Thesis and Approach
The news-sentiment strategy operates on a straightforward premise: buy large-cap equities when recent news carries positive sentiment, exit when tone turns negative. Its universe spans 24 blue-chip names across technology, financials, healthcare, consumer staples, and energy — a deliberately diverse selection designed to maximise the pool of sentiment-rich coverage at any given time. The approach is theoretically grounded; academic literature has long documented short-lived return premia around news events in liquid equities.
Recent Activity
The strategy has been almost entirely silent in recent weeks. Scheduled runs on June 10–17 executed zero trades across six consecutive sessions, with portfolio value drifting between $9,694 and $9,791 against $4,438 held in cash. The lone exception was a buy of 35 shares of BAC at $55.93 on June 12. Prior entries include MSFT (4 shares at $428.23, June 3) and AAPL (6 shares at $312.06, May 31). With three open positions and a large uninvested cash reserve, the strategy is clearly finding few sentiment signals strong enough to meet its entry threshold in the current environment.
Backtest Performance
Over 451 days, news-sentiment generated a total return of 0.19% on exactly two trades, implying a CAGR of roughly 0.10% — barely above flat. The Sharpe ratio of 0.74 suggests marginally positive risk-adjusted returns, and a maximum drawdown of 0.05% reflects the strategy's extreme conservatism when signals are absent.
One figure warrants scrutiny: reported backtest turnover of 36.53% is surprisingly high for a two-trade history. This may point to a position-sizing or measurement artefact in the simulation engine and should be audited before the turnover figure is used in any comparative analysis.
Validation: A Caution Flag
The strategy did not pass validation, and the fold-level breakdown explains why. Of four time-sliced folds covering August 2024 through May 2026, only fold 4 (December 2025 – May 2026) produced any trading activity — the same two trades and 0.19% return that appear in aggregate metrics. Folds 1 through 3 returned zero trades, zero Sharpe, and zero drawdown across approximately 18 months.
The out-of-sample Sharpe of 1.50 and a probabilistic Sharpe ratio of 0.923 look attractive in isolation, but both figures are driven entirely by a single active fold with two trades. The deflated Sharpe ratio of 0.551 — which corrects for trial count and non-normality — is more informative and reflects the genuine statistical uncertainty. With only one positive fold out of four and six total trials, there is not enough evidence to conclude that the strategy carries durable edge.
Strengths and Risks
Strengths
- The 0.05% maximum drawdown confirms the cash-heavy posture limits downside during signal droughts.
- Fold 4's Sharpe of 1.50 is a real, if narrow, data point that the thesis can work in the right regime.
- The universe is well-chosen: liquid, widely-covered names with consistent intraday news flow.
Risks
- Signal frequency is the dominant concern. A strategy that sits idle for 18+ months cannot compound returns or generate meaningful validation data.
- The reported win rate of 0 cannot yet be interpreted — no round-trip trades have closed — but it must be tracked as positions eventually exit.
- Validation failure means this strategy should be treated as experimental; real capital allocation should remain minimal until multiple active folds accumulate.
Outlook
The next three to six months of live operation are the real test. If news-sentiment continues to produce near-zero signals, the sentiment threshold or signal pipeline should be examined: either the bar is too high, the news feed too sparse, or the universe needs rotation toward names with richer intraday coverage. The idea is sound — the implementation simply needs more innings before a verdict is possible.