Backtest your strategy on
18 years of real market data

Bring your own rules — or your own Python. Run them against SPX, NDX and 40+ single names, minute by minute, from 2008 to today. You see the estimated cost before every run, and right now every run is free.

Sign in with Google or email. No card required.

Bring your own strategy

Build entry and exit rules in the browser, or upload a Python file against a small, documented interface. Your logic, not a fixed menu of presets.

Data worth testing on

Minute bars back to 2008 for SPX, NDX and 40+ liquid single names, split- and dividend-adjusted — plus SPX options chains where the history supports it.

Costed before it runs

Every job is priced before you commit: date range, universe and resolution give you an estimate in dollars and minutes. Nothing starts until you say go.

How it works

Four steps, no infrastructure to think about.

01

Define

Write rules in the builder or upload a Python strategy file.

02

Estimate

Pick your window and universe. We price the run before it starts.

03

Run

The job lands on a dedicated spot instance with a hard time limit.

04

Read

Equity curve, drawdown, trade log and per-year breakdown — exportable.

A worked example

The classic RSI(2) mean reversion on SPX — the same strategy shown above. This is the whole file.

# rsi2_meanrev.py — runs as-is
from dfd import Strategy, indicators

class RSI2(Strategy):
    symbol   = "SPX"
    interval = "1d"

    def setup(self, bars):
        self.rsi = indicators.rsi(bars.close, 2)
        self.ma  = indicators.sma(bars.close, 200)

    def on_bar(self, i):
        # only trade with the long-term trend
        if self.close[i] < self.ma[i]:
            return
        if self.rsi[i] < 10 and not self.position:
            self.buy()
        elif self.rsi[i] > 70 and self.position:
            self.close_position()
What you get back
PERIOD2008-01-02 → 2026-08-13
NET RETURN+184.2%
CAGR5.8%
PROFIT FACTOR1.62
SHARPE0.94
MAX DRAWDOWN-18.4%
TRADES2,914
WIN RATE61.3%
EST. COST$0.01

Illustrative figures for a documented public strategy, shown to demonstrate the output format. Your results will differ. Backtested performance is hypothetical and is not a guide to future returns.

Stop guessing whether your edge is real.

Create a free account and run your first backtest in the next few minutes.

Create free account