How to Trade News With Order Flow

News releases are where the most traders get hurt in the shortest time. The initial spike on a CPI print (scheduled on the BLS release calendar) or an FOMC decision (dates on the Fed’s official calendar) is a liquidity vacuum: spreads blow out, the book empties, and price whips both directions before it decides. Trying to trade the first tick is gambling. Order flow gives you a better job to do, wait out the spike, read where the aggression actually settles, and trade the real move once the noise clears.

What actually happens at the number

When the headline hits, algorithms react in milliseconds and price gaps toward the surprise. But that first move is thin. Liquidity providers pull their resting orders a second before the release to avoid getting run over, so the depth of market is a shell. A handful of contracts can move price several points because there’s nothing in the book to stop them. That’s why the spike often reverses violently: it wasn’t backed by real committed size, it was a vacuum filling and then correcting.

The mistake is treating that first spike as the move. On a candlestick chart the spike looks like conviction. On the tape it’s often the opposite, a burst of thin aggression into an empty book, immediately faded. You cannot tell the difference from price alone, which is the entire reason to read the flow.

BIDPRICEASK5,376.00145,375.7595,375.5065,375.25empty book: spread blown out5,375.005,374.7575,374.50105,374.255Liquidity providers pull their orders: a few contracts move price several points.
A second before the print, liquidity providers pull their resting orders and the DOM is a shell. With nothing in the book to stop them, a handful of contracts move price several points, which is why the spike whips and reverses.

Wait for the book to refill

The first discipline of news trading with order flow is patience measured in seconds. Right after the print, the spread is wide, the DOM is thin, and slippage is brutal. Any entry there fills badly and sits at the mercy of the whipsaw.

Give it fifteen to sixty seconds. The book refills as liquidity providers come back in, the spread tightens, and now you can actually read something. The question you’re answering during that wait is simple: once the dust settles, which side is aggressing, and is price holding the move or rejecting it?

Reading the real reaction

Once liquidity returns, the flow tells you whether the initial move was real.

Delta confirming or diverging. If price spiked up on a hawkish surprise and cumulative delta keeps building positive as the book refills, real aggressive buyers are behind the move and it has legs. If price is holding the highs but delta is sagging or diverging, the spike was thin and the aggression has already dried up, that’s a fade setup, not a continuation.

The tape’s character. Tape reading after the print tells you who’s in control. Aggressive, one-sided prints lifting the offer means buyers mean it. Two-sided churn with no progress means the market is undecided and you should stay out.

Absorption at the extreme. If the spike ran into a level and got absorbed, heavy aggression, no further progress, the move is failing right there. Passive orders are fading the spike, and the reversal is the trade.

Trapped spike-chasers. Everyone who bought the top of the spike is offside if it reverses. Those trapped traders become fuel for the fade, exactly the same dynamic as a false breakout, just faster and more violent.

A playbook for the release

  1. Flatten before the number. Don’t hold a position into a scheduled high-impact release unless that’s your explicit strategy. The spike will hit your stop at a random price.
  2. Don’t touch the first spike. Let the vacuum fill. No entries into a thin book with a blown-out spread.
  3. Wait for liquidity to return, roughly fifteen to sixty seconds, until the spread tightens and the DOM refills.
  4. Read the settled flow. Is delta confirming the move or diverging? Is the tape one-sided or churning? Did the spike get absorbed at a level?
  5. Trade the real move. If aggression is real and price holds, enter with the move. If the spike got absorbed and delta diverges, fade it as the trapped chasers bail.
  6. Use a structural stop and smaller size. Ranges are wider after news, so the stop distance is bigger, which means fewer contracts. Let the volatility set your size, per stop placement with order flow.

A worked example

CPI prints hotter than expected. ES spikes down 12 points in three seconds on almost no volume, the book was empty. If you shorted the spike you filled terribly and now sit in a position that’s about to reverse. Instead you wait. Over the next forty seconds the spread tightens and volume returns. Price is sitting near the lows around 5,370, and you watch the tape: aggressive sellers keep hitting the bid but price won’t make new lows, and delta makes a higher low against the spike. The initial down-move got absorbed.

The sellers exhaust, buyers start lifting the offer, and price ticks up off 5,370. You go long 5,373 with a stop at 5,366, below the absorbed spike low, sized down because seven points is a wide stop for the conditions. Price snaps back to 5,388 as the trapped spike-shorts cover. The move you traded wasn’t the spike, it was the real reaction that formed once the flow could be read.

5,370 — passive buyers — absorbed spike low−700−900−500−400Aggressive selling absorbed, delta makes a higher low…and price makes no new lows below 5,370Reversal → long 5,373 (stop 5,366)
Once liquidity returns, sellers keep hitting the bid near 5,370 but price won’t make new lows and delta prints a higher low: the down-spike got absorbed. You go long 5,373 with a stop below the absorbed low, not into the vacuum.

Where news trading fits

News is the extreme version of every order flow skill: liquidity, delta, absorption, and trapped traders all compressed into a few frantic seconds. If you can read the flow through a release, the rest of the session feels slow. It leans on the same entry timing and support and resistance reads as any other setup, just faster. For the full context, see the order flow strategies guide and the order flow trading guide.

Frequently Asked Questions

Should you trade the initial spike on a news release?

No. The first spike happens into an empty order book, liquidity providers pull their resting orders around the release, so a handful of contracts can move price several points on almost no real volume. That thin move frequently reverses violently because nothing committed was behind it. Entries into the spike fill with terrible slippage and sit at the mercy of the whipsaw. The order flow approach is to wait for the book to refill and trade the real reaction that follows.

How long should you wait after a news release before entering?

Roughly fifteen to sixty seconds, until the spread tightens back and the depth of market refills. You’re waiting for two things: tradeable liquidity so you’re not slipped on entry, and enough settled flow to read which side is really in control. The exact time varies with the event and instrument, but the signal is the same, when the book looks normal again and delta and the tape give a clear read, the conditions to trade have returned.

How do you tell a real news move from a fake spike?

Read delta and the tape once liquidity returns. A real move shows cumulative delta building in the direction of the spike with one-sided aggressive prints and price holding its extension. A fake spike shows price holding the extreme while delta sags or diverges, often with absorption at the level where the spike ran out. If the aggression that caused the move has dried up and passive orders absorbed it, the spike was thin and the fade is the trade.

Why do stops need to be wider when trading news?

Because volatility and ranges expand around a release, so the structural invalidation point sits further from your entry. A level that would justify a four-tick stop in a quiet session might need ten or more after the number. You don’t tighten the stop to feel comfortable, that just gets you shaken out by the post-news chop. You keep the stop at the real structural level and reduce position size so the wider distance still equals your fixed dollar risk.