ICT Kill Zones on MT4 and MT5: What the Sessions Show You

If you have seen the words ICT kill zones on a chart or in a forum, you have probably wondered whether a time window can make a difference to your trades. The question is fair. Many traders feel that the same setup works better at one hour than another, and they want a tool that marks the hours for them.

The risk is that the tool becomes the strategy. A coloured box on the chart feels like an edge, and a trader who relies on it may stop checking the entry, the spread and the stop. The box then marks a time when the trader was already losing money on a rule nobody tested.

This guide explains what the kill-zone idea says, the session times most often quoted, how a session tool handles the broker clock, and how to test a rule inside the windows before you risk capital. It does not promise a result, and it treats the method as a claim to check.

What Are ICT Kill Zones?

ICT stands for Inner Circle Trader, the name used by Michael J. Huddleston for his trading education. The method describes the market as moved by large institutions, and it argues that these players are most active in specific time windows. Those windows are called kill zones.

The idea is a filter. It says that price is more likely to move sharply, sweep a recent high or low, or change direction during certain hours. A trader who follows the method looks for signs of a sweep or a structure shift inside the window, and ignores the rest of the day.

The method is a framework of opinion. Its claims about institutions and liquidity are hard to measure directly from a retail chart, and independent testing of the full framework is limited. Treat it as a hypothesis about timing, and test it with the same care you would give any other rule.

The Commonly Cited Session Times

Most ICT material uses New York time as the reference clock. The windows below are the ones most often quoted. Check them against the method you are studying, because some sources use slightly different hours.

The London kill zone is usually given as roughly 2:00 AM to 5:00 AM New York time. Followers of the method say this window often shows early liquidity moves as European trading starts.

The New York kill zone is usually given as roughly 7:00 AM to 10:00 AM New York time. It overlaps with the late London session, and it includes the opening range of the US market and the release of many economic reports.

The Asian session is usually described as the quieter window before London. Followers use its high and low as reference levels, because liquidity often sits just above and below them. The Asian range is a useful place to start, because it is set before the busier sessions begin.

These windows are a map, not a promise. A session that looks active on one day can be dull on the next, and a quiet day can end with a sharp move after the window closes.

Why Your Broker Clock Changes the Result

The most common error with session tools is the clock. Your MT4 or MT5 chart shows the broker server time, which is set by the broker and is often several hours ahead of or behind New York. If a tool reads the chart clock without conversion, every box lands in the wrong place.

A proper session tool converts from server time to New York time, using the broker offset. Some tools detect the offset from the chart, and others ask you to enter it. Either way, you must confirm the offset against a known event, such as the daily candle open, before you use the boxes.

Daylight saving adds a second layer. New York and the broker may change their clocks on different dates, so the gap between them moves for a few weeks each year. A tool that uses a fixed offset will mark the windows an hour off during that period. Check the chart in March and November, when the change happens.

Write the offset down. A line in your notes such as broker server is New York plus seven hours in summer tells you what the tool should show, and it lets you spot a drift before it costs you a trade.

How a Session Tool Draws the Windows

Most session tools do the same few jobs. They read the time on each bar, decide which session the bar belongs to, and draw a box or a shaded area over that session. Many also draw the session high and low as horizontal lines, and some add alerts when price leaves a range.

The box helps you see the window, but the useful part is the high and low it records. A session range gives you two levels that other traders may watch too. A sweep of the Asian high during the London window is a common example that followers describe. Whether that sweep leads to a reversal is a question for your own test.

Check what a tool does not draw. A session box does not tell you the trend, the order flow or the risk on a trade. Combine it with your own rules, and do not let the colours decide the trade.

Some tools include a time filter that blocks an EA from trading outside the windows. This is a different use from a visual box. A filter changes the trade count, and you must test it on the same data with and without the filter, or you will not know whether it helped.

Settings to Review Before You Trade

A session tool has settings that change what you see. Review each one before you rely on the chart.

The session start and end times set the window. Check that they match the method you are testing, and not a set of times copied from a different course.

