Why We’re Building a Trading Research Lab in Public

The internet has no shortage of strategies that look great on paper. We're building something different — a lab where every experiment is documented, every result is published, and the data speaks for itself.

The Problem With Most Trading Research

Most trading content has an incentive problem. Strategies get written up after they’ve already worked. Backtests get tuned until the curve looks good, then the methodology is presented as if it was clean from the start. Losses go unmentioned. The result is an internet full of confident-sounding analysis that was designed, consciously or not, to look credible — not to be credible.

The information looks authoritative. The track record is silent. You have no way to tell the difference between someone who has actually tested what they’re claiming and someone who ran a spreadsheet backward until it confirmed what they already believed.

The Premise

A real research lab publishes the failures.
If you only show the wins, you’re running a marketing campaign.

The moment we publish selectively, the entire exercise becomes worthless. The scorecard means something only when it’s complete — wins, losses, and the ones that are genuinely inconclusive.

What This Lab Actually Is

Code Assassins is a systematic trading research lab. Every strategy we test runs through the same six-stage pipeline — hypothesis, backtest, paper trading, verdict. The results get published regardless of outcome. A strategy that fails gets a failure post. A strategy that looks promising in backtesting but falls apart in paper trading gets documented exactly that way. No spin.

This is not a newsletter with trade ideas. It’s not a course built on theoretical edge. It’s a lab — one where the methodology is transparent, the assumptions are stated upfront, and the scorecard is live and public.

The Lab Pipeline — Six Stages, Every Experiment
  • 01
    Hypothesis & Data Pull Define the exact question before writing a line of code. Pull clean, point-in-time data. Document every assumption.
  • 02
    Backtest Engine Run the strategy against historical data with proper safeguards against the most common sources of false results.
  • 03
    Results Storage Every trade, every metric, every equity curve — stored in full. No cherry-picking the window that looks best.
  • 04
    Labs Post Builder Write up the methodology and results before moving to live testing. Forces honest interpretation before real money is at stake.
  • 05
    Live Paper Trading Run the strategy forward in real market conditions, in real time, before the verdict is published.
  • 06
    Publish the Verdict Pass, fail, or inconclusive — with the full reasoning. Every lab ends here.

Why Build It in Public

Transparency changes the incentive structure. When the scorecard is live, the only way to build credibility is to actually be right — consistently, over time, with the losses included. There’s no selective disclosure, no conveniently chosen start date, no footnote that quietly excuses everything above it.

There’s a secondary reason too. Most of the institutional research infrastructure that professional investors rely on isn’t accessible to individual investors at a reasonable cost. The tools exist. The methodology is documented in academic literature. What’s missing is someone willing to do the unglamorous work of building it properly and publishing the results honestly. That’s the gap this lab is designed to fill.

What’s Coming

The first lab experiments are already in the pipeline. Before those verdicts drop, we’re building out the Concepts library first — because a verdict means nothing if the reader doesn’t know how to evaluate a backtest, what survivorship bias does to results, or what the Sharpe Ratio is actually measuring.

Foundations before labs. Labs before verdicts. That’s the sequence.

The Standard We’re Holding Ourselves To

Every result we publish will be reproducible. Every verdict will include the full methodology, not a summary of it. Every failed experiment will publish alongside every successful one. If a strategy can’t survive scrutiny, the right answer is to say so.

This is not a difficult standard to articulate. It’s a difficult standard to maintain when the results are bad, the methodology has a flaw you didn’t catch until after publishing, or the experiment that took three weeks to build comes back inconclusive.

We’re publishing anyway. That’s the whole point.