How-To Guide
Everything you need to know to get the most out of the platform.
1. Who is this for?
The long-term investor
You hold assets for months or years and want to know if a strategy has worked historically.
→ Recommendation: Weekly candles (W), 5–10 year range, Golden Cross or RSI/SMA Cross.
The mid-term swing trader
You hold positions for weeks to months. You want to spot trends early and exit on time.
→ Recommendation: Daily candles (D), 2–5 year range, RSI/SMA Cross or Stoch-RSI/SMA.
The strategy researcher
You want to understand which strategies worked best on which assets historically.
→ Recommendation: Compare multiple assets and strategies, use the Winners List.
The diversification planner
You want to diversify your portfolio across stocks, ETFs, gold, or currency pairs.
→ Recommendation: ETF and Commodities backtest (Pro/Elite), test counter-cyclical assets.
2. What does the platform offer?
Backtest Engine
Test historical trading strategies on real market data — without risking real money.
Asset classes:
Strategies (13 + reference):
• Free: RSI / SMA Cross · Golden Cross · RSI OB/OS · Buy & Hold · DCA
• Pro+: Stoch-RSI / SMA · OBV-MACD · WMA Trend Signal · EMA Trend Bias · Keltner Channel Breakout
• Pro+, Crypto: Fear & Greed Cadence · 30D SMA > Max Pain · BTC Signal RSI/SMA
→ Detail pages with methodology, pseudo-code, strengths/weaknesses per strategy at /strategies
Alerts (Pro+)
Email notifications when signals trigger.
Winners List
Backtests with high CAGR from other users.
Improvements
Feature proposals from the community.
3. Step by step: Running a backtest
Choose asset class
Go to "Backtest" and choose: Crypto, Stocks, ETFs, Commodities, or Forex.
Select asset
Select the asset — e.g. BTCUSDT for Bitcoin or AAPL.US for Apple. Use the search.
Choose strategy
Choose a trading strategy. Not sure which? Read the explanations in the Wiki.
Set parameters
Adjust RSI period, SMA period etc. Default values are good starting points.
Set time range and interval
Short (1–2y) = current, less significant. Long (5–10y) = multiple market phases. Weekly = less noise.
Enter starting capital
Virtual capital (e.g. €10,000). Results scale proportionally.
Run backtest
Click "Run backtest" — the engine fetches real market data and calculates all signals.
4. Understanding results
Key metrics:
CAGRAnnual growth rate. S&P 500 ≈ 10%, BTC in strong phases 50–100%+.
Win-RatePercentage of profitable trades. Alone not very meaningful.
Number of TradesMore trades = higher costs in reality (not simulated here).
Max. DrawdownLargest drop from peak. You must be able to handle this emotionally.
Good sign
- • CAGR significantly above benchmark
- • Max. Drawdown < 30–40%
- • Consistent across multiple periods
Warning sign
- • Only good in a short period (overfitting)
- • Many trades with small gains
- • Win-Rate < 40% with large losses
5. What to do with results?
Refine strategy
Test different parameters. Look for settings that work consistently across multiple periods.
Compare multiple assets
Does the strategy work better on BTCUSDT or gold? Does Golden Cross behave differently on stocks vs crypto?
Buy-and-hold as benchmark
If the backtest underperforms simple holding, the strategy adds no value.
Set up an alert
Found a strategy? Set up an alert (Pro+) — you'll be notified by email.
Enter the Winners List
Found a particularly strong combination? Enter it and compare with other users.
Further considerations
- Risk management: No backtest should be followed blindly. Set stop-losses.
- Diversification: Combine multiple assets and strategies.
- Transaction costs: The engine doesn't simulate fees. In reality, every trade costs money.
6. FAQ
Is the data real market data?
Yes. Crypto data from Binance, Stocks/ETFs/Forex/Commodities from EODHD — a professional financial data platform with 30+ years of data.
Can I use the results for real trades?
As a decision aid yes, but not a substitute for professional financial advice. Backtests show past performance — no future guarantees.
Why different results for the same asset and strategy?
The start date significantly affects results. Always test multiple time periods.
What does "no signal" mean?
The strategy found no conditions met in the period. Try a longer range or different parameters.
