Johnson Controls Slides to 250th in Liquidity Amid $460M Volume

Johnson (JCI) closed on Sept. 25, 2025, with a 0.72% decline to $XXX.XX, trading on $460 million in volume that ranked it 250th among U.S. stocks by liquidity. The stock's performance came amid mixed market conditions as investors digested sector-specific developments.

Recent disclosures highlighted Johnson Controls' strategic pivot toward industrial automation, with management signaling $200 million in capital expenditures for AI-integrated manufacturing solutions by 2026. Analysts noted the move aligns with broader industry trends but cautioned about near-term margin pressures from R&D investments. The company also confirmed a $50 million expansion of its North American HVAC division, though the announcement lacked specific revenue targets.

Market participants observed that the stock's trading pattern showed increased short-term volatility, with intraday swings exceeding 1.5% despite moderate volume. This diverged from its historical beta of 0.9, suggesting potential positioning shifts among institutional investors. No material regulatory or litigation developments were reported during the period.

It looks like you’d like to test a daily-rebalanced portfolio that simultaneously buys (or equally weights) the 500 most-actively-traded U.S. stocks each day and then closes the position the following day. At present, our back-test engine is designed to evaluate one tradable instrument (one ticker) at a time—so it can’t directly simulate a 500-stock basket that changes constituents every day. To move forward, we have two practical options: 1. Approximation via an ETF or index proxy • Use an ETF such as SPY (cap-weighted S&P 500) or RSP (equal-weight S&P 500) as a stand-in for "owning 500 large, liquid names." • Backtest the 1-day holding rule on that single ticker. • This captures the high-liquidity large-cap universe, though it won’t exactly replicate daily volume re-ranking. 2. Narrow the scope to a smaller, fixed list of high-volume stocks (e.g., the 50 or 100 most-liquid names) and back-test them one by one, then average the results. • This requires many individual runs and is more time-consuming, but more closely follows your original idea. Please let me know which path you’d prefer (or if you have another approach in mind), and I’ll proceed with the appropriate back-testing workflow.