Here's a trading idea that isn't from a guru — it's from a peer-reviewed finance journal. A study called "Market Intraday Momentum" found that how the market moves in the first half-hour of the day tends to predict how it moves in the last half-hour. If the morning was strong, the close was often strong too. This is a real, testable edge — and it maps cleanly onto a strategy you can build.
Who found it, and where
The paper was written by Lei Gao, Yufeng Han, Sophia Zhengzi Li, and Guofu Zhou, and published in the Journal of Financial Economics in 2018 — one of the top three finance journals in the world. It's not a blog post; it went through years of academic peer review.
What they actually did
They took 20+ years of minute-by-minute data on the S&P 500 ETF (ticker SPY) from 1993 to 2013. For each day, they measured the return of the first 30 minutes (from the prior day's close to 10:00am) and the return of the last 30 minutes (into the 4:00pm close). Then they asked: does the first predict the last? The answer was yes — clearly, and in a way that was too strong to be luck. The effect was stronger on high-volatility days, high-volume days, and big news days.
Why it should work
Two plain-English reasons. First, under-reaction: traders who see a strong morning don't fully price it in until the day forces them to, near the close. Second, late positioning: big players who need to get in or out often do it in the last half-hour, pushing price the same way it started. It's a behavioral and structural story, not magic.
Does it still hold — honestly?
The paper is on stocks (an S&P 500 ETF). Futures traders (MNQ, MES) track the same index, so the idea travels well — but you must test it yourself, because an edge on 1993–2013 stock data isn't a guarantee on today's futures. Effects like this can also shrink once they're published and widely known. That's the whole point of testing before trusting.
Build it in TapeScript
The strategy writes itself: "At 3:30pm, if today's return since the open is positive, go long into the close; if it's negative, go short. Exit at the close." Type that in plain English, and TapeScript turns it into PineScript and Python, backtests it with zero lookahead on real index-futures history, and tells you if the last-hour edge is actually there on the instrument and years you care about — then it will show you which sessions and volatility regimes it works best in (the paper says high-volatility days; you can verify that directly).
Citation: Gao, L., Han, Y., Li, S. Z., & Zhou, G. (2018). "Market Intraday Momentum." Journal of Financial Economics, 129(2), 394–414. Read it on SSRN.