7. Understanding Strategy Insights
The Strategy Insights page shows aggregated results across all backtests — from all users and assets. Here's how the calculation works:
Time-weighted CAGR
The overall CAGR is calculated in two stages: first per asset (longer backtests count more), then across all assets. A 6-year backtest carries three times the weight of a 2-year backtest.
Deduplication
Identical backtests (same asset, period, and parameters) count only once — no matter how many users ran them. This prevents skewing from popular combinations.
Effective time span (union)
Overlapping periods are merged. Two runs covering 2018–2024 and 2020–2023 result in 6 effective years, not 9. No market phase is counted twice.
Parameter variants
In the detail panel, you can see different settings separately (e.g., RSI 14/SMA 14 vs. RSI 7/SMA 21). The overall CAGR at the top is the weighted average of all variants combined.
Buy & Hold comparison
The B&H CAGR shows what would have happened without active trading. A strategy only makes sense if it beats B&H long-term — after considering effort and risk.
Minimum data threshold
Cells with fewer than the set minimum number of runs show "Insufficient data". The more users run backtests, the more meaningful the insights become.
Quality filters
Two filters keep extreme outliers out of the average: pair names must be ASCII (BTCUSDT, AAPL.US, EURUSD.FOREX) — vanity tokens with CJK / special characters are excluded. And every run needs ≥6 months of backtest period — otherwise annualized CAGR is statistically meaningless (a pump coin doing 100x in 22 days mathematically becomes ~5000% CAGR, but tells you nothing real).
How do we count the numbers at the top of Strategy Insights?
- Total backtests: The real total of all saved backtests ever — including re-runs of identical configurations. Reflects the platform's real activity.
- Configurations: Unique combinations of strategy + interval + pair + period + parameters. If you run the same backtest twice, it counts once here — this keeps the CAGR aggregates clean from duplicates.
- Assets: Number of distinct pairs/symbols that have been tested (BTCUSDT, AAPL.US, EURUSD.FOREX, ...).
Your runs land in "My Backtests" per user — duplicates are detected automatically there: an identical run shows "Already saved", a longer period replaces the old entry.
→ Tip
Run different backtests with various assets, periods, and parameters — every run improves the data for all users.
8. Using Signal Status
The Signal Status (traffic light) shows you at a glance which of your strategies currently give a buy or sell signal — calculated automatically, no manual backtest needed.
Open Signal Status page
Navigate to /dashboard/ampel — there you see all your configured signals.
Click "+ Add Signal"
Create a new signal configuration.
Choose configuration
Choose asset type (Crypto/Stocks/...), pair, interval, and strategy. Alternatively: "All strategies at once" — every available strategy will be calculated for this combination.
Signal is calculated automatically
The cron job calculates the status automatically: daily at 00:05 UTC for daily candles, Mondays for weekly candles.
Read the result
Green = strategy says BUY (open position). Red = strategy says SELL (no position). Grey = not yet calculated.
Import from Alerts
If you already have alerts set up: click "Import from Alerts" and all your alert configurations will be added as signals.
Plan limits:
Free— 1 signalPro— 15 signalsElite— 25 signals
→ Tip
Use "All strategies at once" for an asset you're particularly interested in — you'll instantly see which strategies are currently bullish or bearish.
9. Understanding Benchmarks: B&H and DCA
Every backtest shows you two reference values alongside your strategy results: Buy & Hold (B&H) and — once available — DCA. Both are useful benchmarks, but each comes with important caveats worth understanding.
Buy & Hold — The Optimistic Benchmark
Buy & Hold assumes you invest your full capital in one go, on the exact first day of the selected backtest period. In reality, almost nobody times their entry that perfectly.
What this means in practice:
- If you set your backtest start date to January 2019, B&H assumes you bought Bitcoin at ~$3,500. Your strategy then has to beat that entry — which is genuinely hard.
- If you set your start date to November 2021 (near the ATH), B&H assumes you bought at ~$65,000. Suddenly almost any strategy looks great by comparison.
- B&H is not a neutral benchmark — it reflects your chosen start date. A strategy that "beats B&H" may simply benefit from a bad B&H entry point. Always ask: would I have actually bought at that exact moment?
DCA — The More Realistic Benchmark
Dollar-Cost Averaging spreads your capital across regular purchases over time, regardless of price. This removes the question of timing and reflects how most real investors actually enter the market.
