Negative after costs

DOGEUSDT · 60m · Verdict Card

Percentile-deviation fade (mean reversion) · mr_percentile · hold 24 · 2025-06-01 → 2026-10-11 · 11929 bars · bybit · 2026-10-12 12:48

Rule: Fade when the 24-bar log deviation |d| ≥ its rolling p80 percentile (threshold recomputed every 168 bars); long & short
Cost model: 5.5bp/side fee + 3bp/side slippage + 0.07bp/bar funding

Trades

129

Win rate

50.4%

Avg / trade

-9bp

Gross avg

10bp

Total

-1119bp

t-stat

-0.22

Permutation p

0.4043 · T0.361/S0.404

95% CI of mean

[-83bp, 69bp]

Max drawdown

40.5%

Exposure

31.2%

Profit factor

0.95

Worst trade

-2309bp

Median

3bp

Equity curve (log)

Buy & hold: -55.3% · missing 0 (max gap 0) · OHLC complete

Stability by segment

Seg	From → To	Trades	Avg / trade	Win rate	t-stat
Seg 1	2025-08-23 → 2025-11-14	33	-66bp	48%	-0.65
Seg 2	2025-11-17 → 2026-02-16	33	-74bp	42%	-1.01
Seg 3	2026-02-23 → 2026-06-15	33	102bp	61%	1.79
Seg 4	2026-06-18 → 2026-10-08	30	5bp	50%	0.06

In-sample/Out-of-sample	Trades	Avg / trade	Win rate	t-stat
In-sample 60%	77	-35bp	49%	-0.63
Out-of-sample	52	31bp	52%	0.58

Trade log · last 12 trades

Time	Side	Entry px	Exit px	Bars held	Exit	Net
2026-10-08 23:00	LONG	0.08261	0.0846	24	time	219.4bp
2026-10-07 10:00	LONG	0.09001	0.08905	24	time	-125.9bp
2026-10-03 16:00	LONG	0.09246	0.09312	24	time	52.4bp
2026-09-28 15:00	LONG	0.09246	0.09501	24	time	253.4bp
2026-09-25 02:00	SHORT	0.09668	0.09754	24	time	-107.2bp
2026-09-23 23:00	LONG	0.09268	0.09563	24	time	294.7bp
2026-09-22 22:00	SHORT	0.09995	0.09481	24	time	509.3bp
2026-09-21 21:00	SHORT	0.09323	0.10094	24	time	-813.2bp
2026-09-20 20:00	LONG	0.08511	0.09345	24	time	916.1bp
2026-09-18 17:00	SHORT	0.08569	0.08724	24	time	-197.9bp
2026-09-16 02:00	LONG	0.08074	0.07952	24	time	-170.9bp
2026-09-10 06:00	LONG	0.08551	0.08349	24	time	-257.7bp

Method & known limits

Exit: fixed 24-bar hold; fill=signal bar close
Costs: 5.5bp/side fee + 3bp/side slippage + 0.07bp/bar funding drag (same both sides; funding sign not modeled)
Non-overlapping: signals during open positions are dropped
Significance: t-stat on per-trade net returns. Two nulls — (1) timing permutation: entry times randomized, direction mix kept, same trade count, same non-overlap suppression, p=0.3613; (2) sign-flip permutation: each trade's realized move kept, direction re-rolled (variance-matched, stricter for fades), p=0.4043. Verdict uses the stricter p=0.4043 (2000 trials, seed 42, reproducible)
The null must match the live procedure trade-for-trade: every surrogate fills the same count (129 entries, 2000 effective trials). A surrogate that only fills ~9 entries widens the null about 4x and can fake p=0.30 out of p=0.01
Uncertainty: mean 95% bootstrap CI [-83bp, 69bp] — spanning 0 means no claim of edge
Sample: 11929 bars of 60m, 0 missing (max gap 0), OHLC complete
Splits: first 60% in-sample, rest out-of-sample, plus 4 time blocks to catch single-segment flukes
Params: k=24 pct=80 side=both warmup=2000 rebalance=168 — if tuned on this same history the verdict carries selection bias; collect out-of-sample before believing it
Known limits: no order-book depth so slippage is assumed; liquidation/margin/liquidity not modeled; rejected or partial fills not modeled
Null sensitivity: the sign-flip test keeps each trade's realized move and re-rolls only the direction, so the long/short allocation itself counts as evidence — in a one-way market that can inflate significance. Hence the verdict requires both nulls to clear, a positive net mean and a positive out-of-sample mean; miss one and it degrades to "not separated from luck"

This card only judges whether a rule has statistical evidence under stated costs. Not investment advice; past results do not predict future returns.