Running Your First Backtest

A backtest is a historical simulation. You give the engine a strategy, point it at historical market data, and it tells you what would have happened — every entry, every exit, every drawdown.

Before you click anything, you need to decide three things: which market you want to test, how far back you want to look, and which strategy you want to run. Everything else is configuration detail.

For your first backtest, keep it simple. Pick a liquid market like BTCUSDT, use a common timeframe like the daily or 4-hour chart, and choose a built-in strategy. The goal is not to find a winning strategy on the first try — it's to understand what the platform gives you back and how to read it.

Quanthop is built around the idea that you own your strategy code. The platform provides a set of templates — ready-made strategies that you can clone, modify, and build from — but the real power is writing your own logic in QSL (Quanthop Strategy Language). Templates are starting points, not black boxes.

For a first run, clone the EMA Cross template. It's a trend-following strategy that goes long when a fast exponential moving average crosses above a slow one, and exits when it crosses back below. The logic is intuitive, the code is readable, and the results are easy to reason about.

You'll also find templates like RSI Mean Reversion (buys oversold conditions, sells overbought), MACD Momentum, and TDR Advanced (a multi-factor system). Each has its own parameter set that you can adjust. For now, clone a template and use the defaults — you can edit the code later once you understand the workflow.

The configuration panel is where you define the scope of the test. Every setting feeds into the simulation engine, and understanding what each one controls will help you design more meaningful experiments.

Once your configuration is set, click Run Backtest. The engine processes every candle in your date range, evaluating the strategy's rules at each bar.

Behind the scenes, the engine computes all indicators defined in the strategy (moving averages, RSI, MACD — whatever the strategy requires), then checks entry and exit conditions on every candle. When conditions are met, it simulates a trade: opening a position at the next available price, tracking it through the holding period, and closing it when the exit signal fires or a stop is hit.

The result is a complete trade log — every position the strategy would have taken, with exact entry and exit prices, timestamps, and profit or loss. This log is the raw material for everything you'll look at next.

The results dashboard is where the backtest comes to life. Rather than a single summary, the platform breaks the analysis into eight distinct cards — each one answering a different question about the strategy's behaviour.

The Metrics card is the quantitative core: payoff ratio, profit factor, trade count, win rate, Sharpe ratio, and expectancy. These numbers tell you whether the strategy has a statistical edge, how efficiently it converts risk into return, and how many trades produced the result.

The Behaviour card goes beyond raw numbers and characterises the strategy's profile — whether returns come from a few large winners, steady small gains, or something in between. It also rates consistency: whether performance was stable across the tested period or concentrated in a single regime.

Trade Structure shows where the profit actually comes from. A "Concentrated Winner Profile" means a small fraction of trades drove most of the gains — common in trend-following strategies. Exposure maps when the strategy was in the market and when it was flat, revealing how much opportunity it captures and how long positions are held.

Regimes breaks the test period into volatility and trend categories. If 90% of your profit came during trending markets and the test period happened to be 80% trending, that's important context. Drawdown shows the worst peak-to-trough decline, how long recovery took, and the recovery factor.

The Drawdown Envelope runs a Monte Carlo simulation — shuffling trade order a thousand times to show whether your observed drawdown was lucky, unlucky, or typical. And the Parameters card displays the current configuration with a validation stage badge: Exploratory means the results haven't been confirmed through Walk-Forward Analysis yet.

On the right, the Score panel distils everything into a single grade out of 100, broken into four sub-scores: profitability, risk management, consistency, and robustness. This is a starting point for comparison, not a final verdict — but it gives you an immediate sense of where a strategy stands.

Each of these cards deserves deeper explanation, and the full dashboard is covered in detail in the Understanding Backtest Results article. For now, focus on reading the dashboard as a whole: a strong result is one where multiple cards tell a consistent story, not one where a single headline number looks good in isolation.

A single backtest on a single asset over a single time period is a starting point, not a conclusion. Good numbers here do not mean the strategy works. They mean it worked in this specific context.

Before you draw any conclusions, you need to ask harder questions. Does it work on other assets? Does it survive different market regimes — trending, sideways, volatile? Do the results hold up if you change the parameters slightly? If changing the moving average period from 20 to 22 cuts the return in half, the original result was fragile.

The next steps are where the real validation begins: testing across multiple symbols, running parameter optimization with Walk-Forward Analysis, and stress-testing with Monte Carlo simulation. Each of these is covered in its own article in this series.