Choose a strategy, adjust its settings, and test it on historical data.
280 coins · 19 presets · up to 2.9yr data · real fees
Choose a preset, review the settings, then run a simulation. Use Edit settings to adjust risk and dates, or open the builder for custom entry rules.
Click a preset. See a real simulation — every result gets an honest verdict, failures included.
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.
Select a preset to open its settings and result together. Close the result to return to your place in the list.
The “split sweep” row on a card is measured with different costs — fee 0.08%/leg x2, slippage 0.02%/leg x2, funding 0.01%/8h (SHORT receives, LONG pays), NET PnL.
The “measured” date on a card is when those figures were taken. The card is a snapshot from that date — clicking re-runs it on current data, so the result can differ.
product default-parameter run
Interval 1.20–1.70 stays above breakeven in this sample · 90% CI over 597 tradesproduct default-parameter run
product default-parameter run
product default-parameter run
product default-parameter run
Interval 1.14–1.56 stays above breakeven in this sample · 90% CI over 936 tradesproduct default-parameter run
None of the cards below passed verification at its shown settings — 1h candles, the SL/TP on the card, the top-10 coins, 5x leverage, fees included. The badge says why: ✗ Washout lost in both windows · ◐ Conditional won only in some · Shelved was dropped as redundant, not as a failure. Change any of those settings and the verdict can move.
Same engine, fees and data as the winners above. “Retired” means this exact setup did not clear our verification bar over the full backtest — most lost money, and the rest won only in one window or broke out-of-sample. It does not mean the idea can never work. Some even look profitable if you widen the take-profit, but that gain fails our out-of-sample test (it's curve-fit, not real). Coin count matters too: an edge on the top 10 coins by market cap can thin out as you add more. Click one to reproduce its result, or load it and change the settings yourself. Don't believe. Verify.
product default-parameter run
product default-parameter run
Bull IS 2023-12-30 → 2025-07-07 · bear OOS 2025-07-07 → 2026-07-12. The 25/40 denominator is the measured coin universe (40 coins) — a different number from the card's trade count n. Edges in both windows, yet the half-year walk-forward shows 2 of 7 adequately-sampled (n≥30) halves at a loss — interval losses the aggregate windows hide. A short edge in the bear window may be bear beta — the direction, not the strategy, may have won.
product default-parameter run
Interval 0.74–0.88 stays below breakeven in this sample · 90% CI over 2,839 tradesBull IS 2023-12-30 → 2025-07-07 · bear OOS 2025-07-07 → 2026-07-12. The 22/40 denominator is the measured coin universe (40 coins) — a different number from the card's trade count n. A short edge in the bear window may be bear beta — the direction, not the strategy, may have won.
product default-parameter run
Bull IS 2023-12-30 → 2025-07-07 · bear OOS 2025-07-07 → 2026-07-12. The 4/40 denominator is the measured coin universe (40 coins) — a different number from the card's trade count n.
product default-parameter run
Interval 0.86–1.00 stays below breakeven in this sample · 90% CI over 3,205 tradesBull IS 2023-12-30 → 2025-07-07 · bear OOS 2025-07-07 → 2026-07-12. The 25/40 denominator is the measured coin universe (40 coins) — a different number from the card's trade count n. A short edge in the bear window may be bear beta — the direction, not the strategy, may have won.
product default-parameter run
Interval 0.77–0.89 stays below breakeven in this sample · 90% CI over 3,862 tradesBull IS 2023-12-30 → 2025-07-07 · bear OOS 2025-07-07 → 2026-07-12. The 16/40 denominator is the measured coin universe (40 coins) — a different number from the card's trade count n. A short edge in the bear window may be bear beta — the direction, not the strategy, may have won.
product default-parameter run
Bull IS 2023-12-30 → 2025-07-07 · bear OOS 2025-07-07 → 2026-07-12. The 11/40 denominator is the measured coin universe (40 coins) — a different number from the card's trade count n.
product default-parameter run
Interval 0.66–0.84 stays below breakeven in this sample · 90% CI over 1,511 tradesBull IS 2023-12-30 → 2025-07-07 · bear OOS 2025-07-07 → 2026-07-12. The 0/40 denominator is the measured coin universe (40 coins) — a different number from the card's trade count n.
product default-parameter run
product default-parameter run
Interval 0.61–0.76 stays below breakeven in this sample · 90% CI over 1,571 tradesBull IS 2023-12-30 → 2025-07-07 · bear OOS 2025-07-07 → 2026-07-12. The 1/40 denominator is the measured coin universe (40 coins) — a different number from the card's trade count n.
Bull IS 2023-12-30 → 2025-07-07 · bear OOS 2025-07-07 → 2026-07-12. The 17/40 denominator is the measured coin universe (40 coins) — a different number from the card's trade count n. A short edge in the bear window may be bear beta — the direction, not the strategy, may have won.
product default-parameter run
Interval 0.68–0.86 stays below breakeven in this sample · 90% CI over 1,387 tradesBull IS 2023-12-30 → 2025-07-07 · bear OOS 2025-07-07 → 2026-07-12. The 18/40 denominator is the measured coin universe (40 coins) — a different number from the card's trade count n. A short edge in the bear window may be bear beta — the direction, not the strategy, may have won.
product default-parameter run
Interval 0.88–0.96 stays below breakeven in this sample · 90% CI over 9,430 tradesBull IS 2023-12-30 → 2025-07-07 · bear OOS 2025-07-07 → 2026-07-12. The 1/40 denominator is the measured coin universe (40 coins) — a different number from the card's trade count n.
Bull IS 2023-12-30 → 2025-07-07 · bear OOS 2025-07-07 → 2026-07-12. The 29/40 denominator is the measured coin universe (40 coins) — a different number from the card's trade count n. A short edge in the bear window may be bear beta — the direction, not the strategy, may have won.
product default-parameter run
Interval 0.79–0.86 stays below breakeven in this sample · 90% CI over 13,308 tradesBull IS 2023-12-30 → 2025-07-07 · bear OOS 2025-07-07 → 2026-07-12. The 1/39 denominator is the measured coin universe (39 coins) — a different number from the card's trade count n.
Bull IS 2023-12-30 → 2025-07-07 · bear OOS 2025-07-07 → 2026-07-12. The 0/39 denominator is the measured coin universe (39 coins) — a different number from the card's trade count n.
Strategies are being re-validated. No live comparison is available; only backtest results are shown.
Only strategies that pass fresh out-of-sample validation qualify. Comparisons return after live trading resumes and at least 30 days of new data accumulate.
Auto-trading is coming soon — real-money execution is on hold. For now, you can test strategies in the simulator.
Measured results
8 profitable in this window · 11 unprofitable
Every result published — failures included
Scope: top 10 coins · 1H · 5x · fees 0.05% · as of 2026-10-06
Choose a preset or pick indicators
14 indicators, AND/OR logic
Set entry conditions and risk
SL, TP, time filters, position size
Run a backtest on the coins you choose
Top 10 by default · seconds · fees included
Simulation only — not real trading. Past performance does not guarantee future results. Futures trading involves risk of loss.
No code. No account. No cost. Results include realistic fees and slippage.