in-time strategy research

AlphaVerdict

Run 0d9692dd01976a73 · synthetic-evidence-composite
fail
Audit verdict, not a trading signal
Evidence score
12
out of 100
Audit findings
8
5 independent reviewers
Evaluated periods
28
weekly rebalance

Equity path

Teal: strategy · gray: configured benchmark. Hypothetical, after configured friction.

Research metrics

Periods28
Years0.538
Total Return-14.50%
Annual Return-25.25%
Annual Volatility10.97%
Sharpe Ratio-2.591
Sortino Ratio-3.204
Calmar Ratio-1.741
Max Drawdown-14.50%
Win Rate17.86%
Best Period1.86%
Worst Period-4.54%
Var 95-3.52%
Cvar 95-4.05%
Skew-1.121
Excess Kurtosis1.036
Average Turnover16.22%
Annual Turnover8.43x
Average Gross Exposure46.43%
Beta0.290
Annual Alpha-20.64%
Information Ratio-0.103
Correlation0.450

Priority validation work

  1. Anchor the field to its public availability timestamp, not its fiscal period or event date.
  2. Reduce turnover or demand a larger gross edge; the current result does not survive plausible friction.
  3. Simplify the rule or broaden the sample; a stable edge should not depend on one contiguous window.
  4. Collect more independent periods and inspect the lower confidence path.
  5. Describe the failing regime explicitly and test a pre-declared exposure gate rather than tuning on the full sample.
  6. Repeat the evaluation without those symbols and inspect point-in-time universe breadth.
  7. Explain what risk or diversification benefit justifies the additional complexity.
  8. Repeat the run with licensed or user-owned real point-in-time data.

Run identity

Data

b22ec972779143b5a7588b83bb90818c0437df2750849a0bedfbc246f4764307

Strategy

ff94886307d997f9e2561d2318780b41e5296b0ce0fdd2493bb7cd120a717141

Findings

critical DATA_TEMPORAL_LEAK · Features appear knowable before they were observed

12 feature rows have available_at earlier than observed_at.

Next: Correct availability timestamps; fiscal periods and publication dates are not interchangeable.

high COST_FRAGILE · Gross edge sits close to configured friction

Approximate breakeven friction is -237.0 bps versus 100.0 bps configured.

Next: Reduce turnover or demonstrate a wider edge on untouched data.

high FOLD_INSTABILITY · Most contiguous evaluation folds were not profitable

0 of 2 folds had positive total return.

Next: Simplify the rule and pre-declare a fresh holdout before further tuning.

medium BOOTSTRAP_UNCERTAIN · Bootstrap evidence is not decisive

Probability of a positive resampled total return is 0.0%.

Next: Collect more independent periods and inspect the lower confidence path.

medium REGIME_INSTABILITY · Performance is concentrated in one coarse regime

Only one evaluated benchmark regime had positive compounded return.

Next: Treat regime dependence as a hypothesis and validate it on a fresh sample.

medium SYMBOL_CONCENTRATION · A few stocks dominate absolute contribution

The top three symbols contribute 89.3% of absolute gross contribution.

Next: Repeat the evaluation without those symbols and inspect point-in-time universe breadth.

low BENCHMARK_UNDERPERFORMANCE · Strategy underperformed its configured benchmark

Net total return was -14.5% versus -14.1% for the benchmark path.

Next: Explain what risk or diversification benefit justifies the additional complexity.

info DATA_SYNTHETIC · Run uses demonstration data

Synthetic fixtures validate plumbing only and cannot support a market claim.

Next: Repeat the run with licensed or user-owned real point-in-time data.

Research boundary

A pass means only that this run survived the configured tests. It is not evidence of future returns, a recommendation, or permission to deploy capital.

Results are research simulations, not investment advice or execution instructions.

Signals are formed after a decision close and applied at the following session open.