Strategies

Every trading idea, the reason it should work, and how far it's gotten through the 5-step proving gauntlet.

Experiments
trend_pullback

Auto-research sweep of trend_pullback

In testing

Why it should work: Confirmed uptrends resume more often than not; buying dips to the MA has positive expectancy.

1. Shows an edge (after costs & tax)2. Holds up on unseen data3. Survives worst-case luck4. Beats just holding BTC & cash5. Proven in live paper-trading
0 parameter combinations tested — the more you try, the higher the bar a result must clear to count as real (guards against false positives).
breakout

Auto-research sweep of breakout

In testing

Why it should work: Range breaks on momentum tend to continue as new participants chase the move.

1. Shows an edge (after costs & tax)2. Holds up on unseen data3. Survives worst-case luck4. Beats just holding BTC & cash5. Proven in live paper-trading
0 parameter combinations tested — the more you try, the higher the bar a result must clear to count as real (guards against false positives).
mean_reversion

Auto-research sweep of mean_reversion

In testing

Why it should work: In range-bound regimes, extreme RSI readings revert.

1. Shows an edge (after costs & tax)2. Holds up on unseen data3. Survives worst-case luck4. Beats just holding BTC & cash5. Proven in live paper-trading
0 parameter combinations tested — the more you try, the higher the bar a result must clear to count as real (guards against false positives).
xsec_momentum

Auto-research sweep of xsec_momentum

In testing

Why it should work: Relative strength persists; capital chases recent winners (a documented anomaly).

1. Shows an edge (after costs & tax)2. Holds up on unseen data3. Survives worst-case luck4. Beats just holding BTC & cash5. Proven in live paper-trading
0 parameter combinations tested — the more you try, the higher the bar a result must clear to count as real (guards against false positives).
trend_pullback

Auto-research sweep of trend_pullback

In testing

Why it should work: Confirmed uptrends resume more often than not; buying dips to the MA has positive expectancy.

1. Shows an edge (after costs & tax)2. Holds up on unseen data3. Survives worst-case luck4. Beats just holding BTC & cash5. Proven in live paper-trading
0 parameter combinations tested — the more you try, the higher the bar a result must clear to count as real (guards against false positives).