Volatility: 26 Backtests

We backtested 26 volatility strategies across crypto, ETFs, US stocks. 2 beat buy-and-hold before correction; best +3.3 pts/yr.

Every result below is the strategy's annualized return minus buy-and-hold on the same assets and days, charged real trading costs and scored only on data up to the discovery cutoff. See the full family index, the full method and every result, or download the raw registry (JSON).

0 of these 26 strategies were significant at p < 0.05 before any correction for multiple testing, and 2 of 26 beat buy-and-hold on the raw excess-return number across crypto, ETFs, US stocks. Across the whole registry we ran 260 tests and applied one Benjamini-Hochberg correction; only 0 of those survived it, which is why a single significant p-value inside one family isn't treated as a working edge here — the correction is explained in full on the evidence page. This family was last re-scored on 2026-09-18, using only data up to the 2025-03-15 discovery cutoff — the same freeze point every family in the registry is held to, so none of them can see data the others couldn't. Volatility is tested across 3 asset classes (crypto, ETFs, US stocks); a strategy only joins this family because of the signal it trades on, not the asset it's tested against.

• Inverse-volatility risk parity across crypto majors (crypto): +3.3 pts/yr, behind buy-and-hold, p=0.336. Source: Qian, E. (2005) 'Risk Parity Portfolios', PanAgora.
• Volatility targeting position scaling on trend signal (crypto): +3.0 pts/yr, behind buy-and-hold, p=0.443. Source: Moskowitz, Ooi, Pedersen (2012) JFE 'Time Series Momentum'.
• Inverse realized-vol percentile position scaling (crypto cross-sectional) (wide US stocks) (US stocks): −0.2 pts/yr, behind buy-and-hold, p=1.000. Source: Asness, Frazzini, Pedersen 2012 'Leverage Aversion and Risk Parity'.
• Inverse-volatility risk parity across crypto majors (wide US stocks) (US stocks): −2.4 pts/yr, behind buy-and-hold, p=1.000. Source: Qian, E. (2005) 'Risk Parity Portfolios', PanAgora.
• Realized-vol skew (down-vol vs up-vol) crash-risk timing (crypto): −2.5 pts/yr, behind buy-and-hold, p=1.000. Source: Ang, Chen, Xing 2006 'Downside Risk' RFS.
• Van Tharp fixed-% risk with ATR chandelier trailing exit (wide US stocks) (US stocks): −2.9 pts/yr, behind buy-and-hold, p=1.000. Source: Van Tharp, 'Trade Your Way to Financial Freedom' (1998).
• Realized-vol skew (down-vol vs up-vol) crash-risk timing (wide US stocks) (US stocks): −3.6 pts/yr, behind buy-and-hold, p=1.000. Source: Ang, Chen, Xing 2006 'Downside Risk' RFS.
• Vol-managed exposure: long QQQ only in low-realized-vol regime (ETFs): −6.6 pts/yr, behind buy-and-hold, p=1.000. Source: Moreira & Muir 2017 JF (Volatility-Managed Portfolios).
• VIX futures contango/backwardation SPY timing (ETFs): −11.5 pts/yr, behind buy-and-hold, p=1.000. Source: CBOE; various QuantConnect community backtests of VIX contango timing.
• VIX futures contango/backwardation SPY timing (wide US stocks) (US stocks): −12.3 pts/yr, behind buy-and-hold, p=1.000. Source: CBOE; various QuantConnect community backtests of VIX contango timing.
• Minervini VCP (volatility contraction pattern) (wide US stocks) (US stocks): −12.7 pts/yr, behind buy-and-hold, p=1.000. Source: Mark Minervini SEPA/VCP.
• VIX spike mean-reversion into SPY (ETFs): −13.1 pts/yr, behind buy-and-hold, p=1.000. Source: Simon & Wiggins 2001 J. Futures Markets; CBOE VIX spike studies.
