When a high-impact economic number prints — Nonfarm Payrolls, CPI, an FOMC decision, GDP — a trader faces a binary tactical choice. "Trading the release" (also called trading with the move or momentum/breakout trading the news) means taking a position in the direction of the initial price reaction, betting the market will keep going that way as the information is absorbed. "Fading the move" means doing the opposite: letting the first knee-jerk spike happen, then taking a position against it, betting the initial reaction overshot, was a liquidity-driven head-fake, or will mean-revert once context is weighed. The core tension is that both can be right — sometimes news triggers a sustained directional move (continuation), sometimes the first 30–120 seconds is a spike that reverses (the "V-shape"). Choosing wrongly in a low-liquidity, high-volatility window is one of the fastest ways to get stopped out, and the deciding variable is rarely the headline number itself.

The two approaches mechanically

Trading the release. Enter in the direction of the move immediately after (or within seconds of) the print. The thesis is that the surprise — actual vs. consensus forecast — is genuinely new information that the market will price in over minutes to hours, not all at once. Practitioners gate this on the size and direction of the surprise relative to the prior distribution of surprises, and on whether the move is confirmed by correlated markets (e.g., for a hot CPI: equities down, 2-year yields up, dollar up, gold down — all agreeing).

Fading the move. Stand aside through the first spike (commonly the first 1–5 minutes), then enter against it once the initial surge "fizzles." The thesis is that the first move is disproportionately driven by stop-runs, algorithmic orders hitting thin liquidity, and forced positioning unwinds rather than by considered repricing — so it overshoots and partially retraces. Fade entries are typically triggered by a failure to make new extremes, a reclaim of a pre-release level, or divergence between the headline move and correlated instruments.

A practical hybrid most professionals actually use: wait, don't predict. Let the algorithms fire their first salvo into available liquidity, watch whether price accepts or rejects the new level, and then trade the confirmed direction — which may be with or against the initial spike.

Why the first move is unreliable: market microstructure

The first seconds after a scheduled release belong to machines that parse the headline against preset thresholds and fire into the order book before humans react. Around the print, bid-ask spreads widen sharply and depth thins, so even modest order flow produces an outsized price move; price discovery and liquidity then recover over the following seconds-to-minutes as quote activity returns (ScienceDirect — Price discovery and liquidity recovery: Forex market reactions to macro announcements). This is the structural reason the initial spike both overshoots and is the most dangerous moment to enter: you cross a wide spread and risk slippage and gapping. It is also why guaranteed stop-loss orders are recommended for those who do trade the print (Investing.com — Trading the News).

Crucially, markets move on the surprise, not the level. The same CPI print can produce opposite reactions depending on consensus positioning, the Fed's prevailing tone, the level of yields, and risk appetite — so "the number was high, therefore sell" is an incomplete model (FBS Academy — Trading Strategy on NFP and CPI Releases).

How it's used in practice

• Surprise vs. consensus first. Both approaches start by comparing actual to forecast; an in-line number rarely justifies either trade. The internals (NFP: average hourly earnings, revisions, unemployment rate; CPI: core vs. headline, shelter) frequently override the headline.
• Cross-market confirmation. A "real" move shows correlated assets agreeing. When equities, rates, FX and gold disagree, the move is more likely to fade.
• Fade only after the spike exhausts. The canonical fade waits for the initial surge to stall, then enters on a failure/reversal signal with a stop beyond the spike extreme.
• Trade the release only on large, clean, confirmed surprises — the case where continuation is most plausible — and size small for the volatility.
• Most disciplined desks avoid the literal print and trade the second move (acceptance or rejection), which sidesteps the worst spread and slippage.

Standing & evidence

The honest picture is that neither "always fade" nor "always trade the release" is an established edge — the literature shows continuation and reversal occur with comparable frequency, so the rule must be conditional, not blanket.

• In favor of continuation/drift: the pre-FOMC announcement drift (Lucca & Moench, 2015) documented ~49 bps of average S&P 500 excess return in the 24 hours before scheduled FOMC announcements over 1994–2011 — but it did not appear before other major U.S. macro releases, and several studies find it largely disappeared after ~2015 (NY Fed Staff Report 512; The disappearing pre-FOMC announcement drift, PMC). It is a pre-release phenomenon, not a release-reaction strategy, and its instability is itself a caution.
• On under- vs. overreaction generally: Fama's (1998) survey of the long-horizon event literature concludes that apparent underreaction to information is about as common as overreaction, and that post-event continuation of abnormal returns is about as frequent as post-event reversal (Fama, 1998, Market efficiency, long-term returns, and behavioral finance; see also the PEAD review, ScienceDirect). Post-Earnings-Announcement Drift is robust evidence of continuation (underreaction) — but it is a multi-week equity-specific anomaly, not transferable to the seconds-to-minutes window of a macro print.
• Practitioner consensus is that the first spike is frequently a trap that chases retail traders into slippage, which is why "wait for the dust to settle" dominates desk practice (FBS Academy; Investing.com). This is widely believed but not rigorously base-rated in public academic work for intraday macro reactions — treat it as informed heuristic, not measured fact.

Strengths & limitations

Trading the release captures the largest, fastest moves when a genuine surprise reprices the market, but it pays the widest spreads, the worst slippage, and is wrong precisely when the spike is a head-fake. Fading profits from the common overshoot and reversal but has unbounded risk on the days the news is a regime change (a Fed pivot, a shock CPI) and price runs without retracing — the fader is then leaning against a freight train. The single most common misuse is treating the rule as unconditional — committing to always-fade or always-trade-with regardless of surprise magnitude, cross-market confirmation, and liquidity. The second most common is entering during the spike rather than after price has shown acceptance or rejection. Both approaches degrade in thin liquidity and around overlapping releases.

System relevance

This node sits under Economic Data Calendar & High-Impact Events and is a tactical companion to the surprise-magnitude and consensus nodes there. For Delvantic, the operative caveat for the Augustus trade-setup agent is a timing exclusion, not a setup: scheduled high-impact releases are a reason to widen stops, reduce size, or stand aside, because the spike/whipsaw window invalidates normal swing entry and stop logic. Augustus should consume the event calendar to avoid initiating or being mechanically stopped during the release window, and defer any "fade vs. follow" judgment to confirmed post-spike acceptance plus cross-market agreement — never to the headline number alone.

Sources

• ScienceDirect — Price discovery and liquidity recovery: Forex market reactions to macro announcements (spread widening, price-discovery and liquidity-recovery dynamics around releases)
• FBS Academy — Trading Strategy on NFP and CPI Releases (V-shape reversal, surprise-not-level, fading the open)
• Investing.com — For the Traders: Trading the News (algorithmic first reaction; GSLOs; trade acceptance/rejection, not the print)
• Lucca & Moench (2015), NY Fed Staff Report 512 — The Pre-FOMC Announcement Drift
• Kurov, Wolfe & Gilbert (2021), The disappearing pre-FOMC announcement drift, Finance Research Letters (PMC / NIH mirror) — drift essentially gone after ~2015, sample extended to 2019
• Fama (1998), Market efficiency, long-term returns, and behavioral finance — underreaction roughly as common as overreaction; continuation roughly as frequent as reversal
• A review of the Post-Earnings-Announcement Drift (ScienceDirect) — PEAD as a robust multi-week continuation anomaly
• Disputes flagged: "always fade the spike" is desk folklore, not academically base-rated for intraday macro; pre-FOMC drift is a contested, possibly-vanished pre-release effect and does NOT generalize to other releases or to reaction trading.