The Best Hour to Trade ES Futures — What 18 Years of Data Shows

Every futures trader asks the same question at some point. When is the best time to trade ES futures?

Not which instrument. Not which setup. When. What hour of the day. What day of the week. What session window gives the best edge.

Most answers to this question are anecdotal. Someone trades the open for a few months, has a good run, and declares it the best hour. Someone else avoids Fridays because of a bad stretch in 2023. Forum threads are full of these personal experiences presented as universal rules.

I needed to answer this question with data. Not a month of data. Not a year. Eighteen years.

The dataset

The numbers in this post come from 89,774 qualifying signal interactions across 4,721 trading sessions, spanning January 2008 through early 2026. Every signal was evaluated against the full Midnight Grid level structure and Turning Points confluence engine that powers the Algorithmic Suite.

If you want the full methodology — how the signals were identified, how the three-layer verification works, how the permutation framework was built — I covered that in The Research Behind the Algorithmic Suite. This post assumes that foundation and focuses on one specific dimension: time.

When does the framework perform best? When does it perform worst? And does any time slot actually need to be avoided?

Hour by hour: the RTH breakdown

Regular Trading Hours for ES futures run from 09:30 to 16:00 ET. But the Algorithmic Suite framework evaluates signals starting at 09:00 ET, capturing the pre-open positioning that often sets the tone for the first move.

Here is the hour-by-hour win rate across 18 years of data.

The 15:00 hour — the final hour of regular trading — posts the strongest win rate of any RTH hour. The 09:00 open hour comes in strong, above the RTH average. The weakest RTH hour is 11:00, the midday lull — but still profitable across all 18 years.

The overall RTH average is well above breakeven.

Read that again. The weakest hour — midday — is still profitable across 18 years. That is not a dead zone. That is not a period to avoid. It is the weakest hour in a framework where every hour is profitable.

This distinction matters. I am not presenting this data to tell you to only trade at 3pm. I am presenting it so you can see the shape of the edge across the trading day and make informed decisions about when to allocate your attention.

Why the open and close are strongest

The pattern is not surprising if you understand market microstructure. But the data confirms what the theory predicts.

The opening hour concentrates institutional order flow. Portfolio rebalancing, overnight position adjustments, and reaction to pre-market news all converge in a narrow window. This creates directional commitment — real money moving with conviction. When the Algorithmic Suite identifies a signal at a key structural level during that window, it is capturing a moment where institutional flow and technical structure align.

The closing hour mirrors this dynamic. End-of-day positioning, options-related hedging flows, and portfolio adjustments create a second concentration of institutional activity. The closing hour's win rate reflects the framework capturing those structural inflection points when they matter most.

Midday is the inverse. Volume thins. Institutions step back. The market often ranges without conviction. The framework still identifies real interactions at real levels — the weakest RTH hour is still profitable — but the quality of the directional resolution is lower.

This is not unique to the Algorithmic Suite. It is a structural feature of how equity index futures trade. But having 18 years of data to quantify the magnitude of the difference is useful.

RTH versus overnight: the quality gap

This is the finding that surprises most retail traders.

There is a persistent belief in the futures trading community that overnight sessions — the Asian and European windows — offer easier trading. Less competition. Thinner order books. More predictable moves.

The data tells a different story for this framework. RTH signals are materially stronger in quality. The win rate differential is clear and consistent across the full 18-year dataset.

RTH averages roughly 6.9 to 7.5 qualifying signals per day. The minimum is 1. The maximum is 50. The median sits around 8. That is enough volume to build a real session around without forcing trades.

Overnight signals exist. They are not worthless. But if you are a prop firm trader managing a finite daily loss limit, or a discretionary trader who needs to choose when to sit down and focus, the data supports prioritizing RTH.

This does not mean avoiding overnight entirely. It means understanding that the structural advantage the framework captures — the interaction between Midnight Grid levels and Turning Points signals — produces its highest-quality resolution during hours when institutional participation is highest.

Day of the week

Every day of the week is profitable. That is the headline. No day needs to be avoided.

But the distribution is not flat.

