Sign inCreate account
← Strategy library

Surprise-Scaled Reactionthe number against the forecast, not the number

Evidence: StrongProfessionalNews · Event

A release on its own carries no information. What moves price is the gap between what printed and what was expected — and the size of that gap, measured against how large surprises usually are for that release.

Andersen, Bollerslev, Diebold and Vega measured this directly across six years of real-time quotes: announcement surprises produce jumps in the conditional mean of spot rates. Not announcements. Surprises.

They also found the reaction is asymmetric — bad news moves price more than good news of the same size — which is built into the sizing rule below rather than left as a curiosity.

Category
News · Event
Timeframe
M1 execution · the surprise computed before the release
Suitable pairs
EURUSD · GBPUSD · USDJPY — tightest-spread majors only
Session
Scheduled US and European release times
Indicators / tools
A calendar with consensus forecasts · the historical standard deviation of that release's surprises · tick volume

Valid setup

NW-01 · SURPRISE-SCALED REACTION — VALID SETUP EURUSD · M1 · standardised surprise of −1.8 — bad for the euro, so full size SHORT 1.08158 1.08223 1.08288 1.08353 1.08418 1.08482 1.08547 BEFORE — flat, waiting UNPRICEABLE THE ADJUSTMENT — thirty minutes ORDINARY MARKET AGAIN TP1 then TP2 1R STOP = beyond the release candle + 0.3 × ATR ENTRY = close of the SECOND M1 candle TP1 = 1R TP2 = 2R, or thirty minutes 1 2 3 4 5 release +30 min 1.08418 1.08344 1.08270 1.08196 ① Consensus and the surprise deviation both recorded beforehand ② Release: standardised surprise −1.8, past the 1.0 gate ③ First candle skipped — the spread is at its widest there ④ Enter at the close of the second candle ⑤ Full size: bad news for the currency moves price more ⑥ Flat at thirty minutes whatever the price

Exact rules

Entry

  • Before the release: record the consensus forecast and the standard deviation of that release's own past surprises. Without the second number there is no way to tell a large surprise from a normal one.
  • Standardised surprise = (actual − consensus) ÷ that standard deviation.
  • Gate: trade only when the absolute standardised surprise is above 1.0. Below that the jump is inside the noise the market already priced.
  • Enter at the close of the second M1 candle after the release, in the direction of the surprise. The first candle is unpriceable — the spread is at its widest there.
  • Asymmetry rule: for a surprise that is bad for the currency, use full size. For good news of the same magnitude, use two thirds — the measured reaction is smaller.

Stop loss

  • Beyond the extreme of the release candle, plus 0.3 × ATR(14) on M1.
  • That is wide — typically two to four times a normal M1 stop. It is wide because the volatility genuinely is, and tightening it only converts a wide stop into a certain one.
  • Risk 0.25% per trade. One release, one position, no adding.

Take profit / exit

  • TP1 = 1R, half the position.
  • TP2 = 2R, or a time exit thirty minutes after the release.
  • Flat at thirty minutes regardless. By then the adjustment the research describes has finished and you are holding a directional bet on an ordinary market.

Build Surprise-Scaled Reaction

Reading this page is free and always will be. Taking it is what costs — the complete MQL5 for the rules above, yours to keep, permanently. Five of them cost $10 and they never expire.

Sign inWhat it costs

More from News trading

Spike Fadewhen the reaction outruns the surpriseLiquidity Recovery Entrylet the spread come back firstThe Pre-Release Straddlethe strategy the research argues against

Trading carries risk and most retail accounts lose money. Nothing on this page is financial advice or a recommendation to trade. The chart is an illustration of the rules on constructed candles — it is not a record of a trade and not a backtest.