Start here
New to testing a strategy? Five minutes here and every report on the site will read plainly.
What this site does, in four lines
You write down a trading strategy as rules: when you buy or sell, and when you get out. We run those exact rules on one-minute bars, since 2008 on most markets, with real costs. We compare the result with 5,000 copies of your strategy that pick the side by coin flip. Then we tell you, in plain words, whether it held up: Passed, Close or Failed. It is free, and it is not advice.
Three ways in
You can say it in a sentence. Open Describe your strategy and write it the way you would tell a friend: “Buy NQ when the first 15-minute high breaks, stop at the low, out at 3:55.” We read it back to you in plain words before anything runs, and ask only if something is missing.
You want to see what a result looks like first. Read a sample report. It is a strategy that failed, which is what most do.
You want to look at strategies people already trade. Open every setup, tested: 50 well-known setups on six markets, 300 tests, every result published, failures included.
The words you will meet
Passed, Close, Failed. The verdict. The score out of 100 is made of six checks, each with its points shown. Passed is 70 or more with no check under half its points, no automatic fail, and two more checks: fewer than 1 in 20 coin flips did as well on the years before 2023, counting every version you tried, and the range of likely results on those years stays above zero. Close is 40 to 69, or 70 and over with one of those checks missed. Failed is under 40, or failed automatically (too few trades, or money only since 2023).
The coin flips. 5,000 copies of your strategy that trade at the same moments but pick buy or sell at random. If many of them did as well as yours, your rules may not be what made the money.
The years since 2023, kept apart. The coin flips are run on 2008 to 2022, and the years since 2023 are shown on their own. A strategy that made all its money in one stretch shows it here.
One contract, after costs. Every dollar figure is for one contract, after commission and a tick of slippage each way. Nothing is sized up to look better.
Longest wait for a new high. How many months the strategy went without beating its own best result. This is the part most people underestimate: could you have kept trading it that long?
A try. Every version you test counts. Change one thing and test again, and the second result is held to a stricter bar, because it was picked after seeing the first.
Any other word: the glossary has every word on the site in plain English.
What it is not
A backtest is a test on past prices, not real trading, and a good one does not promise a good year ahead. We don’t sell signals, courses or strategies, and nobody here earns when you trade. Read what a backtest result is and isn’t before you put money on anything.