InterestRateDecision ECB in 14h 57m
  1. Home
  2. Guide

How to trade interest rate decisions

A central bank decision is not a number to guess. It is a burst of orders at one exact minute, and price does a very particular thing around it. Here is the read, on real decisions, with the counts including the times it did not work.

Decisions measured105one-minute candles, eight banks, since Jan 24, 2025
Closed back67 / 10564% inside the hour-before range within an hour
Then the other side42 / 67of those, within four hours
Median first move21 pipsin the first 15 minutes
In brief
  1. A decision is news-induced liquidity: many orders land in the same minute.
  2. In the hour before, price tends to compress into a range. Stops rest just outside it.
  3. The first spike usually runs one edge of that range: a stop hunt.
  4. A five-minute candle that closes back inside the range says the hunt is done.
  5. The move that follows often belongs to the other side, and the press conference can add a second one.
  6. It does not always happen. Counted honestly, it closed back 64% of the time.

Why a decision is not like other news

Most of the time, price moves because orders arrive a little at a time. A rate decision is different: every trader, fund and algorithm that has been waiting for it acts in the same minute. We call that news-induced liquidity. For a short while there are far more orders than usual, on both sides, and the market has to find where they can be filled.

That is why the first reaction is so often violent and so often wrong. It is not a considered reading of the statement. It is a rush of orders going wherever there is something to trade against, and the places with the most resting orders are the stops that traders have left just beyond the recent high and low. Read the decision as a minute of liquidity first, and as information second.

This guide is built on 105 decisions of eight banks (Fed, ECB, BoE, BoJ, BoC, RBA, RBNZ, SNB), each measured on one-minute candles with the same rules every time. The rules are on the method page. The numbers say how often something happened. They do not say it will happen next time.

Step 1: the hour before

Before a scheduled decision, many traders stop trading. Spreads widen, volume thins and price tends to drift into a narrow range while everyone waits. We mark that range as a box: the high and the low of the 60 minutes before the announcement.

The box matters because of what sits just outside it. Traders who bought in the hour have their protective stops under the low; those who sold have theirs over the high. Those stops are orders waiting to be triggered, which is exactly what a burst of liquidity looks for. The median box across the 105 decisions was 14 pips wide, and the median first move after the announcement was 21 pips: the first spike is usually larger than the whole hour that came before it.

Step 2: the first spike is a stop hunt

At the announcement the first spike goes one way, fast, and runs past the edge of the box. This is a stop hunt: the stops beyond the edge are triggered, filling the opposite side's orders. The market is not saying anything about the decision yet.

In our counts the first spike left the box within 30 minutes in 104 of 105 decisions. In the other 1, price stayed inside the hour-before range for the first half hour. Where it did leave, the spike went through by a distance that is easy to measure and hard to trade by hand, which is the reason to wait for the next step.

Step 3: the close back

The signal is not the spike. It is what the market does when the orders run out. When a five-minute candle (counted from the announcement) closes back inside the box within 60 minutes of the announcement, the stops have been taken and price has refused the new level. We call that the close back.

It happened in 67 of the 104 decisions that broke the box, and 37 of them did not close back within the hour: the first move simply held. A rule that works 64% of the time on this sample fails the rest, and the failures are the large moves. Hold that thought for the mistakes section.

A textbook case: Fed, Sep 16, 2026

The Federal Reserve raised 25 bp to 3.75%–4.00%, as expected. The hour before the 2:00 PM EDT announcement ranged 8.8 pips on EUR/USD. The first spike went above the box by 7.7 pips, a candle closed back inside after 5 minutes, and within four hours price had traded through the low of the box: the other side.

