Strategy library
Browse the research, including strategies that failed. Open a strategy to review its rules, results, and test conditions.
Research record⌄
15 research campaigns · 7M+ simulated trades → 17 strategy families, every one rejected. The simulator track is a separate population: 0 verified there.
280+ coins analyzed · 91 configs re-ranked daily per top-coin group · strategy families and research campaigns listed with their verdicts in the research ledger · real fees
We don't sell strategy count. Failures ship with full data — the kill rate is the product.
22 strategies documented — real market data, failures published alongside the rest.
+ 26 builder templates to start from → open the builder
How to read these metrics
Marks next to a metric: ~ = borderline (PF between break-even and 1.5; win rate ≤5 pts above break-even, or 50–55% if none shown) · ! = below the bar, or the strategy is already retired or shelved — a closed verdict outranks its number. No mark = clears it and is still standing.
No strategies match these filters.
● LIVE TRACK
1Trend Ensemble (Risk-Managed)
LIVE TRACKAbout this strategy⌄
Roughly halved Bitcoin's worst drawdown (−76.7% → −38.5%, BTC daily 2019-09 → 2025-12) — a multi-lookback trend overlay for holding Bitcoin, not beating it. Other windows cut less.
Crisis-alpha, not directional alpha — and we say so plainly.
◐ CONDITIONAL
1BB Squeeze SHORT
CONDITIONALConditional: not suitable for standalone use without regime awareness.
About this strategy⌄
Bollinger Band Squeeze detects volatility compression then enters short when expansion begins. Original Binance-era backtest: PF 2.22, WR 68.6%.
Re-classified CONDITIONAL on 2026-08-17 (3-value verdicts): both-window PF edge is real (IS 1.29 / OOS 1.08) but half-year walk-forward shows 2 structural loss halves — usable only with regime awareness, not standalone. The 2026-06-28 retirement (3 adversarial axes in CHOP) stands as history; see the full condition profile.
68.6%
break-even 49.6%
2.22
+$794
535
as published (1H · 2026-02)
Run settings · SL 10% · TP 8%
Simulate This →⟳ IN TESTING
6ATR Breakout LONG
TESTINGAbout this strategy⌄
Long when the previous close breaks above an ATR band with an EMA uptrend filter. 2026-08-14 live re-measure at preset defaults (sl 3 / tp 7, top-10, 1H): PF 1.08, +46.5%, MDD 36.5%, 648 trades.
Whole-window positive, but it has not passed the OOS + walk-forward bar — testing, not verified.
⚠ Not OOS-validated — research only
41.36%~
break-even 39.5%
1.08~
+46.5%
top-10, fixed nominal sizing
10
as published (1H · 2026-08-14)
Run settings · SL 3% · TP 7%
Simulate This →BB Squeeze LONG
TESTINGAbout this strategy⌄
Long-side variant of BB Squeeze. Retired in the 2026-02 full-universe run (PF 0.98); the 2026-07 re-measure at sl3/tp20 on the top 10 coins by market cap passed (PF 1.35) and it now ships as a curated simulator preset and passed OOS + walk-forward on 2026-07-14 — verified then; downgraded to testing on 2026-09-25 under stricter criteria.
⚠ Not OOS-validated — research only
40.16%
break-even 33.2%
1.35~
+145%
top-10, 2yr, non-compounded
10
as published (top coins · 1H · 2026-07-14)
Run settings · SL 3% · TP 20%
Simulate This →DCA Accumulation
TESTINGAbout this strategy⌄
Buy a fixed amount on a fixed schedule, regardless of price. No prediction, no timing — the most honest strategy there is.
It is market beta, not alpha, and we say so plainly. Not running on a live account at the moment; the mechanics are shown on the paper track.
⚠ Not OOS-validated — research only
Ichimoku Bearish
TESTINGAbout this strategy⌄
Short crypto below the Ichimoku cloud on a Tenkan/Kijun cross. Original 2026-05-04 backtest: 1,625 trades, PF 1.55 (IS→OOS 2.07→1.83).
Failed a fresh out-of-sample regime-robustness re-test on 2026-06-28 — the SHORT result is bear-beta. Backtest reproducible; live directional edge not confirmed.
⚠ Not OOS-validated — research only
51.7%
break-even 40.8%
1.55
50
50 coins · updated 2026-05-04
Run settings · SL 3% · TP 15%
Simulate This →Keltner Squeeze LONG (Bull Regime)
TESTINGAbout this strategy⌄
Long when Bollinger Bands compress inside Keltner Channels then break upward. Card metrics are the full-window run at the card's own settings (4H, 50 coins, sl15/tp30) — the same run the Simulate button reproduces.
The earlier headline PF 2.28 (367 trades, OOS 2.96, measured 2026-05-04) was a bull-regime subset: true for that window, but no button setting selects it. Kept below as history.
