Should you trade individual assets or build a portfolio?

Eighty frozen strategies across 105 Binance perpetuals, one candidate stream, five execution policies and 37 out-of-sample months, separating what a capacity constraint costs from what choosing well...

Memecoin trading, Prediction Markets, Online Casinos: Which is worse?

Does a wider, less liquid universe improve a systematic book?

Thirty Binance perpetuals instead of six, with the strategies, the grids, the schedule and the costs held fixed. What breadth does to the book, whether thin markets pay better, and an addendum asking...

Does tick data beat OHLC for return forecasting?

A controlled comparison on 900 symbol-days of nanosecond US equity trades and quotes: same bars, same folds, same models, differing only by whether the tape-derived columns are present.

Do trading strategies need an economic rationale?

A matched comparison between strategies written from an economic mechanism and strategies enumerated from an indicator catalog, on sixty Binance perpetual contracts with an identical search budget on...

Backtest Overfitting & the Deflated Sharpe Ratio

Quantifying selection bias under multiple testing and gating backtests on the Deflated Sharpe Ratio.

How long should you wait before trading a newly listed perpetual?

How a Binance perpetual changes over its first year in volatility, funding, spread and book depth, and whether a fixed rule library earns anything different on a young contract than on a mature one...

How much can machine learning improve losing trading strategies?

A model layer trained on the trades of 420 losing technical rules, used to filter, size or rank their signals, walked forward for 43 months on US equities and Binance perpetuals against taking every...

A topological way to build portfolios

Select strategies from the edge of the correlation network, the ones orthogonal to the common market mode, and a thin crypto edge sharpens where chasing Sharpe self-destructs.

Should a strategy's recent results change how it manages the next trade?

A pre-registered test of equity-curve management: five families of rule reading a strategy's own completed-trade history, across two markets and two speeds, plus a barrier-only addendum that removes...

What matters more in crypto: the market or the strategy?

A pre-registered contest between picking the market and picking the strategy, run on a balanced six-by-six grid of hourly Binance perpetuals across fourteen out-of-sample quarters.