TradeClaw

Open trading research for testing ideas, modeling costs, and reproducing every result.

Self-hosted by default

Evidence

  • Track record
  • Studies
  • Prospective ledger
  • Methodology
  • Open data

Lab

  • Candidate feed
  • Screener
  • Backtest

Build

  • Self-host guide
  • Documentation
  • API reference
  • GitHub

© 2026 TradeClaw. MIT licensed.

Terms|Privacy|Trading involves risk. Research candidates are not trade instructions or financial advice.
TradeClaw
EvidenceLabBuild
Evidence→

Open trading research lab · 5,815 eligible sized candidates

Test trading ideas.See where they fail after costs.

Current finding: Our average trade lost after costs.

TradeClaw is an open trading research lab. Across 5,815 eligible sized candidates, average modeled net expectancy was negative after stated fee and slippage assumptions. Inspect every result, test your own idea, or self-host the complete evidence trail.

Explore the evidenceSelf-host TradeClaw
  • Self-hosted
  • MIT licensed
  • Open data

Cost-aware evidence field

One dot per OHLCV-resolved sized candidate

Raw data

R = the amount planned to risk on one trade.

Calculation methodology
Before modeled costs, per trade
-0.01R
− Modeled fees + slippage, per trade
0.55R
≈ 0.187% of trade size
= After modeled costs, per trade
-0.56R
Separate simulation: compounded result (1% risk per trade)
-100%

Follow the evidence

From finding to reproduction.

Start with what happened, understand the cost assumptions, test a variation, then reproduce the work yourself.

  • 01 · FindingSee the recordStart with the cost-adjusted result, including the result that failed.
  • 02 · CostsUnderstand the dragSee how fees, slippage, and turnover change the apparent edge.
  • 03 · EvidenceInspect the studiesCompare observed records with separately labeled modeled research.
  • 04 · TestTest an ideaChange the market, range, and method before reading the metrics.
  • 05 · ReproduceSelf-host the labRun the same code, database, and evidence trail on your own machine.