Tell you tomorrow's close.
Nobody can. It shows you how your book behaves when the market moves instead.
Some are impossible. Some are bad ideas. We built everything else.
Each refusal is a promise about how your numbers are made.
Nobody can. It shows you how your book behaves when the market moves instead.
It runs the test. Sometimes that's the same thing. Often it isn't.
Every retry raises the overfitting bar. The fourth-try winner didn't win.
Signals fill at the next bar's open, never the close you signaled on.
Backtests are scored only on data the strategy never trained on.
Data-backed Vega answers come with the chart, the source and the query behind them.
Every figure keeps the method and data version behind it.
Strategies run sealed, with no network and no secrets.
We never place an order and never hold a dollar. That's your broker's job.
It printed the next day's close on a dot-matrix printer. We looked into building it. It can't be done, by anyone.
Risk that updates as your book moves, and research you can check.
Risk that recalculates every time the portfolio changes: VaR and CVaR, factor exposure, stress tests, limits and an optimizer.
Ask the question you'd give an analyst. Get an answer with a chart and a source for each claim.
Write the rules once, in blocks or Python. The backtest and the risk views read the same rules.
Graded on years the strategy never saw, with an overfitting check that gets stricter on every rerun.
Six asset classes, and every dataset shows its dates and whether it can replay a past date.
Open a ready-made screener, or describe the one you need in plain English.
Add your holdings, then run them through past crises like 2008, the 2020 crash and the 2022 rate selloff, or set a shock of your own.
Systematic research software from Kurtosis, Inc. Not investment advice. Backtested results are hypothetical. See the disclaimer.