Monte-Neo strategy verdict REJECT

Certificate 90b86e297855b9d6 · engine v0.43.0 · 2026-09-30T08:39:54+00:00

Rejected: truncation probe: LEAK DETECTED; future-perturbation probe: LEAK DETECTED (and 4 more).

integrity3 pass

lookahead1 pass · 4 fail

economics1 pass · 1 fail

statistics3 pass · 1 fail

Not signed. Sign with monte-neo verify ... --sign KEY to prove who issued it.

Summary

net return-64.97%

Sharpe (annualized)-232.58

max drawdown-65.19%

closed trades1492

Deflated Sharpe0.000

timing p-value—

buy & hold return-1.76%

buy & hold Sharpe-5.01

break-even cost6.5 bps

Sharpe 95% interval-259.89 … -206.61

return 95% interval-69% … -61%

spread from high-low10.4 bps

capacity at 5% of volume13K

positionssign

Equity

strategybuy & holddrawdownlog scale

Return distribution

Per-bar returns after warm-up, bars with a change only.

Rolling Sharpe

Annualized, trailing window of 293 bars.

Sensitivity to costs

Net return as the per-side trading cost grows.

Trades

closed trades1492

win rate20.1%

profit factor0.24

average win+0.11%

average loss-0.11%

best trade+0.55%

worst trade-0.20%

average hold2.0 bars

longest win streak4

longest loss streak27

What to fix

The signal at bar t changes when later bars are removed: compute features only from rows <= t (no shift(-k), centered windows, bfill or full-sample stats).
Past signals change when the future is rewritten: remove whole-series statistics (mean/std/min/max over all rows) and future-dependent fills.
Fix the flagged source lines (negative shift, center=True, backward fill) and re-run verify. Lines: 7.
Next-bar hit rate is too high to be real: look for leakage of the next bar's close/open into the signal.
The strategy loses money after costs: reduce turnover or find a stronger edge.
Sharpe does not survive the number of variants tried: test out-of-sample or reduce the search space.

Evidence

fail lookahead_truncation

cut after bar 300: the signal at bar 300 was 1, then 0
cut after bar 301: the signal at bar 301 was -1, then 0
cut after bar 417: the signal at bar 417 was 1, then 0
cut after bar 425: the signal at bar 425 was 1, then 0

fail lookahead_perturbation

future rewritten after bar 300: the signal changed from bar 300
future rewritten after bar 839: the signal changed from bar 839
future rewritten after bar 1379: the signal changed from bar 1379
future rewritten after bar 1918: the signal changed from bar 1918

fail lookahead_static_lint

line 7 negative_shift: shift with a negative period reads future bars
return np.sign(df["close"].shift(-1) - df["close"])

fail implausible_accuracy

next-bar hit rate 1.000 over 2999 bars (z 54.8); a real edge rarely exceeds 0.70

Checks

check	category	status	summary
data_integrity	integrity	pass	OHLCV is clean
data_quality	integrity	pass	no spikes, frozen prices, split jumps or gaps
determinism	integrity	pass	signal() is deterministic
lookahead_truncation	lookahead	fail	truncation probe: LEAK DETECTED
lookahead_perturbation	lookahead	fail	future-perturbation probe: LEAK DETECTED
external_data	lookahead	pass	outside data: none read (df only)
lookahead_static_lint	lookahead	fail	static lint: negative_shift
implausible_accuracy	lookahead	fail	next-bar hit rate 1.000 over 2999 bars (z 54.8)
costs_modeled	economics	pass	10.00 bps per side modeled
net_profitability	economics	fail	net total return -64.97% after costs
cost_margin	economics	skip	not profitable after costs
delay_sensitivity	economics	skip	not profitable after costs
timing_significance	statistics	skip	not profitable after costs
sample_size	statistics	pass	1492 closed trades (min 30)
deflated_sharpe	statistics	fail	deflated Sharpe 0.000 over 40 trial(s)
trials_disclosed	statistics	pass	n_trials declared
holdout_consistency	statistics	pass	Sharpe/bar train -0.3171 vs holdout -0.3292
period_consistency	statistics	skip	not profitable after costs
walk_forward_stability	statistics	info	0 of 6 equal windows (490 bars each) made money; worst -17.9%, best -11.6%
sharpe_confidence	statistics	info	Sharpe 95% interval -259.89 to -206.61 (return -68.7% to -60.6%); 0% of resamples have a positive Sharpe
track_record	statistics	info	the Sharpe is not positive: no history length makes it significant
benchmark	statistics	info	buy & hold -1.76% (Sharpe -5.01) vs strategy -64.97% (Sharpe -232.58)
spread_estimate	economics	info	rough spread estimate from high and low: about 10.43 bps (5.22 bps per side); the model charges 5 bps slippage per side
capacity	economics	info	capital that keeps 90% of 1,492 fills within a share of the bar's volume: 1%: 2.6K, 5%: 13K, 10%: 26K (median bar trades 2.4M)

By period (segment)

period	bars	return	buy & hold	Sharpe	exposure
bars 60-794	735	-18.72%	-2.05%	-183.71	100%
bars 795-1529	735	-25.75%	-1.75%	-265.99	100%
bars 1530-2264	735	-25.32%	+0.59%	-255.26	100%
bars 2265-2999	735	-22.28%	+1.47%	-225.71	100%

By market regime

market regime	share of bars	strategy	market	Sharpe
rising	54%	-44.28%	-1.06%	-249.02
falling	46%	-36.51%	+1.53%	-225.95
calm	50%	-38.52%	+1.12%	-222.98
volatile	50%	-42.46%	-0.66%	-253.85

Regime of a bar: the market's trailing return and volatility over the previous 150 bars.

Reproduce

Costs: 5.00 bps commission + 5.00 bps slippage per side; fills at the next bar's open; long_short; warm-up 60 bars; n_trials 40; min_trades 30.

data f027f3d330505a357cb91cb7e31f49dd8b02f727f9db7ef9e4dafb0879384e2c
signals 45b3a6f9d973ea59894be14d75d67b1c5546b05091273887b0e6aa6d25d02235
source a86ca8eacaec7c18926b6d34f116bf4438b031bf18d95e84dfb56e99fde3cac1

Re-run with the same data and code: monte-neo verify --recheck CERT --ohlcv DATA --strategy STRATEGY.py

Checks backtest methodology (look-ahead, costs, selection bias), not future profit. Not investment advice.