The time zone or offset setting converts the chart clock. Confirm it on a known bar, as described above.

The colour and transparency change only how the boxes look. They do not change the result, although a bright box can make a session feel more important than it is.

The alert settings can send a notification when price enters or leaves a window. Alerts are useful for watching a session without a screen, but they can fire often, so set a sensible limit.

The high and low line settings decide which levels you see. Some tools draw the previous session’s levels and others draw only the current one. Make sure the level you plan to trade is the one drawn.

Testing a Kill-Zone Rule Without Overfitting

A rule that uses kill zones must be tested the same way as any other rule. The risk is overfitting, which means building a rule that fits the past by chance and fails on new data.

Start by writing the rule before you look at results. For example, a rule might say: during the London window, enter only after a sweep of the Asian high, with a stop above the sweep and a fixed target. A written rule is harder to change after a loss.

Next, choose a test period and do not change it. Split the history into two parts. Build the rule on the first part, then test it on the second part without any change. If the result collapses on the second part, the rule was probably fitted to the first.

Record the spread. A session window is often busy, and spread can widen sharply around news. A rule that wins by a small margin can lose once the spread is counted at the real time of entry.

Count the trades. A rule that takes four trades in a year has not been tested. Aim for a sample large enough that a few trades cannot decide the result, and note the sample size next to every figure you report.

Finally, test the rule on a demo account before any live money. A demo run shows how the rule behaves with your broker’s fills, and it exposes timing errors that a chart review can miss.

Combining Kill Zones With Structure and Risk

A kill zone is a time filter, and a time filter needs something to filter. Followers of the method usually pair the windows with structure, such as a break of structure, a fair value gap or an order block. Each of these concepts needs its own definition and its own test.

Start with fair value gaps, order blocks and break of structure. Read each one before you combine it with the windows, because a rule built on a concept you do not understand will break in the first unusual market.

Risk comes first, whatever the window. A stop that is too tight for the session’s volatility will be hit by normal noise. A stop that is too wide will cost more than you planned. Measure volatility before you set the distance, using a method such as the ATR multiplier for stop loss, and size the position so that one loss stays small. For a separate trend filter you can compare, the Supertrend indicator for MT5 shows a different way to read direction.

The news calendar is a separate filter. Many kill-zone windows overlap with scheduled releases, and a release can move price far beyond any normal range. A plan for news, such as pausing the EA or closing positions before the release, protects the account on the days the chart is least predictable. A guide to disabling an EA during news events lists the practical ways to do that.

What the Kill Zone Cannot Tell You

A session box shows time, not outcome. It cannot tell you whether the next London window will be active, whether a sweep will reverse, or whether your entry is better than the last one. Keep those limits in mind whenever the chart looks convincing.

The method also describes institutions that you cannot see. Their orders are not recorded on a retail chart in the form the method suggests, and the framework is hard to prove or disprove from candles alone. A rule built on it can look right in hindsight and still have no edge.

Past sessions do not set future ones. Market structure changes with the news cycle, the rates outlook and the liquidity of each market. A window that worked well in one year can work poorly in the next.

A Practical Routine for the Windows

A routine keeps the method from taking over your day. Use the same steps each time you trade a window.

Before the window opens, check the time zone on the chart against the broker clock, and confirm the offset for daylight saving. Look at the news calendar for releases inside the window.

When the window opens, note the session high and low, and watch for a sweep. Do not enter because the box is coloured. Enter only if your written rule is met.

When the window closes, record what happened. Log whether the rule triggered, the entry, the stop, the exit and the spread. A log that you review each week is the only way to know whether the rule is working.

After a month, review the log as a whole. Count the trades, the win rate, the average result and the largest drawdown. Decide whether the rule earns more study or should be dropped.

When to Walk Away From a Session Strategy

Walk away if the rule works only in a few cherry-picked weeks. A session strategy that depends on a single trending month is not a robust rule.

Walk away if the spread or slippage removes the edge. A rule that makes money on the chart and loses money after costs has no edge in the market you trade.

