Skip to main content
This guide walks you through a complete backtest from scratch. By the end you will have defined a strategy, run a simulation, reviewed performance statistics, and opened an interactive chart.
1

Install the package

Install backtesting.py from PyPI:
pip install backtesting
See the installation guide for full details including dependencies and environment notes.
2

Import required modules

Open a Python script or Jupyter notebook and import the three things you need:
from backtesting import Backtest, Strategy
from backtesting.lib import crossover
from backtesting.test import SMA, GOOG
  • Backtest — the simulation engine.
  • Strategy — the abstract base class your strategy must extend.
  • crossover — a helper from backtesting.lib that returns True on the bar where one series crosses above another.
  • SMA — a simple moving average function included for testing.
  • GOOG — a built-in DataFrame of daily Google (Alphabet) OHLCV data from 2004 to 2013, used throughout the documentation examples.
3

Define a strategy

Subclass Strategy and implement two methods:
  • init() — called once before the simulation. Declare indicators here using self.I().
  • next() — called once per bar (candle). Place or close orders here.
class SmaCross(Strategy):
    def init(self):
        price = self.data.Close
        self.ma1 = self.I(SMA, price, 10)
        self.ma2 = self.I(SMA, price, 20)

    def next(self):
        if crossover(self.ma1, self.ma2):
            self.buy()
        elif crossover(self.ma2, self.ma1):
            self.sell()
self.I(func, *args) wraps an indicator function so the engine reveals its values gradually as the simulation progresses bar-by-bar, matching how a live system would see the data.The strategy buys when the 10-day SMA crosses above the 20-day SMA and sells (goes short or closes) when the 20-day crosses back above the 10-day.
Strategy parameters can be declared as class-level variables and later swept with bt.optimize(). For example, you could write n1 = 10 and n2 = 20 as class attributes and reference them inside init().
4

Create and run a Backtest

Pass your data and strategy class to Backtest, then call run():
bt = Backtest(GOOG, SmaCross, commission=.002, exclusive_orders=True)
stats = bt.run()
print(stats)
  • commission=.002 — 0.2% round-trip commission per trade.
  • exclusive_orders=True — each new buy() or sell() automatically closes any open position in the opposite direction before opening the new one, keeping position management simple.
bt.run() returns a pandas.Series containing every performance metric.
5

Read the statistics

Running the example above prints output similar to:
Start                     2004-08-19 00:00:00
End                       2013-03-01 00:00:00
Duration                   3116 days 00:00:00
Exposure Time [%]                       94.27
Equity Final [$]                     68935.12
Equity Peak [$]                      68991.22
Return [%]                             589.35
Buy & Hold Return [%]                  703.46
Return (Ann.) [%]                       25.42
Volatility (Ann.) [%]                   38.43
CAGR [%]                                16.80
Sharpe Ratio                             0.66
Sortino Ratio                            1.30
Calmar Ratio                             0.77
Alpha [%]                              450.62
Beta                                     0.02
Max. Drawdown [%]                      -33.08
Avg. Drawdown [%]                       -5.58
Max. Drawdown Duration      688 days 00:00:00
Avg. Drawdown Duration       41 days 00:00:00
# Trades                                   93
Win Rate [%]                            53.76
Best Trade [%]                          57.12
Worst Trade [%]                        -16.63
Avg. Trade [%]                           1.96
Max. Trade Duration         121 days 00:00:00
Avg. Trade Duration          32 days 00:00:00
Profit Factor                            2.13
Expectancy [%]                           6.91
SQN                                      1.78
Kelly Criterion                        0.6134
_strategy              SmaCross(n1=10, n2=20)
_equity_curve                          Equ...
_trades                       Size  EntryB...
dtype: object
The strategy returned 589% over the period versus a buy-and-hold return of 703%, with a CAGR of 16.8%, a Sharpe ratio of 0.66, and a maximum drawdown of -33%. It executed 93 trades with a 53.8% win rate.stats._trades is a full DataFrame of every individual trade with entry/exit bars, sizes, P&L, and commissions. stats._equity_curve is the equity curve as a time series.
6

Plot the results

Open an interactive Bokeh chart:
bt.plot()
The chart shows the OHLC candlestick data, your declared indicators overlaid or in sub-panels, trade entry/exit markers, and the equity curve with drawdown periods highlighted.
bt.plot() opens a browser window when run from a standalone script. In a Jupyter notebook it renders inline. If you are running in a non-interactive environment, pass filename="output.html" to save the chart to a file instead.

​
Complete example

Here is the entire working example in one block: 
from backtesting import Backtest, Strategy
from backtesting.lib import crossover
from backtesting.test import SMA, GOOG


class SmaCross(Strategy):
    def init(self):
        price = self.data.Close
        self.ma1 = self.I(SMA, price, 10)
        self.ma2 = self.I(SMA, price, 20)

    def next(self):
        if crossover(self.ma1, self.ma2):
            self.buy()
        elif crossover(self.ma2, self.ma1):
            self.sell()


bt = Backtest(GOOG, SmaCross, commission=.002, exclusive_orders=True)
stats = bt.run()
bt.plot()

​
Next steps

Now that you have a working backtest, explore the rest of the documentation to go further:
  • Bring your own OHLCV data as a pandas DataFrame with a DatetimeIndex.
  • Add parameters to your strategy and use bt.optimize() to find the best values.
  • Use composable base strategies from backtesting.lib such as TrailingStrategy or SignalStrategy.
  • Plug in any indicator library — TA-Lib, pandas-ta, or your own NumPy functions — via self.I().

Build docs developers (and LLMs) love

Get started for freeTalk to us