EUR/USD · 1-minute · New York time
EUR/USD one-minute chart around the Federal Reserve decision of Sep 16, 2026 The hour before the 14:00 announcement ranged 8.8 pips. The first spike went above it by 7.7 pips, and a five-minute candle closed back inside after 5 minutes. 1.146 1.148 1.150 1.152 1.154 13:00 13:30 14:00 14:30 15:00 15:30 THE HOUR BEFORE Fed 14:00 PRESS CONF. Spike 1.15469 · 7.7 pips past the hour's high Closed back inside · 5 min

One hour later the pair stood −68 pips from where it was before the announcement. The full decision, minute by minute.

Step 4: the other side

After a close back, the stops on one side are gone and the orders that were waiting on the other side become the next place to go. Often the move that follows runs the other edge of the box. Of the 67 decisions that closed back, price traded through the opposite edge within four hours in 42 (63%).

That is a count of what happened, not a promise. In 25 cases price closed back and then did not reach the other edge in four hours, and a trade taken at the close back would have been sitting on a stop or a slow drift.

When the first move holds: RBNZ, Oct 28, 2026

The read has a clear failure case, and a reader who never sees it will over-trade it. On Oct 28, 2026 the Reserve Bank of New Zealand held at 2.75%. The hour before ranged 11 pips. The first spike went below it by 93 pips and no five-minute candle closed back inside within the hour. An hour after the announcement the pair was −60 pips from where it started.

NZD/USD · 1-minute · Wellington time
NZD/USD one-minute chart around the Reserve Bank of New Zealand decision of Oct 28, 2026 The hour before the 14:00 announcement ranged 11 pips. The first spike went below it by 93 pips, and did not close back inside within the hour. 0.5575 0.5600 0.5625 0.5650 0.5675 13:00 13:30 14:00 14:30 15:00 15:30 THE HOUR BEFORE RBNZ 14:00 Spike 0.55813 · 93 pips past the hour's low

Here the first move was the move. No close back, no signal, no trade under this read. The discipline of waiting for the close is what keeps a trader out of the decision that simply runs.

The press conference: a second move

The statement is read in a second; the press conference takes an hour. The Fed's, for example, starts thirty minutes after the statement. A governor's answer can confirm the decision or soften it, and a phrase in an answer can move price as much as the rate did. When the press conference starts, the one-minute chart often shows a second burst of orders, with its own spike and its own question: has it closed back? The decision pages mark the press conference on every chart.

The result against the expectation

A currency does not rise because a bank hiked. It rises when the bank is more hawkish than the market expected. If the market expected 50 basis points and the bank delivered 25, a hike is a disappointment and price can fall. The same applies to the statement, the vote and, four times a year for the Fed, the dots.

In the 105 measured decisions with a consensus, the result matched it exactly 102 times, so for most decisions the number alone was not the news. When the result did differ and the first spike left the box (3 decisions), that first spike went the way the surprise pointed 3 times and the other way 0 times. Even a clear surprise does not guarantee the first move goes its way, because the first move goes where the stops are.

Next up is the European Central Bank, Thu, Oct 29 at 2:15 PM Frankfurt, with hold at 2.50% expected. Its preview has the standing highs and lows where stops are resting.

The counts for all eight banks

Every decision since Jan 24, 2025 that has one-minute candles, measured the same way. "Closed back" counts decisions, out of all of them, not only those that broke the box.

Central bank Decisions Closed back Then other side Median first move Median hour before
USD
14 10 (71%) 9 (64%) 26 pips 11 pips
14 10 (71%) 6 (43%) 15 pips 15 pips
GBP
14 8 (57%) 5 (36%) 29 pips 18 pips
JPY
14 11 (79%) 6 (43%) 28 pips 29 pips
CAD
15 12 (80%) 6 (40%) 12 pips 21 pips
14 8 (57%) 7 (50%) 20 pips 8.7 pips
13 4 (31%) 2 (15%) 30 pips 11 pips
7 4 (57%) 1 (14%) 17 pips 14 pips
All eight banks 105 67 (64%) 42 (40%) 21 pips 14 pips

Pips are measured on each bank's main pair, so sizes are not comparable between rows without care. Small samples per bank swing a lot: read the first column before the others.

