Lorentzian Classification is built around a K-nearest neighbours machine learning model using Lorentzian distance metrics rather than traditional Euclidean measurements, engineered for equity traders operating within D1 to monthly position trading timeframes. The algorithm classifies current price action against thousands of historical analogues, weighting recent data more heavily to account for regime drift over time. Multi-timeframe and webhook capabilities make it compatible with both discretionary and fully systematic trading workflows.
In verified backtesting across a five-year sample, Lorentzian Classification records a 73% win rate and a 2.51 profit factor, with a maximum drawdown of just 10.2%. Equity position traders who want a statistically rigorous, machine-learning-driven classification system rather than indicator-based heuristics will find this among the most technically sophisticated options on TradingView. Its primary edge lies in AI pattern recognition — using Lorentzian geometry to surface historical price analogues that conventional Euclidean distance models structurally fail to detect.