Thesis
The news-sentiment strategy runs a simple, legible rule: buy names when recent news sentiment turns positive, exit when it turns negative. It operates over a 24-name large-cap universe spanning tech (AAPL, MSFT, GOOGL, NVDA), financials (JPM, BAC, V, MA), healthcare (JNJ, UNH, PFE, ABBV), and staples/industrials (PG, KO, WMT, COST, CAT, XOM). It is currently live.
Recent Activity
The book has been quiet. The six most recent scheduled runs — spanning 2026-09-30 through 2026-10-07 — each executed zero trades and rejected zero. Cash has sat unchanged at $2,036.71 while total portfolio value drifted between roughly $10,155 and $10,388, reflecting mark-to-market moves on existing holdings rather than new positioning.
The last real activity was 2026-09-23, when the strategy rotated out of financials and into chips: it sold 35 shares of BAC at $55.96 and bought 4 shares of NVDA at $225.34. Earlier live buys included GOOGL (Jul 1), UNH (Jun 18), the original BAC position (Jun 12), MSFT (Jun 3), and AAPL (May 31). So the strategy does act — but infrequently, and the recent sentiment signal has given it nothing to do for over two weeks.
Backtest & Validation
Here the picture gets harder. Over 451 days the backtest returned just +0.19% (final equity $10,018.55, ~0.1% CAGR) on only 2 trades, with a 0% win rate, Sharpe of 0.74, and a negligible 0.05% max drawdown. The low drawdown is flattering but misleading: with so few trades, the book was largely idle, so there was little exposure to draw down. Turnover of 36.5% against two trades confirms this is a thin, low-activity sample.
Cross-validation did not pass. Of 4 folds, only 1 was positive — and tellingly, folds 1 through 3 logged zero trades each. The entire result lives in fold 4 (2025-12-16 to 2026-05-29), which produced the full +0.19% return, a 1.5 out-of-sample Sharpe, and both trades. A PSR of 0.923 looks strong in isolation, but the deflated Sharpe ratio drops to 0.551 after adjusting for 6 trials — a reminder that the headline Sharpe is not trial-robust.
Verdict
Strengths: the logic is transparent, the drawdown is contained, and the single active fold showed a respectable OOS Sharpe. Risks: the evidence base is dangerously thin. Two trades cannot support a performance claim, three of four folds never triggered, the win rate is zero, and validation failed. This reads as a strategy whose signal is too sparse to evaluate, not one that is proven. It warrants continued paper observation and a richer sentiment feed before any conviction — or capital — is extended.