Research
Essays
Long-form on archetypes, statistics, and the line between edge and noise — written from the11-gate validation work, not from theory.
2026-07-19
Are FX swaps profitable? The spread is real, but the easy carry is not
FX swaps are a huge funding market, but covered interest parity absorbs most headline rate differences. The remaining basis is usually a scale-and-funding business.
Read →2026-07-19
Are Hyperliquid strategies profitable after fees? What the tests actually show
Hyperliquid yield, liquidation-fade and cross-venue funding tests show why venue data and execution costs matter more than attractive gross signals.
Read →2026-07-19
Do Polymarket BTC 15-minute strategies work? Five tests, no validated edge yet
We tested continuation, strike filters, spot lead-lag, complement discounts and BTC-to-altcoin lead-lag. The best result was promising but inconclusive.
Read →2026-07-19
Nine sports betting strategies tested across soccer, the NFL and MLB
Multi-season tests of nine soccer, football and baseball strategies found no validated rule; full results and failed hypotheses are published.
Read →2026-06-25
The placebo test: the one gate that retires the most strategies
Most backtests never face a placebo. We re-run every strategy 200+ times with shuffled labels — and it is the gate that has quietly retired the carry trade, the ETH/BTC pairs trade, and more.
Read →2026-06-24
Edge vs. noise: what eight retired strategies have in common
Eight trading archetypes have failed the gates — basket-grid, Forex Fury, D1 trend, night mean-reversion, trend-pullback, carry, NR7 and ETH/BTC pairs. Here is the pattern across all of them, and what separates them from the survivors.
Read →