⚠ Not OOS-validated — research only
43.62%~
break-even 40.9%
1.12~
50
as published (top coins · 4H · 2026-08-23)
Run settings · SL 15% · TP 30%
Simulate This →MA Cross (50/200 EMA)
TESTINGAbout this strategy⌄
Classic 50/200 EMA crossover. Original 2026-04-22 backtest: 1,111 trades, PF 1.09, the lowest max drawdown (33.5%) of the strategies measured that day.
Failed a fresh out-of-sample regime-robustness re-test on 2026-06-28. Backtest reproducible; live directional edge not confirmed.
⚠ Not OOS-validated — research only
47.79%~
break-even 45.6%
1.09~
10
as published (top coins · 4H · 2026-04-22)
Run settings · SL 5% · TP 10%
Simulate This →✕ RETIRED
12▾
✕ RETIRED
12▾ADX Trend Short
RETIREDAbout this strategy⌄
Short when ADX > 25 confirms a strong downtrend and DMI- crosses above DMI+. Original 2026-05-04 backtest: 487 trades, PF 1.68, OOS/IS ratio 1.01.
Failed a fresh out-of-sample regime-robustness re-test on 2026-06-28 — the SHORT result is bear-beta. Backtest reproducible; live directional edge not confirmed.
65.5%!
break-even 53.1%
1.68!
50
50 coins · updated 2026-05-04
Donchian Breakout
RETIREDAbout this strategy⌄
Turtle Trading 20-period channel breakout. SHORT showed PF 1.27 (bull) / 1.06 (bear) in the original 2026-04 backtest, but failed a fresh out-of-sample regime-robustness re-test on 2026-06-28 (bear-beta artifact).
Backtest is reproducible; the live directional edge is not.
1.27!
updated 2026-04-01
Heikin Ashi Trend
RETIREDAbout this strategy⌄
Enter after 3 consecutive Heikin Ashi candles with no wick against the trend. 2026-08-14 live re-measure at preset defaults (sl 7 / tp 8, top-10, 1H): PF 0.91, 8,944 trades, -100% — account ruin.
Retired; published for education, not for trading.
46.85%!
break-even 49.2%
0.91!
-100%
account ruin, top-10, fixed nominal sizing
10
as published (1H · 2026-08-14)
Keltner Squeeze SHORT
RETIREDAbout this strategy⌄
Short the lower-band breakout after a Keltner squeeze. Original 2026-05-04 backtest: 923 trades, PF 1.61.
A fresh out-of-sample re-test on 2026-06-28 failed regime-robustness — the SHORT result read as bear-beta, not a standalone edge. Backtest reproducible; live directional edge not confirmed.
39.9%!
break-even 29.2%
1.61!
50
as published (4H · 2026-05-04)
MACD Cross SHORT
RETIREDAbout this strategy⌄
Short on a MACD dead cross above the zero line. 2026-08-14 live re-measure at preset defaults (sl 12 / tp 10, top-10, 1H): PF 0.83, 2,656 trades, -100% — account ruin.
Retired; published for education, not for trading.
47.74%!
break-even 52.4%
0.83!
-100%
account ruin, top-10, fixed nominal sizing
10
as published (1H · 2026-08-14)
Mean Reversion LONG
RETIREDAbout this strategy⌄
Buy when price breaks 2.0σ below SMA(20) with RSI under 30. 2026-08-14 live re-measure at preset defaults (sl 10 / tp 8, top-10, 1H): PF 0.71, 1,455 trades, -100% — account ruin.
Retired; published for education, not for trading.
48.18%!
break-even 56.7%
0.71!
-100%
account ruin, top-10, fixed nominal sizing
10
as published (1H · 2026-08-14)
Momentum Breakout LONG
RETIREDAbout this strategy⌄
Enters long when price breaks above the highest close of the last 20 candles with volume confirmation. Retired after failing 2-year validation — the in-sample gain was curve-fit and collapsed out-of-sample (negative expectancy in the original Binance-era run).
37.5%!
break-even 58.8%
0.42!
Negative
updated 2026-02-05
RSI Divergence
RETIREDAbout this strategy⌄
Long on bullish RSI divergence, short on bearish. Retired 2026-07-12; this page was published 2026-08-14 as a postmortem.
The 2026-08-14 live re-measure at preset defaults (sl 7 / tp 5, top-10, 1H) confirmed it: PF 0.67, 1,491 trades, -100% — account ruin. Published for education, not for trading.
48.02%!
break-even 58.0%
0.67!
