The thesis
The idea is about as clean as strategies get: buy names showing positive recent news sentiment, and exit when sentiment turns negative. The strategy runs across a 24-stock large-cap universe spanning tech, financials, healthcare, staples, and energy — AAPL, MSFT, GOOGL, NVDA, JPM, V, JNJ, XOM, and peers. There is real academic and practitioner support for sentiment-driven momentum, so the premise is defensible. The question, as always, is whether the signal fires often enough and cleanly enough to matter.
Recent activity
Live paper trading shows five executed buys between late May and early July 2026 — AAPL at 312.06, MSFT at 428.23, BAC at 55.93, UNH at 402.85, and GOOGL at 360.40. Since that GOOGL fill on July 1, the tape has gone quiet. The scheduled runs from August 21–28 executed zero trades, with several signals rejected outright. Cash sits at just $990.06 against a total book of roughly $10,465, so the strategy is effectively fully invested and short of dry powder — a plausible reason recent candidate buys are being rejected.
That is worth watching. A sentiment strategy that cannot rotate because it has no cash is not really expressing its thesis; it is holding whatever it bought earlier and waiting.
Backtest and validation
Here the picture is more sobering. The backtest covers 451 days and returns just 0.19% total (0.1% CAGR), on only 2 trades, with a Sharpe of 0.74 and a shallow 0.05% max drawdown. The low drawdown is a genuine positive, but it is the low drawdown of a strategy that barely does anything — turnover of 36.5% and a reported win rate of 0 underline how little the engine has had to work with.
Walk-forward validation makes the concern explicit: it did not pass. Across four folds, only one was positive. The first three folds generated zero trades — the signal simply never triggered — and all of the return came from fold 4 (Dec 2025–May 2026), which posted 0.19% and a 1.5 Sharpe on two trades. A probabilistic Sharpe of 0.923 looks encouraging until you deflate it for the six trials run: the deflated Sharpe drops to 0.551. That is the honest number, and it does not clear the bar.
The verdict
Strengths: an interpretable thesis, tight risk control, and a promising most-recent fold. Risks: an extremely thin sample, a signal that stays dormant for long stretches, a failed validation gate, and a near-zero cash buffer that is currently smothering new entries. Nothing here says the edge is fake — but nothing yet says it is real, either. The strategy needs more firings across more regimes before its live status is anything more than provisional. For now, treat it as a hypothesis under observation, not a proven engine.