quantbase.liveRun this gate on yours
Public verdict · gate_check · 2026-09-13 07:24 UTC

Verdict+

donchian 55d breakout did not pass.

6 of 7 blocking checks failed — judged on ETH-USD 1h bars from coinbase, 2024-01-02 → 2026-05-01, with the family's ledger at N = 2 trials.

Familyxsec-donchian
Splituniverse-2024-common-window
DataETH-USD · coinbase · 1h · 24,010 bars
LedgerN = 2 in this family
0.645
Deflated Sharpe · need ≥ 0.95
0.7
P(overfit) · need < 0.5
0.49
t-stat · need > 3
5
Trades · need ≥ 50
Table 1 · 10 checks7 blocking · 3 warning
StateCheckValueHurdleDetail
FAILDeflated Sharpe0.645≥ 0.95deflated Sharpe against N=2 trials
FAILProbability of backtest overfitting0.7< 0.5combinatorially symmetric CV
FAILt-statistic · Harvey–Liu hurdle0.49> 3Harvey-Liu hurdle, not the conventional 2
FAILBacktest length · years1.413≥ 138.4MinBTL = 2·ln(2) / E[max]²
FAILRegime survival · regimes that lost moneybear, chopnonetested: ['bear', 'bull', 'chop']
PASSNet expectancy · bps per trade85.15> 0after the pinned cost model
FAILTrades5≥ 50too few trades to say anything
WARNCost model sourcefallback published schedule= coinbase apifees assumed from the published schedule ($0-$10K: 40/60 bps); 1 cost model(s) in this family are not live reads. Configure COINBASE_API_KEY and COINBASE_API_SECRET and re-run cost_model_refresh
PASSVenue consistencycoinbase advanced= coinbase advanced publishedbars from coinbase_advanced, fees from coinbase_advanced_published
WARNBreadth · effective tests2.94≥ 322 assets provide 2.9 effective tests (mean pairwise correlation 0.5267), M_eff by unlabelled; the estimator was not recorded, so this is judged against the participation_ratio hurdle — entropy-based M_eff on the same universe measures ~2.3x higher and is not comparable
Blocking checks decide the verdict. Warnings are recorded and shown, and do not.
ResultReferee

The referee's answer is no. What failed: deflated sharpe; probability of backtest overfitting; t-statistic; backtest length; regime survival; trades. A strategy fails when the evidence cannot distinguish it from the best of 2 attempts at nothing — not because it lost money. 2 warnings reported but not blocking.

LimitWhat is withheld

What this page does not show: the strategy's parameters. The owner published the verdict, not the spec — only its hash prefix 66a9f9acc391, which is what the ledger binds this result to. Spec still open; holdout registered.

Same seven blocking checks, on your strategy, from any MCP client. The ledger counts every trial you run; the holdout opens once.

Connect · freeHow the gate works
gate c34094c2-15e1-45b2-9a03-d98afed5cfa4 · spec 66a9f9acc391 · json