-100%
account ruin, top-10, fixed nominal sizing
10
as published (1H · 2026-08-14)
Stochastic RSI Short
RETIREDAbout this strategy⌄
Short when Stochastic RSI crosses bearishly in overbought territory on 12H timeframe. Original backtest (2026-05-04): 2,674 trades, PF 1.22, OOS/IS ratio 1.38.
Retired 2026-08-08 on a fresh-universe run that reversed the sign — those kill-run numbers came from a pre-repair engine and have not reproduced since; see the re-measurement note.
53.5%!
break-even 48.5%
1.22!
50
as published (12H · 2026-05-04)
SuperTrend LONG
RETIREDAbout this strategy⌄
Long when the SuperTrend flips bullish (ATR(10) × 3.0 bands). 2026-08-14 live re-measure at preset defaults (sl 5 / tp 12, top-10, 1H): PF 0.92, 2,184 trades, -100% — account ruin.
Retired; published for education, not for trading.
44.09%!
break-even 46.2%
0.92!
-100%
account ruin, top-10, fixed nominal sizing
10
as published (1H · 2026-08-14)
Volume Profile POC
RETIREDAbout this strategy⌄
Mean reversion to Volume Profile Point of Control. Enters when price deviates >3% from POC.
Original 2026-03-27 backtest: PF 1.14, WR 53% on 10 coins. Sample is below our n≥30 standard — too small to confirm an edge.
Backtest reproducible; edge not confirmed.
53%!
break-even 49.7%
1.14!
10
as published (1H · 2026-03-27)
VRP Short Volatility
RETIREDAbout this strategy⌄
Sell volatility (defined-risk iron condor) to harvest the volatility risk premium — IV consistently exceeds realized vol. The first NON-directional, non-beta edge we found after every directional strategy reduced to market beta — and then our own real-quote backtest killed it.
RETIRED 2026-07-22: on Deribit bid/ask, 7DTE iron condors earn a negative Sharpe before cost. The premium is measured and real; this vehicle for harvesting it is not.
⬜ SHELVED
2▾
⬜ SHELVED
2▾ATR Breakout SHORT
SHELVEDAbout this strategy⌄
Short when an ATR spike signals a downside volatility expansion. The one conditional survivor of the 2026-06-28 fresh out-of-sample re-test — a genuine altcoin-short timing edge — but a fine-grained walk-forward shows −321% / 62% MDD in 2024-H1, so it is unsuitable for unsupervised real-money trading. 2,839 trades, PF 1.42.
⚠ Not OOS-validated — research only
35.9%!
break-even 28.3%
1.42!
50
50 coins · updated 2026-05-04
HV Squeeze
SHELVEDAbout this strategy⌄
Same Bollinger Band squeeze detection as BB Squeeze with a lighter filter set (candle color, 1.5x volume). Shelved — it covers 85.9% of BB Squeeze's setups and fires 5.8x as often, with no additional edge.
⚠ Not OOS-validated — research only
58.1%!
break-even 49.4%
1.42!
577
as published (1H · 2026-01)
Why show failed strategies? Radical transparency — we publish everything, wins and losses.
Build your own: browse templates
BUILDER TEMPLATES
All Builder Templates 26
Click any template to open it in the builder with its parameters filled in. Then pick the coins to run it on.
BB Squeeze LONG
LONGClick to simulate →RSI Reversal LONG
LONGClick to simulate →MACD Momentum LONG
LONGClick to simulate →Grid Mean Reversion LONG
LONGClick to simulate →DCA Oversold LONG
LONGClick to simulate →BB Band Bounce LONG
LONGClick to simulate →EMA Crossover LONG
LONGClick to simulate →ADX Trend LONG
LONGClick to simulate →RSI+BB Oversold LONG
LONGClick to simulate →Turtle Breakout LONG
LONGClick to simulate →HV Squeeze Breakout LONG
LONGClick to simulate →Ichimoku Cloud Bullish
LONGClick to simulate →Parabolic SAR Reversal LONG
LONGClick to simulate →Williams %R Oversold LONG
LONGClick to simulate →BB Squeeze SHORT
SHORTClick to simulate →Stochastic Overbought SHORT
SHORTClick to simulate →Stoch RSI Overbought SHORT
SHORTClick to simulate →ADX Trend SHORT
SHORTClick to simulate →RSI+BB Overbought SHORT
SHORTClick to simulate →Turtle Breakout SHORT
SHORTClick to simulate →MACD Crossover SHORT
SHORTClick to simulate →EMA Crossover SHORT
SHORTClick to simulate →HV Squeeze Breakout SHORT
SHORTClick to simulate →Ichimoku Cloud Bearish
SHORTClick to simulate →Parabolic SAR Reversal SHORT
SHORTClick to simulate →Williams %R Overbought SHORT
SHORTClick to simulate →MORE STRATEGIES COMING
We continuously test new approaches. Documented strategies will appear here as they complete backtesting validation.