The thesis
News-sentiment is a long-only signal strategy: buy names showing positive recent news sentiment, exit when sentiment turns negative. It runs across a 24-name large-cap universe spanning tech (AAPL, MSFT, GOOGL, NVDA), financials (JPM, BAC, V, MA), healthcare (JNJ, UNH, PFE, ABBV), staples (PG, KO, WMT, COST) and industrials/energy (XOM, CVX, CAT, HON). The strategy is currently live.
Recent activity
Since late May the book has put on five executed buys: AAPL (6 @ $312.06), MSFT (4 @ $428.23), BAC (35 @ $55.93), UNH (4 @ $402.85) and GOOGL (5 @ $360.40). Notably, these are all entries — no exits have printed, despite the thesis's promise to sell on negative sentiment. The most recent week of scheduled runs (July 22–29) each closed "0 executed, 0 rejected," holding cash flat at $990.06 while total equity drifted between $9,879 and $10,224. At roughly 90% deployed, the strategy is currently sitting on its hands.
Backtest and validation
The headline backtest is modest: +0.19% total return over 451 days, a 0.1% CAGR, Sharpe 0.74, and a shallow 0.05% max drawdown — across just two trades with a 0% win rate. Turnover was 36.5% and fees a trivial $2.
The walk-forward validation did not pass. Of four folds, only one was positive, and folds 1–3 placed zero trades at all. Every ounce of performance comes from fold 4 (Dec 2025–May 2026): +0.19% return, Sharpe 1.5, two trades. Out-of-sample Sharpe reads a flattering 1.5 and the probabilistic Sharpe (PSR) is 0.923 — but the deflated Sharpe ratio (DSR) of 0.551, after six trials, tells the sober story: the edge is barely distinguishable from noise.
Strengths and risks
On the plus side, risk control looks genuine. A sub-1% drawdown and low turnover mean the strategy is not bleeding capital chasing signals, and the single active fold produced a respectable risk-adjusted number.
The risks dominate, though. The sample is tiny — two backtest trades cannot support any statistical claim, and a 0% win rate alongside a positive return implies the gains are unrealized paper marks rather than booked wins. Three of four validation windows generated no signal at all, suggesting the strategy is dormant most of the time. The failed validation gate and marginal DSR argue firmly against committing real capital here.
Verdict
News-sentiment is a low-risk, low-conviction experiment. It hasn't lost money, but it hasn't shown it can reliably make any either. Until it trades more often — and actually exercises its exit rule — treat it as a work in progress, not a deployable edge.