The thesis
news-sentiment is a long-only, event-driven strategy over a 24-name large-cap universe spanning tech, financials, healthcare, staples and energy (AAPL, MSFT, NVDA, JPM, JNJ, XOM and peers). The rule is simple and intuitive: buy on positive recent news sentiment, exit on negative. The appeal is obvious — news moves prices, and a disciplined rule beats reacting to headlines by gut. The question is whether the signal fires often enough, and cleanly enough, to matter.
Recent activity
The book is live, and executed trades have been sparse. A cluster of opening buys ran through the summer — AAPL (May 31), MSFT (Jun 3), BAC (Jun 12, 35 shares @ $55.93), UNH (Jun 18) and GOOGL (Jul 1) — followed by a single rotation on 23 September: it sold its 35-share BAC position at $55.96 (bought at $55.93, essentially a wash) and opened a 4-share NVDA position at $225.34.
Since mid-October the scheduled runs have gone quiet. Six consecutive sessions from Oct 12 to Oct 19 executed nothing at all, with two orders rejected (Oct 12 and Oct 19). Cash sits at $2,036.71, and the paper portfolio has drifted between roughly $10,345 and $10,510 — movement driven by held positions, not fresh signals.
Backtest and validation
Here the numbers turn cautionary. Over 451 days the backtest produced just 2 trades and a +0.19% total return (~0.1% CAGR), with a Sharpe of 0.74, a max drawdown of only 0.05%, turnover of 36.5%, and a trivial $2 in fees. Final equity landed at $10,018.55 — barely off the starting line.
Validation did not pass. Of four walk-forward folds, only fold 4 (Dec 2025–May 2026) traded at all, contributing both of the strategy's trades (+0.19%, Sharpe 1.5). Folds 1 through 3 were completely flat with zero trades. That leaves just 1 of 4 folds positive. The probabilistic Sharpe ratio is encouraging at 0.923, but the deflated Sharpe — which penalises for the 6 trials run — is a modest 0.551.
Strengths and risks
Strengths: the strategy is cheap to run, keeps drawdown negligible, and its logic is transparent and sensible. A high PSR hints the realised Sharpe is unlikely to be pure noise.
Risks dominate the picture. The entire track record rests on two trades inside one fold — statistically meaningless. A reported win rate of 0% despite a positive return is a red flag about how thin and inconsistent the sample is. And the signal simply does not fire often: three of four folds were silent, and recent live sessions have been idle or rejected.
Verdict: this reads as well-built plumbing in search of an edge, not evidence of one. It belongs in paper trading until the signal activates far more frequently — only then will any of these metrics earn trust.