What makes DCA different from B&H:
- B&H is all-in at one moment. DCA is gradual.
- DCA typically produces a lower average entry price in trending markets, but also means your capital works less hard early on.
- A strategy is compared against DCA on equal terms: both deploy capital gradually. B&H has the advantage of full capital from day one.
→ When your strategy beats DCA, that's a meaningful result — you're outperforming a realistic alternative that most investors could actually execute.
The honest summary
| B&H Fixed | Avg B&H | DCA | Strategy | |
|---|---|---|---|---|
| Entry | All-in, day 1 | Avg all days | Gradual | Signal-based |
| Timing risk | High | None | Low | Varies |
| Best used as | Ceiling | Fair benchmark | Realistic baseline | The actual test |
→ Use B&H as the ceiling: "Could I even beat someone who bought at the perfect moment?" Use DCA as the floor: "Does my strategy add value over simply averaging in?"
10. Altcoin Season Indicator
The Altcoin Season Indicator shows at a glance whether altcoins or Bitcoin currently dominate the market.
Layer 1 — Trend Analysis
4 market dominance relations evaluated using proprietary trend analysis. Determines the overall signal: Green, Yellow, or Red.
Layer 2 — CoinGecko Top-50
Additional info: How many top-50 altcoins outperform BTC over 90 days? Does not influence the overall signal.
Layer 3 — Strategy Activity
Coming Soon — measures whether altcoin strategies are currently more profitable than BTC strategies.
The chart shows BTC price and market dominance (BTC.D, Alt.D, Stable.D) with colored background for the current signal. Daily update at 01:15 UTC.
→ Available under Analysis → Altcoin Season
11. Using Filters
Trading filters restrict when a strategy is allowed to open a trade. They only affect entry — exit is always decided by the strategy.
Min. Profit Guard (Pro+)
Blocks a new entry as long as the last closed trade is below the configured profit threshold. Useful to pause a strategy after a losing or break-even streak until market conditions stabilise.
200 WMA Filter (Pro+)
Enable the filter in "Advanced Filters" below the strategy parameters. When active: the strategy only buys when price is above the 200-period WMA. If price is below when a signal fires, the signal is parked — and executed once price crosses back above the WMA.
Altcoin Season Filter (Pro+, altcoins only)
Only available when an altcoin pair (not BTCUSDT) is selected. Choose between Aggressive (from Yellow) and Conservative (Green only).
ATR Volatility Filter (Pro+)
Available for all assets. Entry only during matching volatility regime: Low (quiet), High (active) or Expansion (volatility breakout). Defaults pre-tuned per asset class, adjust lookback and multiplier as needed.
ATR Trailing Stop / Chandelier Exit (Pro+, exit modifier)
The only filter that affects exits. Two modes: ATR Trailing Stop (monotonic, only ratchets up — classic) or Chandelier Exit (rolling N-bar high, can fall again — more flexible). Once close drops below the stop, the trade is closed regardless of strategy signal. Not applicable to DCA / Buy & Hold.
→ Tip: Test your strategy with and without filters and compare results. The result header shows "X of Y trades via Trailing Stop" when a stop has triggered.
12. Reading the Arena Score
The Arena Score appears after every backtest in the result panel. It rates the result on a scale of 0–100.
The four dimensions
Return (30 pts) · Efficiency (25 pts) · Consistency (25 pts) · Sample Size (20 pts)
High CAGR alone is not enough for a good score. A backtest with 5 trades can never achieve a high score.
→ Tip: Look for combinations that score well across multiple dimensions — not just CAGR.
13. Using the Sentiment Dashboard
The Sentiment Dashboard shows where the community is currently focusing its attention.
Use it to:
- See which assets are being tested most actively
- Discover trending assets before they go mainstream
- Monitor the Community Profit Trend
→ Important: Sentiment data is not a buy signal. It shows community activity — not price prediction.
14. Strategy Library
One detail page per strategy with methodology, pseudo-code, strengths/weaknesses, parameter descriptions and FAQ. Plus historical live-stats from real backtests.
What's included?
13 strategies documented: 5 free (RSI/SMA, Golden Cross, RSI OB/OS, B&H, DCA) + 8 Pro+ (Stoch-RSI/SMA, OBV-MACD, WMA Trend, EMA Trend Bias, Keltner Channel Breakout, F&G Cadence, SMA>Max Pain, BTC Signal RSI/SMA).