Walk away if you feel pressure to trade every window. The method does not require a trade in every session, and forcing trades is a common path to losses.

Walk away if the tool you use does not show its time conversion clearly. A tool that hides the clock is a tool you cannot audit, and an audit is the first thing a rule needs.

Reading a Session Chart Step by Step

A session chart is only useful if you read it the same way every time. This section walks through one reading from the open of the window to the close, using the steps most followers describe. The steps are a process for looking, not a promise that the picture will repeat.

Start before the window opens. Note the previous session’s high and low, and the range of the Asian session if you use it. Write both levels on the chart or in your notes. These are the reference points the rest of the reading depends on, so a level you did not record is a level you will argue with later.

Next, watch the first minutes of the window. The question is whether price moves through one of your reference levels and then returns, or whether it accepts beyond the level and keeps going. A sweep and a return is one pattern. A clean break with a pullback that holds is another. Your written rule should say which of the two counts as a trigger.

Then look for confirmation on a lower timeframe. Many followers drop from the hourly chart to a five- or fifteen-minute chart to time an entry. Confirmation should be a visible change in structure, such as a higher low after a sweep of a low. A candle that simply looks large is not confirmation.

Finally, check the context before you act. Is there a news release inside the window? Is the broader trend up, down or flat on the higher timeframe? A sweep against a strong trend has a different meaning from a sweep in a flat range. Record the context with the trade, so the log can show which conditions helped and which did not.

Each step adds a decision point, and each decision point can be wrong. Keep the number of conditions small enough that you can test them. A rule with six conditions is harder to test than a rule with two, and it is far more likely to be fitted to the past.

How to Read a Sweep Without Chasing It

A sweep is a move through a level that is then rejected. Followers describe it as liquidity being taken, and they read the rejection as a sign that the move was engineered. Whether that interpretation is correct cannot be seen on the chart, so focus on what can be seen.

What you can see is the wick and the close. A long wick through a level, followed by a close back inside the range, is the visible form of a sweep. The stronger the rejection, the more the wick tends to matter in the method’s logic, but strength is hard to define, so set a fixed measure in your rule, such as a wick of at least a set number of points.

The common mistake is chasing. A trader who sees a sweep and enters at market, without waiting for a confirmation, usually enters at the worst price of the move. The stop then sits far away, and the risk per trade grows. A written rule that requires a confirmation candle protects the entry and keeps the stop distance sensible.

Another mistake is treating every level as equal. A sweep of the Asian high during the London window is one case. A sweep of a minor intraday high in a quiet hour is another. The method does not rank them for you, so decide which levels your rule uses and leave the rest out of the test.

Why Sessions Differ From Day to Day

A session window is not the same on every day. Some days the London open brings a large move, and other days it drifts for an hour and then stalls. The variation is normal, and it is one of the reasons a session rule can look good on a few days and poor over a year.

Several factors change the character of a window. Scheduled releases move price sharply, and a release can make an otherwise quiet window busy. Holidays thin out liquidity in one market and not another, so a session that is active in London may be dull in New York on the same day. Month-end and quarter-end flows can push price in one direction for a short period.

Volatility also varies. A window in a high-volatility regime has larger moves, so a fixed stop may be too tight. A window in a low-volatility regime may never reach a fixed target. Measuring the average range of the session before you set a stop or a target helps the rule adapt, and it stops one fixed number from failing in every regime.

For these reasons, test the rule across several regimes, not only a calm month. A rule that works in one regime and fails in another has a condition missing, and the test will show which condition it is if you look at the losing trades by regime.

A Worked Example of a Kill-Zone Test

Imagine you want to test a simple London rule. The rule says: mark the Asian high and low before the London window; during the window, enter short after price sweeps the Asian high and closes back below it on the five-minute chart; place the stop just above the sweep; close the trade at a fixed multiple of the stop distance. Long trades follow the mirror rule at the Asian low.

