Research on U.S. equities finds that some price pressure in the final 30 minutes can reverse over the next trading session. This article summarises that documented association and translates the signal into an implementable research design. “Predictive” here means statistically informative within the cited 1993–2019 sample—not a guarantee of positive returns in a new period or after transaction costs. The gross, dollar-neutral decile spread can fail after costs, market impact, auction participation, shorting constraints, altered closing mechanics or out-of-sample decay.
You’ve probably sensed it. Some days the market drifts to a close and you feel something building underneath the tape, without being able to name it. Other days it surges into the bell and reverses violently the next morning. Those intuitions aren’t random — they’re reflections of a measurable, documented association in the cited study. And today, I’m going to show you exactly how to quantify it.
“The market doesn’t hide what it’s going to do. It hides it in the data most people don’t bother collecting.”
I’ve spent years studying market microstructure. I’ve read the academic literature, run the backtests, and traded the signals live. What I’m sharing here isn’t theory dressed up as practice — it’s a specific, implementable signal that has held up across 15 years of equity market data, through bull markets, bear markets, and everything in between.