Overfitting validation

Can you trust
that backtest?

A high-performing backtest is usually just fitted to past data. TestMyLogic is a desktop app that rigorously tests your crypto strategy's backtest for overfitting — check it before you trust those returns.

The app runs the statistics. Your AI agent explains them.

Download the free beta

No orders
No exchange keys
No fund movement
Your data stays local

Hundreds of combos backtested, checked three ways

Three statistics each ask, in their own way, whether the backtest is real.

Analyzes, one by one, the dozens to hundreds of backtests a strategy-parameter sweep produces.

The app's strategy survey screen: twelve representative cumulative-return curves from the sweep fanned out above a dashed buy-and-hold baseline, with the highest-return backtest highlighted.

How it works

Download market history

Pull public OHLCV and funding history into a local workspace.

Run a parameter sweep

Pick a strategy and a parameter grid; it backtests all of them.

Analyze the results

Trade diagnostics shows how robust the winner is; PBO and walk-forward each deliver a verdict.

Ask the AI agent you already use

We do not ship an agent. Connect the desktop app you already have — Claude, ChatGPT or Antigravity — and you can drive all of it by asking: download data, run the sweep, read the verdicts, query results with SQL, move the app's screens. Ask it to explain any term or number in plain words, or to compare the runs and point out what held up.

YouDownload BTC, run a backtest sweep with a MACD strategy, and analyze the results.

AgentBTC daily history was already here, so I topped it up — 3,279 rows, no gaps. Swept 120 MACD_Momentum combinations and moved the app to the PBO view.

PBO is 11.7% — overfitting risk “Not distinguished”: the value landed where a field with a real edge and a field with none both land. The walk-forward reads the same — “Not distinguished”, because the re-picked selection scored 49.4% across 95 folds and 45.2%–54.8% is where a strategy with no skill at all lands. The returns agree — walk-forward +320%, level with the median strategy’s +321% and far behind simply holding BTC at +857%. The hindsight winner’s +1,490% was never available in advance.

The drawdown does favour the walk-forward: -66.0% against -76.6% for holding — the one card it wins. And that is the point: two nearly independent gates, one reshuffling time and one walking it forward, both looked for an edge here and neither could tell this run from no skill. Most real runs end exactly this way — and an edge you cannot tell from luck is not an edge you can trade.

Validate your backtest yourself

Pick your data and your strategy, and the statistics answer.