ICT killzones: times, Silver Bullet and liquidity sweeps
Killzones are the time windows where ICT looks for entries. Get the times in New York time, the rules of each model and what is actually proven.
Killzones are the time windows where ICT (Inner Circle Trader, Michael J. Huddleston) places the moves with the most intent: mainly the London open and the New York open. They are defined in New York time. The NY AM killzone usually runs from 08:30 to 11:00 and London from 02:00 to 05:00.
The killzone is not a signal. It is the "when". The "where" comes from liquidity levels (yesterday's high and low, Asia's, London's) and the trigger is a sweep of that level followed by displacement. Silver Bullet, Judas swing, Power of 3 and the PDH/PDL sweep share that same skeleton; what changes is the window.
A methodological warning before we start: ICT does not publish closed written rules. What you will see here is the most-cited version of each model, gathered from secondary sources, not the canonical one. And none of these models has audited statistics.
Free tool: if you use NinjaTrader 8, you can download the killzones and levels indicator we built at El Trader Financiado for the community. See the indicator and download it.
What ICT killzones are
They are fixed windows of the trading day. The hours change depending on the source and the year of the course, so you will see variants. This is a common table:
| Killzone | New York time | Chicago time (CT) | Variants in circulation |
|---|---|---|---|
| Asia | 20:00–00:00 | 19:00–23:00 | 18–20, 19–23 NY |
| London | 02:00–05:00 | 01:00–04:00 | — |
| NY AM | 08:30–11:00 | 07:30–10:00 | 07–10, 08–11 NY |
| NY PM | 13:30–16:00 | 12:30–15:00 | — |
Within that frame there are shorter windows:
- Silver Bullet: three one-hour windows, 03:00 to 04:00, 10:00 to 11:00 and 14:00 to 15:00 New York time.
- Macros: roughly 20-minute stretches, such as 09:50–10:10 and 10:50–11:10 New York time. No market documents that schedule and the lists vary between sources, so take them as a convention, not a specification.
And some reference prices that go along with the windows:
- Midnight open in New York (00:00): the start of the "real day" in ICT.
- 08:30 open: the time of US macro data.
- Cash open at 09:30.
These three opens are drawn as horizontal lines and are used for bias. On a day expected to be bullish, the model wants to see the session low below the midnight open or the 08:30 open before price spends the rest of the day above it. Even the sources that explain it call it "an expectation, not a certainty".
Liquidity sweep versus breakout
Above a relevant high, sellers' stops and breakout buyers' orders pile up. That is why those prices are called liquidity pools.
- Sweep: the wick exceeds the level and the candle closes back inside.
- Breakout: the candle closes outside and price keeps going.
The difference is objective and can be marked without opinion. It is the first thing to pin down.
How to mark levels and trade a killzone step by step
1. Before the session, mark the "draw on liquidity"
- PDH/PDL: the previous day's high and low.
- Asia and London extremes.
- ONH/ONL: the overnight high and low.
- Equal highs or lows, which concentrate stops.
- Unfilled fair value gaps (we explain them in the fair value gap guide).
- The opens at 00:00, 08:30 and 09:30 New York time.
2. Wait for price to interact with a level inside the window
Outside the window there is no setup. Inside it, it only counts if price reaches one of the marked levels.
3. Require confirmation
- Sweep: wick outside the level and close back inside.
- Displacement: an energetic leg in the opposite direction that closes through the last swing (a change of structure; we detail it in order block and market structure).
- Entry zone: the FVG that displacement leaves behind.
The sources that describe the Judas swing say it clearly: confirmation comes from the failure and the displacement, not from the clock. Many sessions open and run in one direction with no false move at all.
4. If the window closes with no signal, there is no trade
It is the most useful rule in the whole method and the hardest one to follow.
