Explainer
How to backtest a trading strategy for free
You can backtest for free by hand on a chart, with a free platform tier, with your own code, or by replaying history bar by bar, each with its own limits.
Updated
Four free methods
Free methods exist, and each one trades money for something else: time, setup effort, or limits on data and features. Which one fits depends on how exact your rules are and how many trades you need.
- Manual backtest
- Going through past charts by hand and recording each trade your rules would have taken.
- Bar replay
- A chart that shows past data one bar at a time, so you cannot see what comes next.
- Paper trading
- Placing orders against a simulated account instead of a real one.
- Sample size
- The number of trades behind a figure.
Method 1: By hand on a chart
Scroll back, find each setup your rules describe, and write down the entry, stop, target and result. It costs nothing and it forces you to read every trade.
The limits are time and sample size. A few weeks of charts give you few trades, and a few trades say little. It is also easy to let the next candle influence the call.
Method 2: A replay, so you cannot see ahead
A replay hides the future. TradingView describes its Bar Replay as a way to “simulate past price movements for strategy testing,” and notes that intraday history can be limited by plan. Check the vendor for what your plan includes.
Our Trade Lab is a free bar-by-bar replay for NQ, ES and GC. You pick a start date and step forward through 1-minute bars. No later bar is loaded, you can draw, journal each decision and place paper orders, and it needs an account with a verified email. It replays history only, and no real order is ever placed.
Free is fine if the rules, the data and the costs are right. Free does not make a wrong test right.
Method 3: A free platform tier
Some charting platforms include a results report on a free tier. TradingView lists a Basic plan at $0 forever. Which backtesting features sit on which plan is the vendor’s to say, so check its plan table before you plan around one.
Whatever the platform, read its fill rules. They are part of the result.
Method 4: Your own code with free data
If you can code, you can write the simulation yourself in Python. The code is free. The difficulty is the data. Check the bar size, how far back it goes, the time zone, and how contract months are joined. A futures series that ignores rolls can make a roll look like a trade.
What free methods tend to leave out
Check each one for the same four things: exact fills, costs, contract rolls and held-back months. Free tools often leave these to you. A backtest without costs shows a result no account could have had, and a backtest that never held any months back has only been checked against the data it was built on.
On our service a version is $19, and it comes with the assumptions written out and about 30% of months, spread across every year, held back. See the page on what free covers for the details.
Questions and answers
Can I backtest a trading strategy for free?
Yes. You can go through charts by hand, use a replay, try a free platform tier or write your own code. Each has limits on data, fills or the number of trades.
Is a manual backtest reliable?
It can be, if you record every trade your rules describe and cannot see what comes next. The sample is usually small, which limits what the result can say.
What is a bar replay?
A chart that shows past data one bar at a time with later bars hidden. Ours is free for NQ, ES and GC, with an account and a verified email.
What data can I use for a free futures backtest?
Any data you trust. Check its bar size, its history, its time zone and how it joins contract months before you rely on it.
Do free tools include costs?
That depends on the tool. Check whether commission and slippage can be set, and set them before you read a result.