Thesis
The news-sentiment strategy runs on a deliberately simple premise: buy on positive recent news sentiment, exit on negative. It trades a 24-name large-cap universe spanning tech (AAPL, MSFT, GOOGL, NVDA), financials (JPM, BAC, V, MA), healthcare (JNJ, UNH, PFE, ABBV), and consumer/industrial staples (PG, KO, WMT, COST, MCD, HD, XOM, CVX, CAT). It is currently flagged live.
The appeal of a sentiment overlay is that it reacts to information the price hasn't fully absorbed. The risk, as always, is that headline sentiment is noisy, crowded, and quick to reverse.
Recent activity
The live book holds positions accumulated between late May and early July: AAPL (6 sh @ $312.06), MSFT (4 @ $428.23), BAC (35 @ $55.93), UNH (4 @ $402.85), and GOOGL (5 @ $360.40). Total equity has drifted in the $10,329–$10,525 range through early September, with cash pinned at just $990.06.
That low cash balance matters. Across the scheduled runs from September 1–8, the strategy executed zero new trades while rejecting one or two candidates on most days. In other words, the model is generating signals it cannot act on — likely constrained by available cash or risk gates. A live strategy that is fully invested and idle is not really being tested.
Backtest & validation
Over 451 days, the backtest returned just 0.19% (final equity $10,018.55, CAGR ~0.1%), on only 2 trades, with a 0.74 Sharpe, a tight 0.05% max drawdown, 36.5% turnover, and $2 in fees. The reported win rate is 0% despite the positive total return — a red flag worth watching, and a reminder that a two-trade sample tells us almost nothing.
The walk-forward validation is where the story turns cautionary. It did not pass. Of four folds, only one was positive — and every trade lived in that final fold (Dec 2025–May 2026: +0.19%, Sharpe 1.5, 2 trades). Folds 1 through 3 produced zero trades and zero return. So the entire track record rests on a single ~6-month window.
The higher-order stats are mixed: a PSR of 0.923 is encouraging, but the deflated Sharpe (DSR) of 0.551 — which discounts the 6 trials run — is far more sober, and the 1-of-4 positive-fold count is the honest verdict.
Verdict
Strengths: the thesis is intelligible, drawdown is minimal, fees are negligible, and the one active fold delivered a respectable 1.5 out-of-sample Sharpe. Nothing here looks reckless.
Risks: the sample is dangerously small. Two backtest trades, activity clustered in one fold, a 0% win rate, and a failed validation together say this strategy is undertested, not proven. The live account compounds the problem — near-zero cash means recent signals go unexecuted, so we aren't accumulating the fresh evidence we need.
The honest read: an interesting hypothesis that has not yet generated enough independent trades to trust. Until it trades across more regimes — and frees up cash to act on its signals — treat news-sentiment as a candidate under observation, not a validated performer.