What the Spread Does Around a Release — and Why It's the Entry Timer Nobody Watches

Ask a trader what they watch in the minutes around a release and you will hear about levels, candles, volume, maybe an indicator. Almost nobody says the spread. It sits in the corner of the platform, or is hidden entirely behind a setting, and it is the only thing on the screen reporting live on the condition everything else depends on: whether there is anyone on the other side.

This is a companion to the piece on why the first candle after a release lies. That one argued you should let conditions normalise before you transact. The obvious follow-up question is: normalise according to what? The answer is the spread, and it is the most useful thing on the screen precisely because it is the one number that cannot flatter you.

What the spread actually is, in one paragraph

The spread is the distance between the best price someone will buy from you at and the best price someone will sell to you at. It is not a fee your broker invented; it is the market makers' price for standing between buyers and sellers and carrying the risk in between. When that risk is low and predictable, they compete and the gap is small. When they are about to see a number nobody has seen, the risk is neither low nor predictable, so they either widen the gap or step away entirely.

That is why the spread is informative rather than merely annoying. It is a continuously updated read on how much uncertainty the people quoting prices think they are carrying right now. You are getting their opinion for free, several times a second.

The shape it makes around a scheduled release

Because the release time is known, the spread's behaviour around it has a shape. The exact magnitudes depend entirely on your instrument, your broker and the specific release, which is why this describes a shape rather than quoting numbers at you. What you will typically see:

• It starts widening before the release, not after. Quoting desks reduce risk ahead of the print, so the deterioration begins while the chart still looks calm. A position opened in this window pays for the uncertainty without having seen the number.
• It reaches its worst in the seconds immediately around and after the print, when the book is thinnest and the first orders are landing into it.
• It comes back in stages rather than all at once, as quoting desks re-engage and satisfy themselves that the initial move is not running away from them.
• It settles at a level that may be different from the pre-release baseline — sometimes for the rest of the session, if the release genuinely changed the picture.

The useful part is the third item. The recovery is gradual and observable, and it is a much better description of 'the market has calmed down' than anything you can read off a candle. A candle can look decisive while the spread says nobody is willing to quote you a real price.

Why this is an entry timer

Most entry timing rules are attempts to guess when the move is ready. This one is different: it does not tell you what price will do. It tells you whether the market is in a state where acting is expensive or ordinary. Those are separate questions, and conflating them is why people end up with a good idea and a bad result.

Used as a timer, the rule is almost embarrassingly simple. Do not transact while the spread is materially wider than its normal level for that instrument at that time of day. Not because a wide spread predicts anything, but because everything you do in that window costs more and fills worse, and the compensation for that cost is not visible on the chart you are looking at.

There is a second, subtler use. If the spread refuses to come back — if it stays wide well beyond the usual recovery — that is real information. It usually means the release did something the quoting desks are still digesting, and it is a reason to be smaller or to stay out entirely rather than a reason to hurry.

How to record it, so you are using your own numbers

You cannot apply 'materially wider than normal' without knowing what normal is on your instrument. That takes a few sessions of casual observation, not a research project. Two things to establish, in order:

First, your quiet baseline. Over three or four ordinary sessions with no high-impact release, note the spread at a few fixed times — a mid-session lull, the London open, the New York open, and late in the session. You are not after precision. You want a feel for the ordinary range and, importantly, how much it varies by time of day. A spread that looks alarming at one hour may be perfectly normal at another, and confusing the two will have you sitting out good conditions.

Second, the release profile. Extend the release log from the first-candle piece with a few spread checkpoints. For each high-impact release you care about, note the spread at these moments:

• Five minutes before the release — is it already deteriorating?
• Thirty seconds before.
• Ten seconds after the print.
• One minute, five minutes and fifteen minutes after.
• Whether it had returned to your quiet baseline by the fifteen-minute mark, and if not, how long it took.

That is six observations per release. After a handful of releases you will know your own recovery time, and you will have replaced a vague instinct about 'waiting a bit' with a checkpoint you can actually follow when the screen is moving and your judgement is worst.

One habit that makes this much easier: write the numbers down somewhere permanent at the moment you see them. Spread is not reliably reconstructible after the fact — most platforms chart bid or mid, not the gap — so a spread you did not record is a spread you have lost. This is the main reason the exercise is rarer than it should be, and the main reason it is worth doing.

What this does not do

Worth being blunt about the limits, because a rule that is oversold gets abandoned the first time it disappoints.

Waiting for the spread to normalise does not tell you which way to trade, and it will sometimes cost you a genuinely good entry — the cases where the initial move was the move and it never came back. That is a real cost and pretending otherwise would be dishonest. The argument for the rule is not that it never costs you anything; it is that the trades it declines are, on average, the ones where you were paying the worst prices of the day for the least information, and that the ones it costs you are the minority you could not have identified in advance anyway.

It also does not replace having a plan. A normalised spread with no idea attached is just permission to trade badly at a fair price. The spread answers 'is now expensive?', and that is all it answers.

FAQ

My platform doesn't show the spread. How do I see it?

Most platforms will display it if you turn it on — commonly as a column in the market watch or symbol list, or as an option to show both bid and ask lines on the chart, in which case the visible gap between them is the spread. If yours genuinely cannot, showing the ask line alongside the bid gets you there visually, which is enough for timing even if it is awkward for logging. It is worth the few minutes of digging through settings once.

Is a fixed-spread account immune to this?

It changes the shape of the problem rather than removing it. A fixed spread is fixed under normal conditions; providers generally reserve the right to widen or to stop quoting when conditions are disorderly, which is exactly the window under discussion. And the spread was never the whole problem — the thin book and the lag between headline and detail are properties of the market, not of your pricing model. If your spread genuinely does not move, you lose this particular instrument reading and should lean harder on simply waiting a defined number of minutes.

How much wider is 'materially wider'?

Deliberately not answered with a number, because any number would be wrong for most readers. It depends on the instrument, the broker, the account type and the time of day, and quoting a figure without those would be inventing precision. Your quiet baseline is the whole point: 'materially wider' means clearly outside the ordinary range you established for that instrument at that hour. Two or three sessions of casual observation gives you a better answer than any figure someone else could publish.

Does this apply outside news releases?

Yes, and it is arguably more useful there because it is less expected. Spreads widen at session rollovers, into weekends and holidays, in the first minutes of a session, and during unscheduled events. The same rule applies: the spread tells you whether right now is an ordinary time to transact. Scheduled releases are just the case where the effect is largest and most predictable, which makes them the easiest place to learn to read it.

If I'm already in a position, does any of this matter?

It matters for exits, which people forget. A stop triggered in the wide window becomes a market order into a thin book, so the exit fill can be well beyond where you placed it. That is not an argument for removing stops — it is an argument for deciding your exposure through a release before the release rather than during it, and for being honest that a stop is a request, not a guarantee, in those particular seconds.