Mini-backtest on every page
Every detail page ships with a pre-configured example backtest including equity curve, trade list and aggregate metrics. Direct link into the editor for tuning.
15. Bitcoin Hub
All Bitcoin-specific tools in one place, split into three tabs: Charts, Lifestyle Calculator, Market Pulse.
Charts — On-Chain & Cycle
14 charts: Pi Cycle Top, Mayer Multiple, Power Law, Rainbow, MVRV Z-Score, NUPL, SOPR, Realized Price, Puell Multiple, Hash Ribbons, Bull Market Traffic Light, drawdown from ATH, monthly returns, 200W MA Heatmap, HODL Waves. 800-word methodology DE+EN per chart.
Lifestyle Calculator
How long does your BTC stack last at your standard of living? Power-Law-based projection, inflation and withdrawal rate configurable.
Market Pulse — Live Status
Status banner with current cycle phase, bull market stages (0–5), F&G value and BTC price. Three detail cards: Cycle Phase (Z-score + 8 indicators), Bull Market Traffic Light (5 stages from 20W/50W MA), Sentiment (F&G cadence chart, Pro+).
16. Volatility Insights
Aggregates saved backtests by volatility phase at entry. Answers: which strategy works in which volatility regime?
Four phases
Low / Normal / High / Expansion — classified via rolling median ATR + spike check. The volatility phase at entry is recorded per trade.
Light variant
Aggregate over the latest 300 runs — heuristic, not for optimization. Filterable by asset type and minimum trades.
→ /dashboard/volatility-insights (Pro+)
17. Grid Trading Backtest
A standalone system alongside the strategy engine: a grid bot places buy and sell orders at price levels within a range and profits from sideways volatility.
What the grid does
Define a range (e.g. $40k–90k for BTC), set number of grids (e.g. 50). The bot buys when a level crosses down, sells when the next level above crosses up — again and again.
Range exit pauses, does not terminate
When price leaves the range, the bot pauses naturally (no levels outside). Once price returns, the system continues. Stop-loss and take-profit are hard exits.
Plan limits
Free + Pro+ have access. Free: BTCUSDT/ETHUSDT only, start from 2022. Pro+: all crypto pairs, start from 2020. Elite: all pairs, start from 2017.
18. Reports Shop & Custom Reports
Two paths to finished PDF/Excel reports: curated strategy reports in the shop or a self-configured custom report.
Strategy Reports Shop
Ready-made reports on curated universes — Quick Insights (Top-50, freely shareable) and Full Reports with methodology + trade list + hybrid allocator. Releases are listed at /reports.
Custom Report Wizard — 4 knobs
You pick strategy, universe (top-10/50/100/250), timeframe (1d/1w/1M) and period (last-2y / last-5y / since-2020 / maximum). Pre-flight checks listing date of all pairs vs strategy warmup; young pairs are substituted with the next largest — you always get the full universe size.
Pricing — pay-per-use
€0.10 per viable pair, capped €14.99–€49.99 (Top-10 = €14.99, Top-100 ≈ €9.99/pair, floored). Async PDF + Excel by email (~5–10 min). No subscription, no auto-renewal.
Retention code: 20% off
After every successful custom report you automatically get a 20% discount code (max 3 redemptions, 90 days valid). Code by email + on the account reports page + auto-applied in the wizard.
19. BTC Ladder Stacking
Z-Score-aware sat accumulation for BTC: weekly base × multiplier depending on the current cycle Z-Score, plus monthly trim actions at the top. Two modes: calculator (educational, Free + Pro+) and live tracker (operational, Pro+ exclusive).
Default bands (CSH)
0–30 → Buy 4× (deep value) · 30–50 → Buy 2× · 50–70 → Buy 1× (normal DCA) · 70–85 → Trim 5 % monthly · 85+ → Trim 10 % monthly. 4 % APY on un-deployed cash.
Calculator (educational)
Backward: simulates on 3 cycle presets (2018, 2022, current 2025) — Free. Forward: Z-Score analog matching with custom date + tolerance — Pro+. Comparison vs HODL Pure Lump-Sum + Static DCA baselines, fair math.
