Building Your First Trading Bot in Python

Every quant's journey starts somewhere. This tutorial walks you through building a simple momentum trading bot from scratch — the same foundation used in professional systematic trading.

Architecture Overview

A trading bot has four components:

Data Pipeline — Fetch and clean market data

Signal Generator — Compute trading signals from data

Portfolio Constructor — Convert signals to position sizes

Execution Engine — Place and manage orders

Step 1: Data Pipeline

import pandas as pd

import yfinance as yf


def fetch_data(tickers: list[str], period: str = "2y") -> pd.DataFrame:
"""Fetch daily OHLCV data for a universe of stocks."""
data = yf.download(tickers, period=period, group_by="ticker")
# Pivot to get adjusted close prices
closes = pd.DataFrame({
ticker: data[ticker]["Adj Close"]
for ticker in tickers
})
return closes

Step 2: Momentum Signal

The classic 12-1 momentum signal: rank stocks by their 12-month return, excluding the most recent month (to avoid short-term reversal).

def momentum_signal(prices: pd.DataFrame, lookback=252, skip=21) -> pd.DataFrame:

"""12-month momentum with 1-month skip."""
returns = prices.pct_change(lookback).shift(skip)
# Cross-sectional rank (0 to 1)
ranked = returns.rank(axis=1, pct=True)
return ranked

Step 3: Portfolio Construction

def construct_portfolio(signals: pd.DataFrame, long_pct=0.2) -> pd.DataFrame:

"""Long top quintile, equal-weight."""
threshold = 1 - long_pct
weights = (signals >= threshold).astype(float)
# Normalize to sum to 1
weights = weights.div(weights.sum(axis=1), axis=0).fillna(0)
return weights

Step 4: Backtest

def backtest(prices, weights):

"""Simple vectorized backtest."""
daily_returns = prices.pct_change()
portfolio_returns = (weights.shift(1) * daily_returns).sum(axis=1)
cumulative = (1 + portfolio_returns).cumprod()
sharpe = portfolio_returns.mean() / portfolio_returns.std()  (252 * 0.5)
return cumulative, sharpe

What's Next

This bot trades once a month with zero leverage. To make it competition-ready:

• Add risk management (position limits, stop-losses, volatility targeting)
• Combine multiple signals (momentum + value + quality)
• Optimize execution timing (avoid trading at the close)

Upload your strategy to AlphaNova and see how it ranks against thousands of other quants.