Why Quant Research Should Be Open
The secrecy myth
Most quant secrecy protects capacity-constrained execution details, not ideas. The ideas themselves — factor construction, regime detection, validation methodology — improve dramatically when exposed to peer review.
The execution edge — low-latency infrastructure, market microstructure insights, capacity constraints — that’s worth protecting. But the research methodology? That only gets better in the open.
What we share
We publish factor research, evaluation methodology, and tooling. We keep execution infrastructure private. That line has served open-source communities well for decades.
- Factor research: We publish our methodology, data sources, and results
- Validation frameworks: Walk-forward validation, out-of-sample testing, regime-aware backtesting
- Tooling: Open-source research frameworks and data processing pipelines
- Execution details: Low-latency infrastructure, order routing, market microstructure — these stay private
The benefits of openness
- Faster iteration: Peer review catches mistakes you miss
- Better methodology: Collective scrutiny improves validation standards
- Community building: Open research attracts collaborators, not competitors
- Reproducibility: Others can verify your results, building trust
Examples from our research
Our work on earnings revision momentum decay and LLM-extracted earnings sentiment is fully reproducible. The methodology is documented, the data sources are specified, and the code is available.
Join us
If you spend your weekends reading papers, you already belong here. Come build with us.
Related Reading
- Earnings Revision Momentum Decay in the Post-2023 Regime — example of our open methodology
- LLM-Extracted Earnings Sentiment as an Alpha Factor — reproducible financial AI research
- Five Backtesting Pitfalls That Fake Your Sharpe — open methodology for validation