The Short Version

This is a quantitative trading research platform. I look for strategies that have actual research behind them, test them against years of real data, and don't put money in until they pass walk-forward validation and Monte Carlo simulation. If a strategy can't prove itself on unseen data, it doesn't trade.

The Founder

I have a B.S. in Physics, a B.A. in Natural & Applied Sciences, and a Master's in Physics. My research was at Jefferson Lab, working on deep inelastic scattering experiments funded by DOE.

In particle physics, your code is your eyes. You can't watch a parton get knocked out of a proton and recombine into a shower of particles in 10^-23 seconds. You write models and analysis code, and that's how you see what happened. I spent years getting good at pulling real signals out of noisy data and knowing how much to trust the answer.

He wasn't wrong. Simons came from theoretical physics and built Renaissance. Shaw came from computational science and built D.E. Shaw. Derman went from particle physics at Columbia to running quant strategies at Goldman. It's the same problem in a different domain: signal extraction from noise.

That's what this platform does. Same discipline, applied to market data instead of collision data. No gut feelings, no chart patterns. Just math.

The Process

I start with research. Academic papers, AQR white papers, documented strategies from professional traders. I'm not looking for tips. I want edges where someone has done the work and shown a structural reason the opportunity should persist.

Most institutional strategies need capital or infrastructure that retail traders don't have, so I filter for what actually works with a small account. What's left gets backtested across years of real data, parameter-swept, then validated with walk-forward testing and Monte Carlo. I'm not looking for the prettiest backtest. I want the one that holds up on data it's never seen.

If it passes, I build a paper bot and run it on live market data with real slippage and commissions. The ORB futures bots went through this whole pipeline before real money went in. Nothing skips paper trading.

How I Work

• Hypothesis first. I don't optimize to fit historical data and call it edge. The strategy needs a reason to work before I test it.
• Uncertainty matters. A 70% win rate doesn't mean much if the confidence interval is wide. I bootstrap thousands of resamples to check if the edge is real or noise. Same standard as publishing a physics result.
• Real costs included. I model slippage at 66% of bid-ask spread, add commissions, and report P&L after all expenses. Most backtest results you see online skip this.
• Everything is published. Every trade, every loss, every drawdown. Check the journal.