H-005: SMA50 spot long/flat beats buy-and-hold on 8 independent coins

Fail Family: trend · Status: rejected

Does a 50-day moving average strategy work on altcoins? The idea did not beat simply buying the coin and holding it. It is rejected.

Result

full: wins 4/8 (need 6), beats buy-and-hold 4/8; recent: wins 3/8 (need 4), beats buy-and-hold 3/8

Sharpe: return per unit of risk (higher is better). CAGR: average yearly growth. Max drawdown: worst fall from a peak. p-value: chance the result is luck (lower is better). “Hold” columns are buy-and-hold for the same asset and days. Glossary

Asset	Period	Sharpe	Hold Sharpe	CAGR	Hold CAGR	Max drawdown	Hold drawdown	In market	p-value	Beats hold	Counts as win
BNBUSDT	full (2019-01-01 → 2026-10-13)	1.41	1.15	+104.7%	+85.2%	-40.7%	-76.1%	57%	0.002	yes	yes
BNBUSDT	recent (2022-01-01 → 2026-10-13)	0.62	0.42	+18.3%	+8.1%	-36.1%	-62.9%	52%	0.086	yes	yes
XRPUSDT	full (2019-01-01 → 2026-10-13)	0.55	0.65	+16.0%	+19.2%	-81.2%	-83.2%	46%	0.276	no	no
XRPUSDT	recent (2022-01-01 → 2026-10-13)	0.22	0.51	-2.4%	+11.0%	-72.7%	-72.0%	43%	0.547	no	no
ADAUSDT	full (2019-01-01 → 2026-10-13)	0.95	0.71	+53.4%	+25.4%	-75.6%	-95.2%	44%	0.011	yes	yes
ADAUSDT	recent (2022-01-01 → 2026-10-13)	0.04	0.02	-10.7%	-30.1%	-75.6%	-91.0%	36%	0.351	yes	yes
LTCUSDT	full (2019-01-01 → 2026-10-13)	0.36	0.55	+2.0%	+9.6%	-94.0%	-89.4%	50%	0.460	no	no
LTCUSDT	recent (2022-01-01 → 2026-10-13)	-0.53	0.12	-33.4%	-16.6%	-89.3%	-72.9%	45%	0.962	no	no
LINKUSDT	full (2019-03-07 → 2026-10-13)	0.86	0.95	+44.9%	+56.3%	-86.8%	-90.2%	53%	0.185	no	no
LINKUSDT	recent (2022-01-01 → 2026-10-13)	0.25	0.33	-2.3%	-8.3%	-76.7%	-81.7%	46%	0.352	no	no
DOGEUSDT	full (2019-08-24 → 2026-10-13)	0.82	0.77	+97.7%	+62.2%	-85.3%	-92.3%	44%	0.082	yes	yes
DOGEUSDT	recent (2022-01-01 → 2026-10-13)	0.33	0.27	+1.0%	-13.6%	-74.9%	-85.2%	40%	0.256	yes	yes
XLMUSDT	full (2019-01-01 → 2026-10-13)	0.56	0.54	+18.0%	+7.3%	-73.3%	-90.3%	43%	0.162	yes	yes
XLMUSDT	recent (2022-01-01 → 2026-10-13)	0.31	0.32	+0.6%	-6.8%	-68.4%	-75.7%	38%	0.288	no	no
TRXUSDT	full (2019-01-01 → 2026-10-13)	0.64	0.84	+25.9%	+44.4%	-78.2%	-78.2%	59%	0.426	no	no
TRXUSDT	recent (2022-01-01 → 2026-10-13)	0.26	0.74	+2.1%	+35.9%	-61.3%	-51.1%	60%	0.858	no	no

Equity curves

Growth of 1 unit, log scale, fees included. Only days on which the rule is fully defined are counted.

Asset and period:

The same rule on every coin

Exploratory check of this rule on all coins since 2022-01-01. It is not part of the verdict above.

✓works (passed the chance check)
↑better than holding, could be luck
✗worse than holding
·not enough data
original coin of the hypothesis

Coin	✓	Result	Sharpe: rule / holding	CAGR	From
BTC	↑	better than holding, could be luck	0.75 / 0.48	+21% / +12%	2022-01-01
ETH	↑	better than holding, could be luck	0.76 / 0.20	+27% / -9%	2022-01-01
SOL	↑	better than holding, could be luck	0.81 / 0.35	+37% / -10%	2022-01-01
BNB	↑	better than holding, could be luck	0.62 / 0.42	+18% / +8%	2022-01-01
XRP	✗	worse than holding	0.21 / 0.51	-2% / +10%	2022-01-01
ADA	↑	better than holding, could be luck	0.02 / -0.01	-11% / -32%	2022-01-01
LTC	✗	worse than holding	-0.57 / 0.11	-35% / -18%	2022-01-01
LINK	✗	worse than holding	0.24 / 0.30	-3% / -10%	2022-01-01
DOGE	↑	better than holding, could be luck	0.33 / 0.26	+1% / -15%	2022-01-01
XLM	↑	better than holding, could be luck	0.30 / 0.30	+1% / -9%	2022-01-01
TRX	✗	worse than holding	0.26 / 0.74	+2% / +36%	2022-01-01
EVER	↑	better than holding, could be luck	0.09 / -0.23	-20% / -72%	2022-05-07
GRAM	↑	better than holding, could be luck	0.54 / 0.18	+16% / -14%	2023-05-02

Why it might work

Confirmation of the rule fixed after H-003 on coins that played no part in choosing it.

Verdict history

2026-09-24: Fail (F1MEBOT (2026))
2026-10-14: Fail (this engine, data to 2026-10-13)

Pre-registered specification

kind: timing_rule
rule:
  type: price_above_sma
  window: 50
mode: long_flat
fee: 0.001
universe:
- BNBUSDT
- XRPUSDT
- ADAUSDT
- LTCUSDT
- LINKUSDT
- DOGEUSDT
- XLMUSDT
- TRXUSDT
periods:
- name: full
  start: '2019-01-01'
- name: recent
  start: '2022-01-01'
significance:
  method: circular_shift
  n: 1000
  alpha: 0.05
criterion:
  min_wins:
    full: 6
    recent: 4
  require_significance: false