Wednesday is the strongest day. Across the full dataset, Wednesday averages the highest average daily return on a single MES contract after all friction — commission and slippage included. Mid-week in general shows the strongest performance, with Tuesday and Thursday also running above average.

Friday is the weakest day. Not negative. Not a loss day. But the edge is thinner — slightly flat to marginally positive. This aligns with known microstructure dynamics. Friday afternoon sees reduced institutional participation, early position squaring ahead of the weekend, and lower volume in the final hours. The framework still captures real interactions at real levels, but the directional follow-through is shorter.

Monday performs in the middle of the range. The overnight gap from the weekend sometimes creates dislocations at key levels, which can produce strong signals. But Monday also carries the uncertainty of weekend news digestion, which dampens consistency relative to mid-week.

The practical takeaway: if you trade five days a week, keep trading five days a week. Every day is profitable. But if you need to pick three days — for lifestyle reasons, or because your prop firm restricts session count — Tuesday through Thursday is where the data is strongest.

Month of the year

There are no dead months.

March leads the calendar. Every other month is profitable. There are no seasonal gaps where the framework breaks down or requires you to stop trading.

This finding was important to me during the research process. Many trading approaches show seasonal fragility — strong in high-volatility months, weak in summer doldrums, or vice versa. The Algorithmic Suite shows stability across the calendar year.

I covered the regime-conditional testing — how the framework performs across 5 distinct volatility environments — in Does This Strategy Work in Every Market Regime?. The monthly data reinforces those findings from a different angle. Whether you are trading in a historically calm January or a volatile October, the structural edge holds.

What this means for prop firm traders

I built the Algorithmic Suite with serious futures traders in mind. A significant portion of those traders are working within prop firm evaluation frameworks — TopStep, Apex, FTMO, and similar programs.

These traders face constraints that most retail traders do not. Daily loss limits. Maximum drawdown thresholds. Trailing drawdowns that never reset. Session requirements. Some programs require a minimum number of trading days. Others cap your maximum position size.

For these traders, the time-of-day data is not academic. It is operational.

If your prop firm evaluation gives you a $1,500 trailing drawdown and you need to pass within 10 trading days, you cannot afford to trade low-probability hours and burn through your cushion. The data shows you where the framework's edge is sharpest: the opening hour, the closing hour, and mid-week sessions. That is information you can build a session schedule around.

It also shows you what not to worry about. Every hour is profitable. Every day is profitable. Every month is profitable. You do not need to overthink this. You do not need to avoid Mondays or skip July. The framework holds across the full calendar.

The optimization is at the margins. And at the margins, the data says: show up for the open. Show up for the close. Trade Tuesday through Thursday if you can only pick three days. And prioritize RTH over overnight.

The danger of over-optimization

I want to be careful here. Because the data invites a tempting conclusion that I do not think is correct.

It would be easy to read this post and decide: I will only trade at 15:00 on Wednesdays in March. That is where the numbers are highest.

That is not what the data supports. That is overfitting a filter to the best-performing slice and ignoring the base rate. The base rate is strong everywhere. Every hour, every day, every month.

The year-by-year consistency — which I detailed in the year-by-year breakdown — shows that the edge is structural, not concentrated in a few lucky time windows. The hour-of-day and day-of-week data adds nuance to that finding. It does not replace it.

Use this data to inform your schedule. Not to build a fragile rule set that breaks the moment conditions shift.

What I actually do with this data

I trade RTH. I focus on the open and the close. I trade every day of the week but give myself more patience on Fridays.

That is it. No complex session filter. No monthly rotations. The framework does the work of identifying structural interactions at key levels. The time-of-day data tells me when those interactions resolve most cleanly.

The simplest application of this data is the most robust one.

The Algorithmic Suite

Midnight Grid. Quantum Vision. Turning Points.

Three indicators. One framework. Tested across 89,774 signals, 4,721 sessions, and 18 years of every market environment the modern era has produced.

The data shows when it works best. It also shows that it works everywhere.

Start Your 7-Day Free Trial

Algorithmic is charting software for decision support on TradingView. It is not financial advice. Trading involves risk. Outcomes depend on your rules, risk management, and execution. Past performance does not guarantee future results.