Getting started

QuantWright is in private beta. Apply with a short form about how you trade. We read every application and admit traders in small groups. Nothing is charged when you apply.

When your place is ready, we email you a link to create your account. The link is tied to the address you applied with and works for 30 days. If that address already has an account, the email asks you to sign in instead.

The first time you sign in, you see a short summary of the Terms of Service and choose I agree before the workspace opens. The summary says what matters most: QuantWright is research software, its results are hypothetical, and it never places orders or holds your money.

There is one plan: €39 a month, excl. VAT. EU consumers pay 25.5% Finnish VAT on top: €48.95 a month in total. The exact total is shown at checkout before you pay. Checkout runs on Stripe, and QuantWright never sees your card. Tick the box to start the subscription now, then subscribe. Beta members keep this price for as long as they stay subscribed.

The first payment of each new subscription has a 14-day money-back guarantee. Cancelling, refunds and the right to withdraw are covered in Limits and billing.

Type the strategy into the chat as you would explain it to another trader. Name the market and the bar size, when to enter, when to exit, and how much to trade. For example:

A good first promptBacktest QQQ on daily bars. Buy when the 2-period RSI closes below 10. Sell when the close is above the 5-day simple moving average, or after 10 trading days. Put 20% of equity into each trade. Commission 0.01% per round trip, slippage 0.02% per side.

What happens next:

1. The AI writes the strategy as Python. You can read it in the Code tab.
2. If your description left a number open, a card lists the values the AI chose and why. Nothing is backtested until you run it, and you can change any value first. Why this happens.
3. The backtest starts with the account size you stated, or $100,000, with commission and slippage charged: yours if you stated them, defaults if you did not.
4. The result shows returns, drawdown, trade statistics and the Honesty Score. A first score is capped at 64 (grade C) until a held-out test has run: the card you choose, re-run once over years you kept out of sight from the start (see caps). Reading results explains each number.

• /validate re-splits the run’s own trades into many train and test folds of the history to show how steady they are, and measures how much of the return is just the market. It re-fits nothing, so it is not an out-of-sample test, and it does not lift the cap on the Honesty Score.
• Run Monte Carlo, under the result, re-runs the strategy and then redraws its own trades 10,000 times, with replacement and in blocks of consecutive trades, to show the spread of returns and drawdowns those same trades could give. It takes the trades as they are, so it says nothing about whether the edge is real or luck.
• /deploy ftmo simulates a prop-firm evaluation with this strategy’s own trades. See Prop-firm simulator.
• To change the strategy, say what to change, for example “tighten the stop to 1.5%”. Each change runs as a new version, and later commands act on the newest one.