• Larry Williams %-range volatility breakout (0.7x) (wide US stocks) (US stocks): −15.2 pts/yr, behind buy-and-hold, p=1.000. Source: Larry Williams, 'Long-Term Secrets to Short-Term Trading' (1999). Same rule as trader-books-larry-williams-volatility-breakout; separate catalogue entry, run independently at the same fixed params. Adapted for close-to-close: breakout confirmed by today's high is held into tomorrow's close-to-close return instead of closed out same day (the harness has no intraday exit)..
• Larry Williams volatility breakout (day-trade) (wide US stocks) (US stocks): −15.2 pts/yr, behind buy-and-hold, p=1.000. Source: Larry Williams, 'Long-Term Secrets to Short-Term Trading' (1999). Adapted for the harness's close-to-close position model: the book's same-day open-to-close day-trade is not representable without intraday data, so a breakout confirmed by today's high is held into tomorrow's close-to-close return instead of closed out same day..
• Minervini VCP (volatility contraction pattern) (US stocks): −15.7 pts/yr, behind buy-and-hold, p=1.000. Source: Mark Minervini SEPA/VCP.
• Van Tharp fixed-% risk with ATR chandelier trailing exit (crypto): −15.8 pts/yr, behind buy-and-hold, p=1.000. Source: Van Tharp, 'Trade Your Way to Financial Freedom' (1998).
• VIX spike mean-reversion into SPY (wide US stocks) (US stocks): −17.2 pts/yr, behind buy-and-hold, p=1.000. Source: Simon & Wiggins 2001 J. Futures Markets; CBOE VIX spike studies.
• Low-volatility anomaly (direct, 60m realized vol) (wide US stocks) (US stocks): −17.4 pts/yr, behind buy-and-hold, p=1.000. Source: Baker-Bradley-Wurgler / Frazzini-Pedersen low-vol anomaly.
• Inverse realized-vol percentile position scaling (crypto cross-sectional) (crypto): −18.1 pts/yr, behind buy-and-hold, p=1.000. Source: Asness, Frazzini, Pedersen 2012 'Leverage Aversion and Risk Parity'.
• Low-volatility anomaly (direct, 60m realized vol) (US stocks): −20.9 pts/yr, behind buy-and-hold, p=1.000. Source: Baker-Bradley-Wurgler / Frazzini-Pedersen low-vol anomaly.
• Low-realized-vol regime filter for trend entries (crypto): −50.0 pts/yr, behind buy-and-hold, p=1.000. Source: Ang, Hodrick, Xing, Zhang 2006 JF 'The Cross-Section of Volatility and Expected Returns'.
• 10% annualized volatility targeting overlay (crypto): −52.0 pts/yr, behind buy-and-hold, p=1.000. Source: Moreira & Muir 2017 JF; EDGE RESEARCH.md §12 (SMA-trend variant already failed there).
• Larry Williams %-range volatility breakout (0.7x) (crypto): −61.6 pts/yr, behind buy-and-hold, p=1.000. Source: Larry Williams, 'Long-Term Secrets to Short-Term Trading' (1999). Same rule as trader-books-larry-williams-volatility-breakout; separate catalogue entry, run independently at the same fixed params. Adapted for close-to-close: breakout confirmed by today's high is held into tomorrow's close-to-close return instead of closed out same day (the harness has no intraday exit)..
• Larry Williams volatility breakout (day-trade) (crypto): −61.6 pts/yr, behind buy-and-hold, p=1.000. Source: Larry Williams, 'Long-Term Secrets to Short-Term Trading' (1999). Adapted for the harness's close-to-close position model: the book's same-day open-to-close day-trade is not representable without intraday data, so a breakout confirmed by today's high is held into tomorrow's close-to-close return instead of closed out same day..
• ATR-normalized breakout, 1h bars (crypto): −74.8 pts/yr, behind buy-and-hold, p=1.000. Source: EDGE RESEARCH.md §21/§23/§42 (ATR_MULT=1.5, RR=3, 1h) — retested here in hunt2's harness.
• ATR percentile filter on breakout entries (crypto): −83.6 pts/yr, behind buy-and-hold, p=1.000. Source: builds on EDGE RESEARCH.md §21/§23/§42 ATR breakout base rule, adding a percentile gate.

Related families: Trend, Mean reversion, Momentum.