Write the rule down and keep it unchanged. Choose a test period of, for example, twelve months that ends before you began designing the rule. Use the same data source for every trade, and the same spread assumption.

Run the rule on the whole period. Record each signal, the entry, the stop, the exit and the result. Count the signals. If the rule triggers only a handful of times in the year, the sample is too small to judge, and you should widen the period or drop the rule.

Split the results. Look at the long and short trades separately, at the results by month, and at the results on days with a news release compared with days without one. A rule that depends on one month, one direction or one news pattern has a weakness that a single total will hide.

Compare with a baseline. Run the same stop and target rule without the session condition, on the same period. If the session version does not beat the baseline after costs, the session filter adds nothing, and the honest conclusion is that the window is not the source of any edge.

Finally, repeat the test on a second period you did not use. A rule that survives the second period has earned more attention. A rule that fails on it should be dropped or rebuilt, and the failure is a result you should record, not a detail to hide.

Mistakes That Make a Session Rule Look Better Than It Is

Several errors make a session rule look better in a test than it will in live trading. Watch for each one.

Hindsight levels. A test that marks the high or low of a session using the full session data will know the high before it happened. The level must be known at the time of the trade, which means marking it from the data available at the moment of the entry.

Ignoring spread on the window. The spread often widens at the open of a session and around releases. A test that uses a fixed spread through the day overstates the result for the entries that happen when the spread is widest.

Optimising the times. Changing the window start by an hour until the result looks good is a form of fitting. Fix the times before the test, and change them only with a new test period and a stated reason.

Counting the wrong trades. A test that takes the first signal of the window and ignores the rest may be hiding a second entry that lost money. Define the rule so that each window produces at most one trade, or count every entry.

Stopping at the first good month. A rule that looks strong after one month can fall apart in the next. Keep testing until the sample is large enough to show the rule across more than one market condition.

Building a Written Rule Card

A rule card is a one-page record of the rule. It makes the test repeatable and gives you a fixed reference when a losing streak tempts you to change the plan.

On the card, write the session window and the time zone, with the conversion you use. Write the reference levels, how you mark them and when. Write the trigger, the confirmation, the stop rule and the target rule, each as a single sentence with a number. Write the position size rule and the maximum loss per trade.

Add the conditions that stop trading. These might include a maximum number of trades per day, a pause after a set loss, or no trades on a release day. A stop condition written in advance is easier to follow than one invented during a bad session.

Keep the card with the test log. Each time you change a rule, write a new version of the card with the date and the reason. A trail of versions shows whether you are refining the rule or chasing the last loss.

Review the card every month. Check each condition against the log. If you cannot find a trade in the log that followed a condition, the condition is either never triggered or not being followed, and either finding is worth knowing.

Keeping the Method in Proportion

Session timing can be one useful input among several. It is not a substitute for a tested rule, a sound risk plan or a clear view of your own mistakes.

Many traders who study kill zones spend more time reading about the method than testing it. The reading feels like progress, and it can delay the one step that gives real information, which is a test on a fixed period with recorded costs.

A session tool that draws boxes can also make a chart feel more informative than it is. The chart then tells you a story about the market, and the story can persuade you to trade when the rule does not say to. Keep the rule card in front of you, and let it decide.

If a rule survives a fair test, trade it small and keep the log. If it does not, drop it without regret. Both outcomes teach you something real, and a clear answer is worth more than a rule you cannot stop second-guessing.

Next Steps: Test the Idea on a Demo Account

Start with the timing question. Load the session tool on a demo chart, confirm the conversion against a known bar, and check the windows for one full week. Write down what you see, not what you expect.

Then write one rule that uses a window, and test it on a fixed period you did not use to design it. Use the sizing and stop approach from the guides linked above, so the test uses the same risk logic as your live account would.

If the rule holds on demo over a meaningful sample, consider a small live position before you scale. Keep the log, and set a limit on loss before you start.

Trading carries a real risk of loss. Past results do not guarantee future performance, and no session window removes that risk. Start small, test on demo, and only commit capital you can afford to lose.