Thesis
News-sentiment runs a simple, legible rule: buy names 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), staples (PG, KO, WMT, COST), and industrials/energy (CAT, HON, XOM, CVX). That's a liquid, defensible sandbox — sentiment signals are cleaner and cheaper to trade in mega-caps than in thinly-covered small caps. The strategy is currently live.
Recent activity
The live book has gone quiet. Every scheduled run from July 20 through July 27 executed zero trades and rejected zero — cash has sat unchanged at $990.06 while total portfolio value drifted between roughly $9,879 and $10,175, closing July 27 at $10,059.95. The positions themselves were opened earlier: buys in AAPL (6 @ $312.06, May 31), MSFT (4 @ $428.23), BAC (35 @ $55.93), UNH (4 @ $402.85), and GOOGL (5 @ $360.40, July 1). Notably, every recorded trade is a buy — the "exit on negative" half of the thesis has not yet fired in the live book, so what we're really watching so far is the entry engine, not the full round-trip.
Backtest and validation
Over 451 days the backtest returned 0.19% (final equity $10,018.55, ~0.1% CAGR) on a Sharpe of 0.74 and a shallow 0.05% max drawdown, with modest 36.53% turnover and only $2 in fees. The headline caution: just two trades, and a reported win rate of 0 — which sits awkwardly beside a positive return and most likely reflects closed round-trips (of which there are effectively none) rather than mark-to-market gains.
Validation did not pass. Across four walk-forward folds, only one was positive — and it's the same fold that holds all the activity. Folds 1–3 (Aug 2024 through Dec 2025) produced zero trades and zero return; fold 4 (Dec 2025–May 2026) delivered the entire 0.19% at a Sharpe of 1.5 on its two trades. The out-of-sample numbers therefore are the full-sample numbers. Multiple-testing statistics tell the honest story: the Probabilistic Sharpe Ratio is a healthy 0.923, but the Deflated Sharpe Ratio falls to 0.551 once six trials are accounted for.
Strengths and risks
Strengths: a transparent thesis, a liquid universe, genuinely low drawdown, and a respectable OOS Sharpe where the strategy actually traded. Risks: the record is dangerously thin — one fold, two trades — so almost none of it is statistically load-bearing. Three of four folds never engaged, suggesting the sentiment trigger rarely clears its threshold, and the exit rule remains untested live. Until news-sentiment trades more often and completes some round-trips, treat these numbers as a promising sketch, not a verdict.