(Est. 2026)
Our work

High level predictive quantitative math has never been this accessible in history. Anyone who is dedicated to their role as a student has the tools in the palm of their hand to understand and benefit from the most complex ecosystems of inventions in modern history.

Quant etc. is a publication that researches how predictive mathematics is used on a day to day basis to produce value. We use our research in combination with trading APIs and other execution tools to test our findings first hand.

Our Work

Market microstructure · 7 min read Why Wall Street Pays So Much to Save a Microsecond The same order book in three languages, agreeing byte for byte. Compiled builds ran 14–19× the Python reference — and much of the gap between C++ and Rust turned out to be a hash function. Portfolio construction · 8 min read Why the “Perfect” Portfolio Suddenly Stops Working The best regime strategy beat a plain volatility target by 0.03 of a Sharpe point — by earning less. It survives a placebo test and wins only 20 of 33 specification changes. Reported as fragile. Crypto derivatives · 8 min read The Hidden Interest Rate Holding Crypto Markets Together Screens showed 7.4% annualised funding; the trade collected 11.5% gross and kept 10.3%. Maximum drawdown −10.9% on a book described as market-neutral. Machine learning · 8 min read Can ChatGPT Beat a Coin Flip on Wall Street? No model’s confidence interval excluded 50%, at any horizon, on any of six markets. The machine-learning models finished bottom. The language model never ran — and we say so. Market design · 7 min read Would the Stock Market Be Fairer If It Paused? Batching removed the latency race entirely — 44% of volume to zero — and cost real price accuracy, fill rate and delay. The interval’s effect is U-shaped, not monotone.