Crypto Research Labs
ZH ▾
GitHub

← 全部假设

H-020: 当 7 日交易所净存入处于最高 20% 时 BTC 空仓

通过  类别: 资金流 · 状态: 通过

该假设通过了预注册检验。这意味着应当进行独立验证和模拟交易,而不是用真金白银交易。

比特币流入交易所能预测价格下跌吗? 该想法达到了检验前设定的标准,值得独立复核。这不代表一定盈利。

结果

技术细节由引擎生成,以英文显示。

oos: wins 1/1 (need 1), beats buy-and-hold 1/1

Sharpe:单位风险的收益(越高越好)。CAGR:年均增长率。最大回撤:从峰值的最大跌幅。p 值:结果来自运气的概率(越低越好)。“持有”列为同一资产、同一时间段的买入持有结果。 术语

资产区间Sharpe持有 SharpeCAGR持有 CAGR最大回撤持有回撤持仓时间p 值跑赢持有计为成功
BTCUSDToos (2022-01-01 → 2026-10-13)0.830.50+30.6%+13.1%-44.2%-66.9%83%0.010是是

资金曲线

1 个单位的增长,对数坐标,已计入手续费。仅统计规则完全确定的交易日。

同一规则在所有币上的表现

自 2022-01-01 起对该规则在所有币上的探索性检验。不属于上面的结论。

币种✓结果夏普:规则 / 持有CAGR起始
BTC↑优于持有,可能是运气0.80 / 0.48+29% / +12%2022-01-01
ETH↑优于持有,可能是运气0.55 / 0.20+17% / -9%2022-01-01
SOL↑优于持有,可能是运气0.55 / 0.35+14% / -10%2022-01-01
BNB↑优于持有,可能是运气0.44 / 0.42+10% / +8%2022-01-01
XRP↑优于持有,可能是运气0.62 / 0.51+23% / +10%2022-01-01
ADA↑优于持有,可能是运气0.08 / -0.01-18% / -32%2022-01-01
LTC↑优于持有,可能是运气0.34 / 0.11+1% / -18%2022-01-01
LINK↑优于持有,可能是运气0.31 / 0.30-5% / -10%2022-01-01
DOGE↑优于持有,可能是运气0.52 / 0.26+11% / -15%2022-01-01
XLM↑优于持有,可能是运气0.47 / 0.30+9% / -9%2022-01-01
TRX↑优于持有,可能是运气0.80 / 0.74+40% / +36%2022-01-01
EVER✗不如持有-0.21 / -0.15-65% / -68%2022-03-19
GRAM✗不如持有-0.26 / 0.13-32% / -17%2023-03-14

为什么可能有效

Idea by F1DevBin: coins moved to exchanges before price moves carry information. Deposits usually precede selling, so heavy net inflow should precede weaker returns (CryptoQuant and others read large inflows as bearish). Data: Coin Metrics community exchange flows (free, full history since 2011 for BTC, 2015 for ETH); collected daily into datasets/exchange_flows.csv. Spec revised on 2026-10-14 before any test was run: the first version waited a year for data, until it turned out the full history is available.

结论历史

预注册规范

kind: timing_rule
rule:
  type: feature_vs_rolling_quantile
  feature: exnetflow_btc_7d
  q: 0.8
  long_when: not_above
  lookback: 365
  min_periods: 180
  lag: 1
mode: long_flat
fee: 0.001
universe:
- BTCUSDT
periods:
- name: oos
  start: '2022-01-01'
significance:
  method: circular_shift
  n: 2000
  alpha: 0.025
criterion:
  min_wins:
    oos: 1
  require_significance: true