The thesis
The news-sentiment strategy runs a simple, legible idea: buy names with positive recent news sentiment and exit when sentiment turns negative. It trades a 24-stock large-cap universe spanning tech (AAPL, MSFT, NVDA), financials (JPM, V, MA), healthcare (JNJ, PFE, ABBV), and staples (KO, PG, WMT). The logic is easy to reason about and easy to audit — a genuine strength in a book of strategies where opacity is the usual complaint.
Recent activity
The strategy is marked live and runs on a scheduled daily cadence. Activity is sparse. The 29 October run executed three buys — V (4 shares @ $410.29), NVDA (9 @ $213.49), and PFE (66 @ $29.92) — followed by a single KO buy (21 @ $95.13) on 3 November. Every other recent run reported zero executions.
The portfolio has drifted below its $10k base: the 5 November run logged total equity of $9,758.90 with $2,387.18 in cash, down from roughly par at the start of the window. That is a modest drawdown, but the direction is the wrong one, and the cash balance shows the book is only partly deployed.
Backtest and validation
Here is where caution is warranted. Over a 451-day backtest the strategy returned just +0.19% (CAGR 0.1%), with a Sharpe of 0.74, a tiny 0.05% max drawdown, 36.5% turnover — and only 2 trades. A recorded win rate of 0 alongside a small positive return suggests the gains are unrealised rather than booked.
Walk-forward validation did not pass. Of four folds, only one (Dec 2025–May 2026) traded at all; the other three produced zero trades and zero return because the signal simply never fired. That means the entire out-of-sample result — +0.19%, Sharpe 1.5 — comes from a single fold. The probabilistic Sharpe ratio looks reassuring at 0.923, but the deflated Sharpe, which penalises the 6 trials run, sits at just 0.551. One good window does not make a validated edge.
Strengths and risks
Strengths: the thesis is transparent, drawdowns are shallow, and fee drag is negligible ($2 total). It is a well-behaved, low-risk sleeve.
Risks: the sample is far too thin to trust. Two backtest trades and three idle folds mean the metrics are closer to noise than evidence, and failed validation should keep expectations low. The practical problem is activation — the sentiment gate rarely opens — so the strategy is neither clearly profitable nor clearly tested. Until it trades enough to generate a real distribution of outcomes, treat the numbers as provisional, not proof.