The thesis
News-sentiment runs a simple, legible rule: buy on positive recent news sentiment, exit on 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). The appeal is interpretability — every entry and exit maps to a headline-driven signal you can audit after the fact. The risk is equally obvious: sentiment is noisy, often lagging, and prone to whipsaw.
Recent activity
Live paper trading tells a two-speed story. Between May 31 and July 1 the strategy executed five buys — AAPL (6 sh @ $312.06), MSFT (4 @ $428.23), BAC (35 @ $55.93), UNH (4 @ $402.85), and GOOGL (5 @ $360.40). Since then it has gone quiet. The six most recent scheduled runs (July 9–16) executed zero trades, rejecting one candidate on July 13, 14, and 15. Cash has sat pinned at $990.06, with total account value drifting between roughly $9,990 and $10,251 — the latest run on July 16 marked $10,251.44.
That pattern — heavy deployment followed by a hold — is consistent with the thesis: once sentiment-positive names are bought, the strategy simply waits for a negative trigger to exit.
Backtest and validation
Here the picture cools. Over 451 days the backtest booked a +0.19% total return (final equity $10,018.55), a 0.10% CAGR, Sharpe of 0.74, and a strikingly shallow 0.05% max drawdown. Turnover was 36.53% on just two trades and $2 in fees. Win rate reads 0% — no realized winners, likely because positions remain open rather than closed at a loss.
The validation gate failed. Across four walk-forward folds, only one was positive — fold 4 (Dec 2025–May 2026), which alone carried the entire 0.19% return and a 1.5 out-of-sample Sharpe. The other three folds produced zero trades and zero return. Deflated metrics are mixed: PSR is a healthy 0.923, but the Deflated Sharpe Ratio sits at 0.551 across six trials, and one-in-four positive folds is the red flag the validator rightly caught.
Verdict
Strengths: tight drawdown control, low turnover, low cost, and a transparent rule set. Risks: the entire track record rests on a single fold and two trades — a sample far too thin to trust. Three of four folds saw the signal never fire, suggesting the sentiment threshold may be too conservative or the news feed too sparse to generate opportunities. Until it demonstrates consistency across regimes, news-sentiment reads as a promising sketch, not a validated edge. Keep it live, keep it small, and watch whether it can trade more than twice a year.