Thesis
News-sentiment is an information-momentum play: it screens a 24-name universe of large-cap U.S. equities spanning technology, financials, healthcare, consumer staples, energy, and industrials, initiating buys when aggregate headline sentiment turns positive and exiting when it reverses. The core bet is that a brief window exists between a catalyst hitting the news wire and the market fully pricing it in — a window the strategy attempts to exploit systematically, running on a daily schedule after market close.
Recent Activity
The past week has been quiet. Scheduled runs on July 2, 3, 6, and 8 all logged zero executions. July 7 produced one rejection, indicating the signal did evaluate a candidate and actively screened it out rather than staying idle. The most recent executed trade was a buy of 5 shares of GOOGL at $360.40 on July 1.
Since late May, the strategy has built five open long positions: AAPL (6 shares at $312.06), MSFT (4 shares at $428.23), BAC (35 shares at $55.93), UNH (4 shares at $402.85), and GOOGL (5 shares at $360.40). No exits have been recorded in the live period. As of July 8, total portfolio value is $9,956.74 with $990.06 remaining in cash. Intra-week equity ranged from roughly $9,861 to $10,073, driven entirely by mark-to-market moves in existing positions rather than new activity.
Backtest Performance
The 451-day backtest produced a total return of 0.19% and a final equity of $10,018.55, with an annualized CAGR of 0.10% and a Sharpe ratio of 0.74. Maximum drawdown was a contained 0.05%, and total fees amounted to $2 across just 2 executed trades — reflecting the strategy's extremely low trading frequency. Reported turnover of 36.53% suggests that when positions are entered, they represent meaningful capital deployment relative to the portfolio.
The win rate reads 0%, but this is likely an artifact of all backtested positions remaining open without a logged exit — a data nuance rather than evidence of poor stock selection.
Validation: One Bright Fold, Three Silent Ones
The walk-forward validation did not pass. Of four sequential folds covering August 2024 through May 2026, only Fold 4 (December 2025 – May 2026) generated any trading activity or positive returns: 0.19% return on 2 trades, with a Sharpe of 1.5. Folds 1 through 3 recorded zero trades each — the signal never fired.
The Deflated Sharpe Ratio (DSR) of 0.551 captures this inconsistency. After accounting for the 6 strategy variants trialed and the return distribution's characteristics, the edge does not clear a conventional statistical bar. The Probabilistic Sharpe Ratio (PSR) of 0.923 is more encouraging — it implies a high probability the true Sharpe is positive — but PSR does not penalize for multiple comparisons the way DSR does. Both figures should be read together.
Strengths and Risks
Strengths: Minimal drawdown, negligible fee drag, and a deep-liquidity universe make this a capital-efficient vehicle when it does act. The OOS Sharpe of 1.5 in Fold 4 is a genuine data point worth tracking.
Risks: Three silent folds raise a fundamental question about signal frequency: a strategy that rarely fires cannot generate alpha at scale. The exit mechanism is entirely unproven in live conditions — all current positions are open buys with no observed sell trigger. The failed validation gate and a DSR below 0.6 argue for extended paper-trading observation before any capital commitment.
Verdict: The thesis is coherent and the Fold 4 result is encouraging, but the strategy needs to demonstrate that its sell logic works and that Fold 4 was not an isolated regime. Watch for the first live exit.