Premarket Futures: How to Read the Overnight Session

How the Overnight Session Builds

The premarket isn't one long stretch of nothing. It has phases, and each one leaves a different kind of mark on the chart.

The reason I split it up like this is that not all overnight moves deserve the same respect. A 15 point rally built at 2:00 AM on nothing gets faded constantly. The same 15 points built from the European open onward, with volume increasing, is a different animal and I treat it as real.

One more time worth knowing about, even though it lands after the open: European cash markets close at 11:30 AM ET. You'll often see a flush or a sharp reversal right around then as European traders square up, and it catches people who assume the late morning is dead.

The Levels I Mark Before the Bell

This is the actual output of my premarket work. Not a prediction, just a short list of prices where I expect something to happen.

Overnight high and low. The two levels the premarket session creates itself, and the most important ones on this list. The first hour of the cash session tests one of them the majority of the time, and how that test goes tells you a lot. A clean break with volume behind it and price holding above the overnight high is a market that wants higher prices. A poke through that immediately gets sold back inside is a failed breakout, and those give some of the cleanest trades of the morning.

Yesterday's high and low. Obvious ones, and they matter because they define whether today is opening inside or outside the previous day's range. Everything about my expectations changes on that question.

Yesterday's close. The settlement price is where every position in the market got marked overnight, and it acts like a magnet during the session more often than almost any other single price. When a gap starts to fail, this is the first target I look at.

Yesterday's value area and point of control. These come off my market profile, and they're how I know whether price is at a fair price or an extreme one. I also keep any naked POCs still sitting above or below within reach, because unfinished business from previous sessions tends to get revisited.

That's five or six prices most mornings. It takes about ten minutes, and the discipline is in stopping there. Early on I marked everything I could find and ended up with a chart like a spider web, and a chart like that doesn't make you sharper, it makes you hesitate. If every price is a level then nothing is.

Reading the Overnight Range

Here's where most premarket analysis stops short. People check the direction and skip the two things that actually shape the day: how wide the overnight range is, and where it sits.

Width, compared to a normal day. If ES typically covers 40 to 80 points in a cash session and the overnight already produced 60, a big chunk of today's expected movement has been used up. Those sessions frequently turn into rotation inside the overnight range instead of continuation, and traders who chase the open get chopped. The reverse is more interesting to me: a tight overnight range, say 15 points on a contract that usually does 50, is a coiled market. The range breaks in the first hour a lot of the time, and the break tends to run because so little of the day's movement has been spent.

Where it sits against yesterday's value. An overnight range still sitting inside yesterday's value area means nothing changed while I slept, so I come in expecting balance: rotation, edges getting respected, fades working. An overnight range that's travelled entirely above or below yesterday's value means the market repriced overnight, and my whole approach shifts to figuring out whether the new prices get accepted.

The shape of it. Did price trend all night in one direction, or chop around a middle? A one-way overnight session means the move is already mature and the easy part is behind you. A choppy, balanced night means the day still has fuel in the tank. This is why "futures are up 12 points" is meaningless on its own. Twelve points of quiet drift inside yesterday's value is a nothing morning. Twelve points from a data-driven gap out of a two-day balance area is a completely different day, and I'd trade it completely differently.

Overnight Inventory: Who's Already Long

This one took me a while to appreciate, and it explains a lot of opening behaviour that otherwise looks random.

Compare where price traded overnight to yesterday's closing price. If the whole overnight session happened above the close, then everyone who bought during those hours is holding a winner into the bell, and the market is what's called long inventory. If it all happened below, inventory is short.

Now think about what those holders do at 9:30 when real liquidity finally arrives. A lot of them take the money. That's why a market can open, sell off hard for ten minutes, and then turn around and spend the rest of the day going higher. The early drop wasn't weakness, it was overnight longs getting flushed out so the day can start from a cleaner position.

The more extreme the inventory, the more likely that early correction. So when the entire night traded well above the close and price starts dropping at the open, I'm not shorting into it. I'm waiting to see where the correction stops, because that's usually a much better long than anything the first five minutes offers.

This is also part of why I don't take a trade in the first few minutes. I let the inventory sort itself out, let the opening range establish, and then trade the market that's left.

Gaps and What the First Hour Decides

When the open lands outside yesterday's range, you've got a gap, and gaps are the one premarket situation where I genuinely slow down.

Two questions get answered in the first half hour or so. Does price build out there? If the market starts trading time and volume above the old range, holding the gap and stacking activity at the new prices, the gap is being accepted and fading it is expensive. Or does it get sold straight back in? If price can't hold and slips back inside yesterday's range, the fill scenario is live, and yesterday's close is the first place I'd expect a reaction.

What I don't do is take a position in the first minute based on the gap alone. I've watched plenty of traders short a gap-up at 9:30 because "gaps always fill" and then sit through a trend day that never comes back. Gaps fill often enough to be worth knowing about, and not often enough to be a strategy.

There's a decent rule of thumb from the profile world here too. When price opens outside yesterday's value area and then trades back inside it and stays, it will often rotate all the way across the value area to the other side. It's not a law, but it gives you a target and it beats guessing.

The 8:30 Window

Everything above can be rewritten in a single second at 8:30 AM. That's when CPI, the jobs report, jobless claims and GDP land, and a big miss will produce a 30 point candle in ES on a night that was otherwise flat.

What happens after is usually worth waiting for. The first reaction is algorithms and it frequently overshoots, then a chunk of it gets given back in the next ten or fifteen minutes as humans actually read the report. That second move is the one with information in it. On big data days I'll often skip the first fifteen minutes of the cash session entirely and let the market decide what the number meant before I risk anything on it.

The practical premarket job is smaller than that though. I check the calendar the day before, so I know what's scheduled and at what time before I mark a single level. If there's a Fed day on the calendar, 2:00 PM changes my whole plan for the afternoon, and I'd rather know that at 9:00 AM than find out at 1:55.

The Relative Strength Read

This is the part of my premarket I'd miss most if someone took it away, and it costs nothing to do.

I have ES, NQ, YM and RTY on one screen, and before the open I look at how far each has moved overnight in percentage terms. Not the point moves, the percentages, because that's the only way to compare them fairly.

When NQ is out in front and RTY is flat, that's a narrow tape being carried by a handful of tech names, and breakouts in the broader indexes tend to fail. When RTY is leading, money is going into domestic risk and the move usually has broader participation behind it, so I trust continuation more. When YM leads and NQ lags, money is rotating out of tech into the older names. And when all four are moving together in the same direction by similar amounts, that's the cleanest kind of premarket, because whatever moved the market overnight moved all of it.

The divergences are the useful part. If I'm about to buy a breakout in ES and NQ is red on the day, I want a much better reason than the chart in front of me. I trade ES, RTY and YM daily, and this read is a big part of how I decide which one to actually be in on a given morning.

Overnight, it's also worth glancing at what Europe did. If the DAX has been strong all morning and US futures are following, the move belongs to Europe, and there's a reasonable chance it slows down or reverses once European traders close their books at 11:30 ET.

My Premarket Routine, In Order

It's the same sequence every session, and the times below are relative to the 9:30 open.

The whole thing is ten to fifteen minutes, and the written version lives in my pre-market checklist. Then the levels get handed over to order flow, which is what tells me whether a level is holding or breaking when price finally gets there. Premarket work gives me the where. Order flow gives me the when.

Premarket Mistakes I See Constantly

Questions I Get About the Premarket