Negative after costs

BTCUSDT · 60m · Verdict Card

RSI overbought/oversold fade · rsi_fade · hold 24 · 2025-06-01 → 2026-10-11 · 11929 bars · bybit · 2026-10-12 12:48

Rule: Long when RSI(14) ≤ 30, short when RSI(14) ≥ 70
Cost model: 5.5bp/side fee + 3bp/side slippage + 0.06bp/bar funding

Trades

164

Win rate

43.3%

Avg / trade

-42bp

Gross avg

-24bp

Total

-6892bp

t-stat

-2.39

Permutation p

0.9115 · T0.912/S0.902

95% CI of mean

[-76bp, -6bp]

Max drawdown

55.4%

Exposure

39.6%

Profit factor

0.60

Worst trade

-967bp

Median

-39bp

Equity curve (log)

Buy & hold: -20.6% · missing 0 (max gap 0) · OHLC complete

Stability by segment

Seg	From → To	Trades	Avg / trade	Win rate	t-stat
Seg 1	2025-08-25 → 2025-12-01	41	-66bp	37%	-2.12
Seg 2	2025-12-02 → 2026-03-16	41	-44bp	46%	-1.09
Seg 3	2026-03-17 → 2026-07-02	41	-41bp	37%	-1.27
Seg 4	2026-07-04 → 2026-10-08	41	-17bp	54%	-0.45

In-sample/Out-of-sample	Trades	Avg / trade	Win rate	t-stat
In-sample 60%	98	-45bp	42%	-1.95
Out-of-sample	66	-38bp	45%	-1.38

Trade log · last 12 trades

Time	Side	Entry px	Exit px	Bars held	Exit	Net
2026-10-08 12:00	LONG	82606.9	82208.7	24	time	-66.8bp
2026-10-07 09:00	LONG	84328	83062.3	24	time	-169.7bp
2026-10-04 19:00	SHORT	85273.1	86171.8	24	time	-123.3bp
2026-10-02 11:00	SHORT	85483.7	84647.9	24	time	79.8bp
2026-09-30 20:00	SHORT	85262.9	83703.4	24	time	166.2bp
2026-09-27 17:00	SHORT	84923	82645.4	24	time	253.4bp
2026-09-23 12:00	SHORT	87132.1	83997.4	24	time	348bp
2026-09-21 16:00	SHORT	83724.2	86171.4	24	time	-306.5bp
2026-09-19 16:00	SHORT	81291.6	80228.3	24	time	113.2bp
2026-09-18 11:00	SHORT	77343.7	81094	24	time	-491.9bp
2026-09-15 00:00	SHORT	78731.3	76346.3	24	time	289.2bp
2026-09-13 18:00	LONG	76644.8	77861.5	24	time	139.1bp

Method & known limits

Exit: fixed 24-bar hold; fill=signal bar close
Costs: 5.5bp/side fee + 3bp/side slippage + 0.06bp/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=0.9115; (2) sign-flip permutation: each trade's realized move kept, direction re-rolled (variance-matched, stricter for fades), p=0.9015. Verdict uses the stricter p=0.9115 (2000 trials, seed 42, reproducible)
The null must match the live procedure trade-for-trade: every surrogate fills the same count (164 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 [-76bp, -6bp] — 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: period=14 lo=30 hi=70 side=both warmup=2000 — 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.