Executive summary
\n\n\n\nThis post documents a systematic backtest run on 9 crypto assets from the PAXOS exchange, including Bitcoin (BTC), using hourly data and a multi-signal technical framework combining Bollinger Bands, Ichimoku, and Chande Oscillator logic.
\n\n\n\nBetween January 2022 and October 2025, the strategy compounded $100,000 into ~$485,000, delivering a 52.3% CAGR with a 15.5% maximum drawdown, under realistic constraints (leverage = 1, interest costs, position limits).
\n\n\n\nThis article explains:
\n\n\n\n- \n
- The strategy logic \n\n\n\n
- The portfolio & execution constraints \n\n\n\n
- The risk metrics that actually matter \n\n\n\n
- What the results mean (and what they don’t) \n
\n\n\n\n\n⚠️ This is a research and engineering report, not financial advice. We would not trade it as-is: it is research only..
\n
Strategy overview
\n\n\n\nSignal architecture
\n\n\n\nThe system is rule-based and fully systematic, combining entry and exit confirmation from multiple independent indicators.
\n\n\n\nBuy logic
\n\n\n\n- \n
bband2a(volatility & mean reversion context) \n\n\n\nichimoku3a(trend regime & structure) \n
Sell logic
\n\n\n\n- \n
ichimoku3a(trend invalidation) \n\n\n\ncho3a(momentum exhaustion) \n
The goal is regime-aware trend capture, not scalping.
\n\n\n\n\n\n\n\n
Timeframe & universe
\n\n\n\n| Parameter | Value |
|---|---|
| Exchange | PAXOS |
| Assets | 9 crypto pairs BTC ETH LTC BCH AAVE LINK UNI MATIC SOL |
| Bar size | 1 hour |
| Simulation start | 2022-01-01 |
| End date | 2025-10-03 |
| Base currency | USD |
\n\n\n\n
Portfolio & risk constraints
\n\n\n\nThis is not an unconstrained backtest.
\n\n\n\nKey constraints:
\n\n\n\n- \n
- Initial capital: $100,000 \n\n\n\n
- Leverage: 1.0 (spot-like behavior) \n\n\n\n
- Interest rate: 7% annual \n\n\n\n
- Target position weight: 40% \n\n\n\n
- Max position weight: 60% \n\n\n\n
- Trailing stop: 35% \n\n\n\n
- Max entries per bar: 10 \n\n\n\n
- Freeze logic: enabled (anti-overtrading) \n
The system is designed to survive real market conditions, not optimize paper metrics.
\n\n\n\nPerformance summary
\n\n\n\nCore metrics
\n\n\n\n| Metric | Value |
|---|---|
| Final equity | $485,138 |
| Total return | +385% |
| CAGR | 52.3% |
| Max drawdown | -15.5% |
| Sharpe | 1.74 |
| Sortino | 2.57 |
| Calmar | 3.37 |
| Stability | 0.97 |
| Trades | 214 |
| Win rate | 56.1% |
Benchmark CAGR assumed: 5%
\n\n\n\n\n\n\n\n
Why these numbers matter
\n\n\n\n- \n
- CAGR > 50% with DD < 16% → strong convexity \n\n\n\n
- Sharpe < 2 but Sortino > 2.5 → downside risk controlled \n\n\n\n
- Calmar > 3 → drawdown efficiency is the real alpha \n\n\n\n
- Stability ~0.97 → equity curve is not luck-driven noise \n
This is not a fragile, over-optimized curve.
\n\n\n\n\n\n\n\n
Return distribution (important)
\n\n\n\nKey distribution stats per trade (%):
\n\n\n\n- \n
- Worst trade: -13.49% \n\n\n\n
- 1% quantile: -8.80% \n\n\n\n
- Median: +0.17% \n\n\n\n
- 95% quantile: +17.03% \n\n\n\n
- Best trade: +58.8% \n
➡️ Losses are bounded, winners are fat-tailed.
