The Rubik’s Cube Secret to Better Trades

TradeSmith’s CEO pitches a new AI-driven “signals” system that scans ~2.09 million potential trades daily to identify recurring factor combinations that historically predict high-probability short-term moves. The firm claims some signals show 90%+ historical accuracy and that a one-year backtest of a signals portfolio outperformed the S&P 500 by roughly 3-to-1. In internal beta tests, the top 100 trades averaged 2.6% gains in nine trading days versus the S&P 500’s 0.4% (roughly 7x performance). The piece highlights example stock moves (LRCX, DECK, CAT, NVDA, LMT, HCA) and promotes a public launch event and beta access on April 22, 10 a.m. ET. Market impact: the article frames the system as a potential short-term alpha generator versus the US SP 500, suggesting increased trading activity and interest in AI signal-based strategies if results hold in wider use.