Trading scheduled events: CPI, decisions and earnings

Scheduled events — inflation prints, central bank decisions, earnings releases — are the rare moments when everyone knows exactly when new information arrives, just not what it says. That foreknowledge transforms the market around the release in ways that are visible, repeatable, and brutal to traders who treat 8:29 and 8:31 as the same market. They are not. For a few minutes, the market you practiced on is replaced by a different one with thinner books, wider spreads, and faster opponents.

The good news: the transformation follows a script. Learn the script and scheduled events become, at minimum, hazards you stop walking into.

The book before the number

Watch a live order book in the minutes before a major release and you can see the market brace. Makers face a stark asymmetry: quoting through the release means their resting orders are free options for whoever reads the number fastest. If the print is a shock, every stale quote gets picked off at pre-news prices. So, rationally, they retreat — quoted size shrinks, the levels thin out, spreads stretch from their normal width to several multiples of it. Volume often goes quiet in the final minutes. The market is not calm; it is holding its breath. The crucial implication: liquidity is at its worst at the exact moment the most people will demand it.

The release: the first second

When the number hits, the sequence unfolds in tiers of speed. Machines parse the release and trade within milliseconds, sweeping whatever thin liquidity remained on the wrong side. Price gaps — it does not travel smoothly through the intervening levels, it teleports. Spreads blow out further as makers wait to see where fair value lands, then gradually tighten as quotes return over the following minutes. Anyone who sent a market order into that window paid a spread several times normal, in a book several times thinner, against counterparties several thousand times faster. The fill reports read like a penalty notice.

Why first moves reverse

Veterans of event trading repeat the same observation: the first move is frequently wrong, or at least drastically overdone. The reversal has mechanical causes, not mystical ones:

• Headline first, nuance later. Algorithms trade the headline figure instantly; the qualifying detail — the composition of the number, the one paragraph in the statement that changes everything, the guidance buried below the earnings headline — gets read by humans minutes later and often points the other way.
• Thin books exaggerate. The initial move happens in a near-empty book, so modest aggressive flow produces an outsized price change. As liquidity returns, price renegotiates from a fuller market.
• Stop cascades overshoot. The first leg triggers clustered stops, whose forced orders extend the move beyond what the news justifies. When the forced flow exhausts, the snap-back is the market removing the overshoot.
• Positioning unwinds. If the crowd leaned one way into the event, even a confirming number can reverse: those positioned correctly take profits into the move, and there is no one left to buy the news everyone already owned.

None of this means fading every first move is a strategy — sometimes the first move is exactly right and keeps going all day. It means the first print of post-event price discovery is a draft, not a verdict, and acting on the draft costs the worst execution of the session.

Macro releases versus earnings

The mechanics rhyme across event types, but the blast radius differs. A major inflation print or rate decision moves everything — equities, FX, gold and crypto reprice together, because the number feeds the shared engine of rate expectations. Earnings are concentrated: one stock gaps violently while the broader market shrugs, though large bellwethers can drag their whole sector. Single-name event risk is also lumpier — a stock can reopen 15% from its prior price, a scale of gap that diversified macro instruments rarely produce. Concentration sizing matters accordingly: an event that can gap an instrument double-digit percent deserves a fraction of the size you would carry through a routine macro release.

Sizing and order choice around events

Practical discipline for scheduled releases compresses to a short list:

• Decide in advance: trade the event, or stand aside. Improvising at 8:30:05 with a live position is how accounts produce their worst fills of the year.
• If holding through, size for the gap, not the average day. Assume the spread will be several times normal and your stop, if hit, fills with serious slippage. If that arithmetic threatens the account, the position is too big.
• Prefer limit orders in the aftermath. Once the dust begins settling, limits let you set your worst price in a market still quoting wide; market orders keep paying the event premium for minutes after the headline.
• Let the spread normalize before judging the move. Spread width is a live gauge of maker confidence — while it remains stretched, price discovery is still in progress and the chart is writing in pencil.

Key takeaways

• Before a release, makers withdraw: depth thins and spreads widen, so liquidity is worst precisely when demand for it peaks.
• The release itself gaps prices through thin books — market orders in that window pay multiples of normal cost.
• First moves frequently reverse for mechanical reasons: headline-first algorithms, thin-book exaggeration, stop cascades, and crowded positioning unwinding.
• Stops held through events provide gap-shaped protection: triggered at the trigger, filled at the far side of the gap.
• Earnings concentrate the blast in one name and can gap double-digit percent; size single-name event risk far smaller than macro exposure.
• Decide before the event whether you are participating; the worst trades are improvised during the spread blowout.