First and Last 30 Minutes of Trading: What the Research Shows

Most intraday "patterns" you see online never get tested. This one has been tested twice, both times in the Journal of Financial Economics. Here's what the researchers found, what they think causes it, and what it can and can't tell you as a Nasdaq-100 futures trader.

The first study: the half-hours at each end of the day

In 2018, Lei Gao, Yufeng Han, Sophia Zhengzi Li and Guofu Zhou published Market intraday momentum in the Journal of Financial Economics (vol. 129, no. 2). Using high-frequency data on the S&P 500 ETF from 1993 to 2013, they split each trading day into half-hours.

Their main finding: the first half-hour return, measured from the previous day's close to 10:00 ET, predicts the return of the last half-hour, from 15:30 to 16:00 ET. When the market is up overnight and through the first half-hour, the last half-hour tends to be up too.

The effect was stronger on four kinds of days:

• days with higher volatility;
• days with higher volume;
• days during recessions;
• days with major macro releases.

The authors found the same pattern in ten other heavily traded U.S. and international ETFs.

The second study: 60+ futures and a cause

In 2021, Guido Baltussen, Zhi Da, Sten Lammers and Martin Martens published Hedging demand and market intraday momentum in the same journal (vol. 142, no. 1). They looked at intraday returns on more than 60 futures contracts across stocks, bonds, commodities and currencies, from 1974 to 2020, and found strong intraday momentum "everywhere": the return from the previous close through most of the day predicts the return of the last 30 minutes, in the same direction. The effect then partly reverses over the next few days.

Their explanation is mechanical, not psychological. Some big players are "short gamma", mainly options market makers who've sold options and leveraged ETFs that have to rebalance every day. To stay hedged, they have to buy after the market goes up and sell after it goes down, and a lot of that rebalancing happens near the close. The authors call it hedging demand that trades in the same direction as the day's move.

1. 01Does the early-day move predict the last half-hour?

   S&P 500 ETF, 1993–2013 (Gao et al.); 60+ futures, 1974–2020 (Baltussen et al.) · Yes, in both studies, in the same direction

A visual summary of the article's table — not a performance chart or a new backtest.

Short gamma, explained without equations

Picture a leveraged ETF that promises twice the daily return of an index. With $1 billion of assets in the morning, it needs $2 billion of index exposure. Now say the index is up 2% by the afternoon. The fund's assets grow 4%, to $1.04 billion, so it needs $2.08 billion of exposure. But its existing exposure has only grown to $2.04 billion. To get back to twice its assets, it has to buy about $40 million more of the index near the close, after a rise.

If the index had dropped 2%, the same math would force the fund to sell after a drop. These are round numbers to make the point, not a description of any real fund. What matters is the direction: these funds have to trade with the day's move, every day, and they do it late in the session.

Options dealers who've sold options deal with something similar. As the market moves, the hedge they need shifts in the direction of the move, so they end up buying into rallies and selling into drops to stay neutral.

Why the cause matters

A pattern with a believable cause is more likely to last than one found by trying a ton of rules. Here, you can actually observe the cause: leveraged ETF assets, options open interest and estimates of dealer positioning are all published. When those positions are big, so is the end-of-day flow.

The cause also explains why the effect fades over the next few days. End-of-day buying or selling is a hedge, not new information about value, so prices tend to give some of it back.

What this means if you trade NQ

The pattern lives at the end of the day. Our Daily Bias covers 9:30 to 12:00 ET. The research doesn't say anything directly about that window. It describes the link between the morning and the last 30 minutes of the cash session.

The morning still matters. In both studies, the predictor is the move from the previous close into the trading day. A strong morning, either way, is the setup the research describes. So even if you're flat by noon, the morning's direction gives you context for the afternoon.

News days are the strongest case. Gao and his coauthors found the effect was stronger on days with big macro releases. Those are also the days with the biggest morning moves, which is why our pre-market checklist starts with the calendar.

It's a tendency, not a daily rule. The studies report statistical relationships across thousands of days. On any given day, the last half-hour can go the other way.

How you could check it on your own data

You don't need high-frequency data to take a rough look:

1. For each day, record the return from the previous close to 10:00 ET.
2. Record the return from 15:30 to 16:00 ET.
3. Count how often the two have the same sign, and compare that with how often any two half-hours do.
4. Split the days into release days and quiet days.

If the link looks stronger on release days, that matches the research. If it looks weaker since 2020, that's worth knowing too: published patterns often fade once everyone knows about them.

Why we're cautious

• Costs. The last 30 minutes are liquid in index futures, but every round trip costs commission and spread, and a small average effect can disappear after costs, as we've seen in other replications.
• Crowding. Once a pattern is published, more traders try to grab it, which can shrink it.
• Changing positioning. The whole thing depends on how much short-gamma exposure is out there, and that changes with options volumes and leveraged ETF assets.
• Our own data. We haven't measured the effect on NQ intraday data yet. We plan to, with the method stated ahead of time.

Limits

• The first study used the S&P 500 ETF, not the Nasdaq-100. The second included equity index futures among more than 60 markets but mostly reports pooled results.
• Both studies describe historical periods, ending in 2013 and 2020.
• Neither one tests a strategy for retail traders after their own costs.

Sources

• Gao, L., Han, Y., Li, S. Z., & Zhou, G. (2018). Market intraday momentum. Journal of Financial Economics, 129(2), 394–414. Washington University profile
• Baltussen, G., Da, Z., Lammers, S., & Martens, M. (2021). Hedging demand and market intraday momentum. Journal of Financial Economics, 142(1), 377–403. IDEAS/RePEc

FAQ

What is intraday momentum? The tendency for the market's direction early in the day to carry into the last 30 minutes of the session.

Does the first half-hour predict the close? In the S&P 500 ETF from 1993 to 2013, the first half-hour return predicted the last half-hour return, according to Gao et al. (2018).

Why does the market often move the same way at the end of the day? The leading explanation is hedging: options dealers and leveraged ETFs that are short gamma have to trade in the direction of the day's move, often near the close.

Does this work on Nasdaq futures? Baltussen et al. (2021) found intraday momentum across more than 60 futures markets, including equity indexes. We haven't published our own NQ test yet.

Can I trade the last half-hour pattern? Maybe, but the published results don't include your costs. Test it on your own data first.

---

Research and education only. Not financial advice. Futures trading involves substantial risk of loss.

How this article was made: drafted with AI assistance from the research papers listed above; facts and figures checked by Gabriel Décarie and Karel Panneton.