This is exactly what a trend-following crypto system should look like.
\n\n\n\n
Exposure & execution behavior
\n\n\n\n- \n
- Net exposure: ~10x cumulative over time, not constant leverage \n\n\n\n
- Trades cluster during high-volatility regimes \n\n\n\n
- Flat or reduced exposure during chop → capital preservation \n
The strategy earns volatility, it does not fight it.
\n\n\n\n\n\n\n\n
What this backtest does not prove
\n\n\n\nLet’s be clear:
\n\n\n\n❌ No guarantee of future performance
❌ No claim of optimality
❌ No curve-fitting denial (all models have bias)
This is one robust configuration, validated across:
\n\n\n\n- \n
- Multiple assets \n\n\n\n
- Multi-year period \n\n\n\n
- Different volatility regimes \n
\n\n\n\n
Why this matters for KreamEdge
\n\n\n\nThis backtest is part of the KreamEdge research pipeline:
\n\n\n\n- \n
- Modular signal design \n\n\n\n
- Strict risk constraints \n\n\n\n
- Reproducible backtests \n\n\n\n
- Live-ready execution logic \n
The goal is signal quality over signal quantity.
\n\n\n\nMore research notes, live experiments, and tooling breakdowns will follow.
\n\n\n\n\n\n\n\n
Final thoughts
\n\n\n\nIf you’re serious about crypto trading:
\n\n\n\n- \n
- Focus on drawdown control \n\n\n\n
- Optimize for stability, not peak return \n\n\n\n
- Prefer simple signals + strong risk logic \n\n\n\n
- Always assume regime shifts \n
This system passes the first filter: it survives reality.
\n\n\n\nOut-of-sample cutoff. This strategy’s rules and parameters were frozen on 3 July 2025. All performance shown after that date is genuine out-of-sample / forward-tracked data – it post-dates the freeze, so no hindsight or selection could have shaped the rules.
\n\n\n\nJoin our free community channels for backtest output and strategy commentary.
\n\n\n\n
Frequently asked questions
\n\n\n\nWhich assets and timeframe does Backtest 106 cover?
\n\n\n\nNine crypto pairs from the PAXOS exchange, including BTC, ETH, LTC, BCH, AAVE, LINK, UNI, MATIC and SOL, on 1 hour bars from January 2022 to October 2025, starting from 100,000 USD of capital.
\n\n\n\nWhat were the results?
\n\n\n\nFinal equity of about 485,000 USD, a total return of 385%, a CAGR of 52.3% and a maximum drawdown of 15.5%. Sharpe was 1.74, Sortino 2.57 and Calmar 3.37 over 214 trades at a 56.1% win rate.
\n\n\n\nDoes the run use leverage?
\n\n\n\nNo. Leverage is fixed at 1.0, which is spot like behavior. Idle cash is charged 7% annual interest, the trailing stop is 35%, the target position weight is 40% and the maximum position weight is 60%.
\n\n\n\nWhat does the trade distribution look like?
\n\n\n\nLosses are bounded and winners are fat tailed. The worst trade is -13.5%, the 1st percentile is -8.8%, the median is +0.2%, the 95th percentile is +17.0% and the best trade is +58.8%.
\n\n\n\nWhy report Calmar rather than raw return?
\n\n\n\nBecause Calmar divides growth by the worst drawdown, which is the figure that decides whether a strategy is survivable in practice. A high CAGR quoted next to an undisclosed drawdown tells a reader nothing about risk.
\n\n\n\n\n\nRelated KreamEdge research
\n\n\n\n- \n\n
- A realistic systematic 1D crypto strategy \n\n\n\n
- Systematic 1D crypto strategy (Backtest 238) \n\n\n\n
- EUR 1H trading strategy (Backtest 85) \n\n\n\n
- Why most trading backtests are misleading (methodology) \n\n\n\n\n
- Position sizing null result \n\n\n\n
- Risk management after the AI rout: volatility and bias \n\n
0 Comments