Backtest Basics

What is a Backtest?

A backtest simulates how a trading strategy, called a "symphony," might have performed in the past using historical data. It’s important to note that these simulations make certain assumptions and differ from live trading symphonies.

Key Points to Understand

Hypothetical Nature: All backtest results are hypothetical and benefit from hindsight.
Past Performance: Past performance is not a reliable indicator of future performance.
Estimation: Backtests estimate how a model performs under specific market conditions and are not recommendations for actual trading decisions in the past or the future.

Data Sources

Backtests: For stocks, we use daily adjusted closing prices, accounting for corporate actions like dividends or splits, to simulate historical trading decisions. These prices don’t represent real quotes from any point during the trading day. For crypto, we use daily historical prices from 4:00 PM Eastern to simulate historical trading decisions.
Live Trading: When a symphony is traded live, trading decisions are made during the trading period using real-time data, which represent real quotes on the market.

Trading Period

Time of Day: Live trading happens during the trading period. Decisions made during this window may differ from those suggested by a backtest due to volatility.
Intra-day Variations: High volatility can lead to different decisions even for different instances of the same symphony due to varying real-time quotes.

Symphonies Trading Stocks — Dividends and Distributions

Backtests and Live Trading: Both scenarios assume reinvestment of dividends from stocks held by your symphony and distributions into the symphony.

Costs Considered

Trading Pass (Only Applies to Modeled Symphonies Trading Stocks): Backtests include estimates for Composer’s Trading Pass ($30 monthly or $288 annually at a 20% discount). Remember, the Trading Pass is charged per user, not per symphony. If you invest in multiple symphonies, the effective fee per symphony is significantly lower than what is modeled.
Regulatory Fees (Only Applies to Modeled Stock Trades): These mandatory fees are collected on sales of securities and are forwarded directly to regulatory bodies–the US Securities & Exchange Commission (SEC) and the Financial Industry Regulatory Authority (FINRA), not retained by Composer:
- SEC Fee: Charged at $27.80 per $1,000,000 of principal sold, rounded up to the nearest penny per transaction.
- FINRA Trading Activity Fee (TAF): $00.000166 per share sold, rounded up to the nearest penny per transaction, with a maximum of $8.30 per transaction.
Spread Markup (Only Applies to Modeled Crypto Trades): Composer charges a transparent fee for crypto trades: a spread markup of 20 bps (basis points) on either side of a trade. This spread markup is shared between Composer Crypto LLC and our trade execution partner, Alpaca Crypto LLC.
Slippage (Applies to All Modeled Trades): Slippage reflects the potential cost difference between hypothetical trades in your backtest and actual trades that might encounter price movement and spread. The amount of slippage is influenced by market conditions, liquidity, and bid-ask spreads. The default slippage setting in backtests is 1 basis point.

Benchmarks and Indices

Purpose: Used only for illustrative purposes and provided for the purpose of making general market data available as a point of reference only.
Limitations: Benchmarks and financial indices are unmanaged, do not reflect the impact of any trading commissions and costs, management and incentive fees.
Benchmark choice: The S&P 500 is not the only index used as a benchmark for measuring the performance of a portfolio. Depending on the holdings in your portfolio, your investment objectives, and your risk tolerance, it may be more appropriate to measure performance against a different benchmark.

Backtesting Compared to Other Platforms

Assumptions Vary: It’s crucial to understand that different platforms may use different assumptions in their backtests. We recommend users familiarize themselves with these details before making comparisons.

Always consider economic factors, market conditions, and investment strategies when evaluating the potential performance of a portfolio. There are no guarantees that a strategy will match or outperform any particular benchmark.