TradeHouse Overfit Validator Research Report Copilot Pricing Honest backtests. We deflate the Sharpe leaderboards inflate.

Your backtest is probably lying to you.

Leaderboards rank strategies by raw Sharpe. We rank them by the Sharpe that survives purged cross-validation and trial-deflation — the same gate that promotes our own live quant fleet. Paste a strategy. Get the honest verdict in seconds. Free.

Free: Is your strategy overfit?

Probability of Backtest Overfit (PBO), Deflated Sharpe (DSR), and a multiple-testing-corrected verdict — no signup.

Hypothesis → p-value

Test a signal idea with honest, multiple-testing-corrected significance. Open research tool →

NL → strategy copilot

Describe a strategy in plain English; we draft the spec. preview Join the waitlist →

Why we're different

Most retail quant tools optimize for a number that looks good in-sample. A strategy with Sharpe 12 on 30 trades and a -49% drawdown will top a leaderboard and lose your money. We expose the integrity layer (purged-CPCV, PBO, DSR, sample-uniqueness) as the product. Own the data, not the dream.