Algorithmic Trading Guides
Implementing Your Algorithmic Trading Strategies
Learn how to test, revise, and deploy a strategy whose securities, rules, sizing, and risk settings you select or adopt.
If you haven't already, review Algorithmic Trading Concepts for definitions used in the configuration workflow.
Once your strategy is defined, use the strategy details page to review its parameters, run hypothetical backtests, test in a simulated virtual portfolio, and, if eligible, authorize live use.
Accessing Your Algorithmic Trading Strategy Dashboard
To begin managing your automated trading strategy, you first need to access its detailed dashboard:
- Navigate to the algorithmic trading strategy list in your Investfly dashboard
- Click on the automated trading strategy name you wish to manage
Optimizing Your Algorithmic Trading Strategy
You can refine any aspect of your automated trading strategy, including entry signals, exit conditions, and risk parameters. Follow these steps to optimize your trading algorithm:
- Access the algorithmic trading strategy dashboard as described above
- Click on the pencil icon in the section of your automated trading strategy you want to modify
- A dialog (the same interface you used to define the strategy) will appear
- Make and document the rule changes you choose
- Click "Save" on the dialog to close it
- Note: The algorithmic trading strategy is not yet saved on the server at this point. This allows you to modify multiple sections before finalizing your changes
- Click the "Save" button at the top of the page to apply all changes to your automated trading strategy
Backtesting Your Algorithmic Trading Strategy
Investfly can run a user-authored strategy against selected historical data and display hypothetical metrics such as profit/loss, win rate, drawdown, and Sharpe ratio. Results depend on the selected period, available data, rules, and modeling assumptions.
Duration of Backtest
The time frame for backtesting depends on the bar size parameters used in your strategy:
- DAILY bars: The backtest will run with the last 5 years of data (approximately 1,260 trading days)
- MINUTE bars: The backtest will run with the last 5 days of minute-bars (approximately 1,950 data points)
Completion time varies with strategy complexity, selected dates, data availability, and system load.
Reliability of Backtest Results
Backtest results are hypothetical and should not be interpreted as actual trading or a prediction of future performance. Limitations include:
- Historical performance does not guarantee future results
- There is a difference in data granularity between live trading (real-time quotes) and backtesting (1-minute bars)
- The non-deterministic nature of strategy execution as described in Automation Concepts. If 100 stocks match your open conditions but you can only buy 10, which stocks are selected is non-deterministic due to parallel processing
Repeating the same backtest on the same strategy could yield different results. This is actually beneficial because if multiple executions produce favorable results, it's a strong indication that your strategy is robust.
Starting and Stopping a Backtest
- Access the strategy detail page
- Click the "Start" button in the backtest section
- The backtest will go through the following states:
- QUEUED: The backtest is waiting to be executed by one of our backtest servers
- INITIALIZING: The environment is being prepared for the backtest
- RUNNING: The backtest is actively running
- You can stop the backtest at any time by clicking the "STOP" button
Backtest Results
The results section displays standard metrics to evaluate your strategy's performance, such as:
- Maximum drawdown
- Percentage of profitable trades
- Overall profit/loss
- Risk-adjusted return metrics
It also provides a detailed table of all trades that were simulated during the backtest run.
Deploying Your Strategy
The final step is to deploy your automated trading strategy to a trading account so it can begin making trades based on your defined criteria.
- Click the "Deploy" button on the strategy detail page
- Select the trading account to which you want to deploy your strategy
- Click "Confirm Deploy"
Once enabled, the strategy evaluates conditions as supported data and system processing permit. Review automation logs and account activity through the Portfolio Dashboard. Evaluation and order execution are not guaranteed.
Advanced Tips for Strategy Management
- Regularly monitor your strategy's performance and make adjustments as market conditions change
- Consider running multiple versions of a strategy with slight variations to identify the most effective approach
- Review the automation logs in the Portfolio Dashboard to understand why specific trades were triggered or not
- If your strategy isn't performing as expected, check for common issues like overly restrictive conditions or timing issues