Alcoa Shares Jump 2.94% on Strategic Shifts $280M Volume Ranks 354th in Market Activity

On September 12, 2025, Alcoa (AA) saw a 2.94% rise in its share price amid a 40.98% surge in trading volume to $0.28 billion, ranking 354th in market activity. The move followed a strategic shift in its energy portfolio and a production update that signaled short-term operational efficiency gains.

The company announced plans to divest non-core renewable energy assets, reallocating capital toward high-margin aluminum production. Analysts noted this could streamline operations and reduce debt burdens, though the impact on long-term earnings remains uncertain. Separately, Alcoa reported a 7% increase in third-quarter smelter output, driven by maintenance cycle optimizations at its Canadian facilities.

Market participants interpreted the volume spike as a reaction to the production report rather than broader sector trends. The stock's performance outpaced the S&P 500 industrials index, with technical indicators showing renewed buying interest above key resistance levels. Short-term volatility is expected as the company finalizes its asset restructuring timeline.

To run this strategy rigorously I need to build a daily, cross-sectional portfolio made up of the 500 NYSE/Nasdaq stocks that traded the highest dollar volume each day since 2022, buy them at that day’s close (or open) and liquidate them the next day. Our current tool-chain can back-test a single security or an event list quite easily, but it does not yet support a “rolling universe” portfolio out-of-the-box. To approximate (or to code around the limitation) we have two practical options: 1. Simplify the request to a single security (for example an ETF or index) – e.g. test an S&P 500 ETF with daily re-entries. 2. Treat the “top-volume” selection as a daily event, construct a composite equal-weighted return series outside the engine (by pulling prices & volumes for every U.S. listing), then feed that composite into the back-test engine. This is feasible, but means downloading price & volume history for thousands of tickers – a heavy data job that will take time and may hit external-data caps. Please let me know which direction you’d prefer (or if you have another idea), and also confirm: • Universe definition (all U.S. common stocks, or a subset?) • Execution price you want to assume (same-day close vs next-day open) • Whether equal weighting and zero transaction costs are acceptable defaults. Once I have that, I can lay out the data-retrieval plan and start the test.