Separate strategies in different files

Here is an example setup based on your requirements, where you have multiple strategies in separate .py files, each with its own initialization (datasets, indicators, etc.) and trading logic. The main algorithm imports and orchestrates them.

main_algorithm.py

from AlgorithmImports import *
from strategy_one import StrategyOne
from strategy_two import StrategyTwo

class MultiStrategyAlgorithm(QCAlgorithm):
    def initialize(self) -> None:
        # Set desired start/end dates
        self.set_start_date(2021, 1, 1)
        self.set_end_date(2022, 1, 1)

        # Initialize each strategy
        self._strategy_one = StrategyOne()
        self._strategy_two = StrategyTwo()

        self._strategy_one.initialize(self)
        self._strategy_two.initialize(self)

    def on_data(self, data: Slice) -> None:
        self._strategy_one.on_data(self, data)
        self._strategy_two.on_data(self, data)

strategy_one.py

from AlgorithmImports import *

class StrategyOne:
    def __init__(self) -> None:
        self._symbol = None
        self._sma = None

    def initialize(self, algorithm: QCAlgorithm) -> None:
        self._symbol = algorithm.add_equity("SPY", Resolution.DAILY).symbol
        self._sma = algorithm.sma(self._symbol, 5, Resolution.DAILY)

    def on_data(self, algorithm: QCAlgorithm, data: Slice) -> None:
        if not self._sma.is_ready:
            return

        price = algorithm.securities[self._symbol].price
        if price > self._sma.current.value:
            if not algorithm.portfolio[self._symbol].is_long:
                algorithm.set_holdings(self._symbol, 0.2)
        else:
            if not algorithm.portfolio[self._symbol].is_short:
                algorithm.set_holdings(self._symbol, -0.2)

strategy_two.py

from AlgorithmImports import *

class StrategyTwo:
    def __init__(self) -> None:
        self._symbol = None
        self._ema = None

    def initialize(self, algorithm: QCAlgorithm) -> None:
        self._symbol = algorithm.add_equity("AAPL", Resolution.DAILY).symbol
        self._ema = algorithm.ema(self._symbol, 10, Resolution.DAILY)

    def on_data(self, algorithm: QCAlgorithm, data: Slice) -> None:
        if not self._ema.is_ready:
            return

        price = algorithm.securities[self._symbol].price
        if price > self._ema.current.value:
            if not algorithm.portfolio[self._symbol].is_long:
                algorithm.set_holdings(self._symbol, 0.3)
        else:
            if not algorithm.portfolio[self._symbol].is_short:
                algorithm.set_holdings(self._symbol, -0.3)

How to use this template:

• Place all three files in the same QuantConnect project.
• The main algorithm file (main_algorithm.py) is the entry point.
• Each strategy is encapsulated in its own Python file/class, taking care of its setup and actions in dedicated methods.
• The main algorithm coordinates initialization and calls to strategies.

You can scale this pattern to as many strategies or asset classes as you want!

Sources: Multiple Symbol Consolidation Algorithm, Dual Thrust Trading Algorithm, Hybrid Algorithms - QuantConnect.com, Basic Template Option Strategy Algorithm, Multi-Asset Modeling - QuantConnect.com

Thanks so much Mia Alissi . You're a superstar!!

Thank you so much for your kind words! I'm thrilled I could help. If you ever have more questions or need assistance with QuantConnect, please don't hesitate to ask. I'm here to support you, and I truly appreciate the shoutout!

Sorry Mia Alissi - is there a constraint when creating new files? They seem to disappear after I close and reopen the project. Same thing occurs if I clone an existing project and try to create a new file. Looked the the doc below and I don't think I'm doing anything wrong.