Decision-day checklist

  1. Check the time on the bank's own site, and know which timezone your chart uses. The calendar shows both.
  2. Read the preview: the rate now, what is expected, and the statement's last wording.
  3. An hour before, draw the box: the high and low of the last 60 minutes, on the one-minute chart.
  4. Mark where stops rest: the day's high and low, the previous day's, the week's. The decision pages project these.
  5. Decide before the announcement what you will do if the first spike does not close back. Writing it down is the point.
  6. Decide your size and where you are wrong. A news minute can move farther and faster than your stop.
  7. At the announcement, do nothing. Watch the first spike, and wait for the five-minute close.
  8. If a candle closes back inside the box, the plan is yours to run. If none does, you have no trade and that is a result.
  9. Mark the press conference start. It is a second event with its own rules.
  10. Afterwards, save the chart and write what happened. Your own journal beats any count on this page.

Common mistakes

  • Trading the headline. Buying because the bank hiked, without asking what was expected. The price already held the expectation.
  • Chasing the first spike. It is the part that runs the stops. 37 of 104 first spikes did not close back within the hour, and chasing the others is how a trader becomes the liquidity.
  • Treating a count as a promise. 64% is not a signal on any single decision.
  • Ignoring the spread. Spreads widen sharply at the minute of a decision, and slippage on stops can be several times the box.
  • Using one bank's behaviour for another's. The table above shows the banks do not behave alike.
  • Sizing for the average. The median first move is 21 pips; some are many times that.
  • Trading every decision. With eight banks on the calendar, the best use of this read may be to skip most of the decisions.

An honest word on risk

Everything on this page is a count of what price did after decisions in the past, measured with fixed rules. It is not a forecast, a signal or a promise, and nothing on this site is investment advice. Trading news is among the riskiest ways to trade: prices gap, spreads widen, stops fill far from where they were placed, and leveraged positions can lose more than the money you put in. If you trade, size so that a bad decision day is an annoyance and not a crisis.

Where the full method is taught

This site measures the read. Forexia teaches it: where the stops rest before a release, the hunt, the close back and what to do with it, in courses built on real days. The same site's tools and community are there too.

Questions

Trading central bank decisions: questions answered

How do you trade an interest rate decision?

The read this guide teaches is to wait, not to guess the number. Mark the high and low of the hour before the announcement, let the first spike run past one edge, and only act when a five-minute candle closes back inside that range: the hunt is then done and the move often belongs to the other side. Across 105 decisions the first spike closed back inside within the hour 67 times, and in 42 of those price then took the opposite edge within four hours. It did not close back 37 times, so every plan needs a stop and a size you can afford. Trading news is risky.

How far do prices move after a central bank decision?

The median of the bigger 15-minute move across all 105 measured decisions is 21 pips, against a median hour-before range of 14 pips. The figure differs by bank: Fed 26 pips, ECB 15 pips, BoE 29 pips, BoJ 28 pips, BoC 12 pips, RBA 20 pips, RBNZ 30 pips, SNB 17 pips.

Why does a currency fall after a rate hike?

Because the price already holds what was expected. Only the difference between the result and the expectation is news. 102 of 105 measured decisions came out exactly as the consensus expected; in the rest the result was more hawkish or more dovish than that. A hike that is smaller than hoped, or a statement that hints it is the last, reads as dovish and can sink the currency.

Which central bank decisions are the most volatile?

By median first-15-minute move since Jan 24, 2025, the biggest is the Reserve Bank of New Zealand (30 pips on NZD/USD) and the smallest is the Bank of Canada (12 pips on USD/CAD). Pips are not comparable between pairs without care, and a median hides the outliers.

By Forexia

Learn the read behind every projection.

Forexia teaches the same read this site measures: where the stops rest before a release, the hunt, and the close back that says it is done. Courses, live signals for members, and a community of traders who chart the same way.