freqtrade-strategies: a free strategy library for the Freqtrade bot

What freqtrade-strategies actually gives you

Freqtrade is a free and open source crypto trading bot written in Python, controlled through Telegram, with backtesting, plotting and money management tools. The bot itself lives in a separate repository. This repository holds only the strategy files: Python modules that define buy signals, sell signals, indicators, a minimal ROI table and a stoploss. The README lists exactly those five items as what each strategy includes.

The audience is narrow. The README states that users should have coding and Python knowledge, and that the strategies mostly serve as a starting point for your own work rather than as ready-to-use systems. If you want a bot you configure and forget, this is not that. If you want readable Python that shows how a Freqtrade strategy is structured, this is the reference set.

The disclaimer is blunt: the strategies are for educational purposes only, and the authors assume no responsibility for trading results. The README also notes that some strategies only work in specific market conditions while others are more general purpose, and that optimizing for your exchange and pairs usually produces better outcomes. Nothing in the repository promises a profit.

How a strategy file plugs into the bot

There is no build step and no service to run. The data flow is a file copy followed by a command-line flag. All strategies live under user_data/strategies in the repository, and the README's install steps are: pick a strategy, copy the file, paste it into your own user_data/strategies folder, then start the bot with --strategy followed by the class name.

The class name matters more than the file name. Freqtrade loads the strategy by class, so if a file defines class Strategy001, the flag must say Strategy001. The README's example uses freqtrade trade --strategy Strategy001. Getting this wrong is the most common first failure, and the error surfaces at bot startup, not at copy time.

Once loaded, the strategy supplies the signals and the risk parameters the bot's engine consumes. The README groups those under minimal ROI, stoploss, buy signals, sell signals and indicators. A strategy that references an indicator the installed Freqtrade version does not provide will fail to load; the README pins compatibility at version 2022.4 or newer, which is the only version statement in the repository.

Installing a strategy and running a first backtest

You need a working Freqtrade installation first. The README points to freqtrade.io for that and gives no install steps for the bot itself. Once the bot runs, copy the strategy file you want from the repository's user_data/strategies directory into the same folder inside your own Freqtrade project, then start the bot by class name.

freqtrade trade --strategy Strategy001

That command starts the bot with the named strategy class. In dry-run mode it simulates trades; the README advises running in dry-run before engaging money. Before that, backtest the strategy against historical data.

freqtrade backtesting --strategy Strategy001

The backtest command prints a summary table for the strategy. If you have no data yet, download some first. The README's example pulls 100 days.

freqtrade download-data --days 100

One note from the README is worth repeating: static backtest data from a defined period gives comparable results between runs, so fixing a timerange is better than always downloading the latest window. The README says results depend heavily on the pairs, timeframe and timerange used, and that you should run backtests that mirror your own use case.

Why a strategy that backtests well can still lose money

The repository makes no performance claim, and that is the honest position. The README says results above (in the context of the repository's own summary) should serve as a general outline of the number of trades to expect, and that actual performance will differ. Nothing here is a ranking.

The deeper limitation is structural. A strategy file contains fixed numbers: a stoploss percentage, a minimal ROI table, buy and sell conditions tied to indicators. Those numbers were tuned by someone, on some exchange, on some pairs, over some period. Move them to your exchange and your pairs and the tuning no longer applies. The README says as much when it notes that optimization to the exchange and pairs used usually results in better outcomes, and that you should fine tune the strategy to the markets you are trading.

There is also no maintenance promise per strategy. The repository's last push was on 2026-09-08, but that says nothing about whether any individual file has been revisited. A strategy written for market conditions in one year may simply stop producing signals in another. Nothing in the README documents per-strategy review dates, and there are no releases to track, so you cannot tell from the repository alone which files are current.

Freqtrade strategies versus writing your own from the docs

The real alternative is not another strategy repository. It is Freqtrade's own strategy customization documentation, which the README links directly. That path means writing the class yourself from the interface definition rather than starting from someone else's tuned numbers.

The difference in approach is where the parameters come from. A repository strategy hands you a stoploss and a minimal ROI table that someone else chose. Writing your own, or running hyperopt as the README mentions in the context of how these strategies were produced, means the numbers come from your data. The README states that buy and sell signals in these strategies are the result of hyperopt or are based on existing trading strategies, which is a direct admission that the parameters are inherited, not universal.

Starting from a repository file is faster, and reading several of them is a legitimate way to learn the interface. But the moment you treat the inherited numbers as final, you have adopted someone else's assumptions about pairs, timeframe and market regime. The README's own framing supports this: strategies mostly serve as a starting point for your own strategies.

Licence and the cost of keeping up

The repository is GPL-3.0. That matters if you plan to redistribute a modified strategy or ship it inside a product: the licence carries copyleft obligations, and the LICENSE file at the repository root is the authoritative text. This is a description of the licence identifier, not legal advice, and if redistribution is part of your plan you should read the licence or ask someone qualified.

Upgrade cost is low but not zero. The README pins compatibility at Freqtrade 2022.4 or newer, so the repository does not promise that strategies keep working across every future bot release. There are no releases in this repository, so there is no changelog to read before upgrading Freqtrade. The practical check after a bot upgrade is to load each strategy you use and confirm it starts, since an indicator or interface change will fail at load time.

Contribution is open. The README invites strategies, comments, optimizations and pull requests through the issue tracker or as a pull request. That means the set can grow, but it also means quality varies by contributor, and the repository offers no review standard you can point to.

Editorial conclusion

Adopt freqtrade-strategies if you already run Freqtrade 2022.4 or newer and want Python strategy files to read, backtest and modify. Skip it if you expect a ready-to-trade system, if you do not write Python, or if you need a documented performance ranking, because the README provides none. Before risking money, verify the strategy class name matches the file, run freqtrade backtesting on your own pairs and timerange, then keep the bot in dry-run until the results make sense to you.

Frequently asked questions

Official sources

1. freqtrade/freqtrade-strategies on GitHub
2. Issues
3. License: GPL-3.0
4. README