It sounds frustrating, but you’re not alone—this issue can happen due to a few common reasons:

• Only supported file types (such as .py, .cs, .ipynb, .md) will persist in QuantConnect projects. Unsupported file types may disappear when you refresh or sync the project. Here’s the list of supported file types.
• If you’re working locally with the Lean CLI, make sure you run lean cloud push and lean cloud pull to sync changes between your machine and the cloud. Unpushed files may not persist.
• Ensure that you save any new files before closing or switching projects.
• Full cloud synchronization is only available for users in paid organization tiers; make sure your workspace is correctly linked.
• If you’re cloning or copying a project, double-check that all files are saved and pushed before closing.

If you’ve double-checked the above and still see issues, let us know your workflow and we’ll help dig deeper. Don’t hesitate to reach out if you need more assistance!

Sources: Cloud Synchronization - QuantConnect.com, Libraries - QuantConnect.com

Thanks Mia Alissi  just realised I need to specify the file extension. Sorry for the extra questions. Now the backtest seems to have gone into a Runtime error. Any tips from here?

Great catch on the file extension issue! For the Runtime Error, here are some actionable steps to help you debug:

1. Carefully read the stack trace—the filenames and line numbers often point directly to where things went wrong.
2. Double check all your imports, file names, and extensions, especially since you’re using multiple files and custom logic.
3. Add self.Debug() or self.Log() statements right before the lines where you suspect errors; printing out variable values can reveal unexpected results.
4. If you’re loading custom data, make sure the file is uploaded, named correctly, and that you have permission to access it from your code.
5. Check QuantConnect documentation and forums for similar errors—there are lots of helpful threads!
6. Make one change at a time and retest so you can quickly isolate and fix the problem.
7. If you're still stuck, feel free to share the exact error message or stack trace—happy to dive deeper!

You're making fantastic progress—debugging is a skill, and you’re sharpening it!

Sources: Debugging - QuantConnect.com, Debugging Tools - QuantConnect.com

Hey Mia Alissi  you're so quick! I shared the stack trace in the previous post as an image. It doesn't appear to be related to the code (I copied exactly your code above), but related to Lean. Here is the stack trace text:  

[ERROR] FATAL UNHANDLED EXCEPTION:PythonInitializer.AddAlgorithmLocationPath(): Unable to find algorithm location path: .,Engine.Main(): Sorry we could not process this algorithm request.,Unhandled exception. System.ArgumentException: Engine.Main(): Sorry we could not process this algorithm request., at QuantConnect.Lean.Launcher.Program.Main(String[] args) in /LeanCloud/CI.Builder/bin/Release/src/QuantConnect/LeanEnterprise/Launcher/Program.cs:line 75, runLauncher.sh: line 18: 7 Aborted (core dumped) dotnet QuantConnect.Lean.Launcher.dll --data-folder /Data --config /QuantConnect/backtesting/airlock/config.json --results-destination-folder /QuantConnect/backtesting/airlock/

This error usually means the algorithm entry point file is missing, misnamed, or not set up correctly. Here are some things to check:

1. In the QuantConnect Cloud, your main entry point should be named exactly main.py (all lowercase, .py extension included).
2. If you’re using a multi-file setup, make sure all your supporting files end in .py, are present, and are properly saved. Also, your main algorithm class needs to be defined directly in main.py (not nested in a function).
3. Double-check that you didn’t accidentally rename, move, or delete main.py, and scan for any typos in file or class names.
4. If new files or changes don’t appear, try restarting the editor or re-uploading your files to the project.
5. If you’ve checked all the above and still see the error, it’s worth reaching out to QuantConnect Support with your project structure and the full error message.

If you need help checking these, or want to share your current file list, I’m happy to walk through it with you!

Sources: Debugging - QuantConnect.com

Thanks for you patience Mia Alissi  Just realised the main entry point should be named exactly main.py! Works now thanks to you !!

Congratulations! Your dedication and perseverance have truly paid off. I'm so glad to hear that renaming the main entry point to 'main.py' solved your issue. Thanks for sticking with it, and for your kind words too! Please don't hesitate to reach out again if you ever need more help or run into any other questions—I'm always happy to lend a hand!