Live tracker (operational)
Mon 08:00 UTC the cron runs, writes a pending action for each active portfolio, and sends email + optional Telegram alert. You click confirm/skip/edit (own fill price). Plus: ⊕ manual action for ad-hoc buys + ⚙ custom bands editor (1–10 own Z thresholds).
Setup (2 minutes)
1) Have Pro+ → open the "Live Tracking" tab. 2) Create portfolio: initial capital, weekly base, trim cadence (1st Mon/month or rolling 4 wks), currency (EUR/USD). 3) Optional: initial BTC holdings if you already hodl. 4) Email + Telegram alerts on/off. First alert fires next Monday.
Plan limits
Calculator: Free has 3 cycle presets + standard bands. Pro+ has custom date + custom tolerance. Live tracker: Pro = 3 active portfolios, Elite = unlimited. Free sees the tab with an upgrade CTA.
20. Portfolio Simulator
5-asset allocation tool for historical what-if analyses: Bitcoin · S&P 500 · Ethereum · Gold · Cash. Distribute percentages via sliders, pick a period, see what €100 would have grown to — against 100% Bitcoin as reference. Plus Sharpe, max drawdown and volatility per mix. Pro+.
Setup (1 minute)
1) Open `/dashboard/portfolio/simulator` (Pro+ login). 2) Adjust the default mix (BTC 25 · SPX 45 · ETH 10 · GOLD 15 · CASH 5). 3) Pick start and end date — min-date adapts automatically to active assets (ETH from 2015, BTC from 2010, S&P from 1993). 4) Choose rebalancing mode (default yearly). 5) Chart and metrics auto-refresh after 300 ms.
Slider logic
When you change one slider, the others keep their ratio between each other but get proportionally scaled. Example: mix 25/45/10/15/5, Gold from 15 → 30 → BTC 20.6 · SPX 37.1 · ETH 8.2 · CASH 4.1 · Gold 30 (sum 100). Sliders at 0 stay at 0 (asset deliberately excluded). Float drift is pushed onto the largest other slider.
Rebalancing
Never: mix drifts, BTC can grow from 20% to 50%. Yearly: reset on first trading day of each new year (Bogleheads standard). Quarterly: reset on first trading day of each new quarter.
Cash asset
Cash earns the daily 3-month US T-Bill rate (FRED DTB3). At 4% annual, cash grows ~0.016% per day. Max drawdown = 0, Sharpe ≈ 0 — cash carries no market risk, but also no premium. Useful as a comparison anchor and as a risk-off component in volatile phases.
Reading the metrics
The better value per row wins (orange highlight). For end value + Sharpe higher wins; for drawdown + vol smaller wins. Key insight: a mix can lose on end value and still have the better Sharpe — that's the point. 100% BTC beats almost every mix on end value over 10 years, but Sharpe + drawdown look much more comfortable for a 60/30/10 mix.
What it isn't
Not a future prediction. Not a trade simulator with entry/exit timing. No DCA mode (lump-sum on start date). No fees or taxes modeled — both reduce the actual end value.
21. Account & Setup
Everything tied directly to your account — security, plan, notifications, demo entry.
2FA / TOTP
Optional two-factor authentication via any standard authenticator app (Google Authenticator, Authy, 1Password, …). Enable from the account page.
Telegram Bot
@SifuBacktest_bot — link your account with /connect and run backtests, fetch signals and share results directly from Telegram. Plan-aware rate limits.
Newsletter (double opt-in)
Optional newsletter with new reports, strategies and platform updates. Sign up via checkbox at registration or footer form, confirmation by email.
Hybrid demo on the landing page
Right on the landing page: interactive slider for start month (Jan 2018–Dec 2024), 5 historical quick scenarios and a mix slider (0–100% strategy/avg-B&H hybrid). Save or alert clicks store the intent in localStorage and execute the backtest in your account automatically after sign-up.
Plan & billing
Free / Pro (€9.99/mo) / Elite (€19.99/mo) plus 12-month packages (Early Bird €9.99/year / Early Adopter €19.99/year / Supporter €29.99/year). Stripe subscription, cancellable anytime. EU right of withdrawal via account button (§ 356a BGB) — withdrawal-waiver checkbox at purchase makes the selected plan date final.
Backtesting Arena is an analysis tool. All content is for informational purposes only and does not constitute financial advice.
← Start a backtest