Products Research

Strategies that show their evidence.

A marketplace for qkt trading strategies. Every listing carries evidence the system generates, not claims the seller makes.

Beta · test-mode payments
Replay
Bid/ask ticks, pinned costs
Account sizes
$25k, $50k, $100k
Held out
Last 20% of the window
Sign-in
Google
A strategy goes through the Labs pipeline: upload, tick-by-tick runs with the last 20 percent held out, walk-forward folds, doubled costs, a deflated Sharpe check and an automatic publishing gate.

The pipeline

No result is typed in by hand.

Every strategy goes through the same automated steps before it can be listed. One failed check and it is not published; the failure report goes privately to its author.

  • Tick replayOrders fill at the bid or ask, stops trigger on the tick that crosses them, and costs come from a pinned file taken from a real broker account.
  • A window it never sawThe last 20% of the test window is held out. It must be profitable there, at every account size.
  • Walk-forward and doubled costsRe-tested on rolling windows, then again with commission, slippage and swap at 1.5× and 2×.
  • A luck checkThe deflated Sharpe probability discounts a result by the number of variants the author tried.

The real storefront

One listing, in full view.

A real screenshot of a house-research listing, taken on 12 October 2026. Figures on the public page are from the in-sample window; the out-of-sample, walk-forward and cost-stress results unlock with the purchase. They are simulated results, shown with their drawdown.

The qkt Labs page for Gold Trend Long H1 on XAUUSD: eleven of eleven publishing checks passed, an in-sample return of 16.7 percent with a 24.0 percent maximum drawdown over 150 trades, and locked tabs for the out-of-sample, walk-forward, cost-stress and Monte Carlo evidence.

Why you can check it

You get the files, not a promise.

Buyers receive the exact strategy file, the exact cost profile and every raw report the pipeline produced, each with its SHA-256. Re-running the file on the same data with qkt gives the same trades.

  • Open engineEvery test runs on qkt, so how fills, costs and stops are simulated is public.
  • Price from the evidenceCalculated from out-of-sample results and reduced when the evidence is thin. Never set by hand.
  • Limits statedSimulations miss slippage, outages and broker differences. Stops can fill worse in live markets.
Read the risk disclosure

Support the project

Labs keeps qkt free.

The engine, gateways and dashboards stay open source. Labs sales help pay for them, and donations on Patreon are planned for people who just want to help.

Patreon: coming soon

Write your first strategy today.

One command to install. Eight lines to a backtest. Free and open source.