Killzone strategies: rule-based systems
The common skeleton
| Step | What you do |
|---|---|
| 1. Level | Mark a liquidity pool before the window |
| 2. Window | Wait for price to arrive inside the killzone |
| 3. Sweep | Wick outside and close inside |
| 4. Displacement | Strong leg that breaks the last opposite swing |
| 5. Entry | Retracement to that displacement's FVG |
| 6. Stop | Beyond the swept extreme |
| 7. Target | The liquidity on the opposite side |
The models, one by one
| Model | Window (NY time) | Trigger | Stop | Target | What the source itself says |
|---|---|---|---|---|---|
| Silver Bullet | 03–04, 10–11, 14–15 NY | Sweep or change of structure and FVG formed inside the window; no FVG, no trade | Beyond the manipulation swing or the far edge of the FVG | The opposite pool, with partials at intermediate structure | No audited public statistics exist |
| Judas swing | London start (≈02–05 NY) and the NY open | False push through a marked level, rejection and displacement back | Beyond the swept extreme | Opposite-side liquidity | "An expectation, not a law" |
| Power of 3 (AMD) | The whole day | Reads the day as accumulation (Asia), manipulation (the Judas) and distribution | Same as the Judas | Same as the Judas | Reading framework, traces back to Wyckoff |
| PDH/PDL sweep | NY AM | The same skeleton with yesterday's high or low as the pool | Beyond the wick | Opposite pool | No study. It is the Turtle Soup idea applied to daily levels |
| Asia sweep in London | London | Sweep of one extreme of the Asia range and close inside | Beyond the sweep | Opposite side of the range | No study |
The Silver Bullet has a practical advantage that doesn't depend on it working: one window, one setup or nothing. That limits the number of trades.
If you are interested in the sweep over the range of a higher-timeframe candle, that is CRT, which we cover in CRT trading.
The mechanical alternative: the ORB
The opening range breakout is the only system in this family with 100% mechanical rules and papers behind it. The most-cited version (Zarattini and Aziz, on QQQ) is this:
| Rule | Detail |
|---|---|
| Range | First 5-minute candle after the cash open (09:30 NY) |
| Entry | Bullish candle: buy at the open of the second candle. Bearish: sell. Doji: no trade |
| Stop | Opposite extreme of the first candle |
| Target | 10 times the risk (10R) or exit at the close of the day |
| Management | One trade per day |
You will see it in the evidence: the result depends almost entirely on costs.
Common killzone mistakes
- Taking the clock for granted. There are several versions of the hours. Pick one table, write it down and don't change it.
- Using a fixed offset from your local time. If you trade from outside the US, your clock and New York's may not change on the same date. Europe and the US, for example, change clocks on different dates, so for a few weeks in March and in October–November the gap with New York shrinks by one hour. Always convert from New York time.
- Mixing up New York and Chicago. ICT defines everything in New York time. Chicago time (CME's) runs one hour behind.
- Ignoring CME holidays. Early-close days distort the NY PM killzone and the next day's PDH/PDL. Check the official CME calendar.
- Calculating the PDH/PDL without knowing which session. With the regular session or the full session you get different levels. The overnight high and low only make sense with the full session.
- Trading every sweep. Some indicators publish high percentages of "returning to sweep" a level. Those are hit rates: they say price gets there, not that you make money trading it.
- Confusing volatility with direction. A lot of movement in a window doesn't mean the direction is predictable.
What to combine it with
Killzones and levels are the base of the rest of the tools: they give the "where" and the "when". The trigger comes from another piece:
- Fair value gap: the mandatory entry zone in the Silver Bullet.
- Order block and market structure: the change of structure that confirms displacement.
- SMT divergence: NQ makes a new extreme and ES doesn't; used as confirmation of the sweep.
- CRT trading: the sweep over the range of a higher-timeframe candle.
- VWAP and volume profile: references independent of ICT. In ES and NQ, the NY AM killzone covers the same window as the Initial Balance.
A combination with objective rules would be: a sweep of the PDL or the overnight low inside NY AM, a close back inside, a reclaim of VWAP or yesterday's value area, entry at the FVG and stop beyond the wick. It is explainable, but its edge is not measured. And each confluence you add reduces the number of trades and raises the risk of fitting the rules to the past.
What the evidence says
Let's separate what the method says from what is proven:
- "Things happening" in killzones has an explanation without ICT. Intraday volatility concentrates at the open and the close, and a good part of the time-of-day pattern is due to scheduled announcements (Andersen and Bollerslev, 1998; NBER). 08:30 and 10:00 New York time are macro data times. More volatility does not imply that direction is predictable.
- ICT models have no audited statistics. Not even the sources that document these models the most provide audited public statistics for the Silver Bullet.
- The ORB depends on costs. The paper reports a 24% win rate and +0.13 R per trade before spread or slippage. An independent replication with costs, on Nasdaq CFDs, leaves +0.002 R per trade, practically zero.
How killzones fit a prop firm
News. The NY AM killzone (08:30–11:00) contains CPI and NFP (08:30), the 10:00 data releases and, for CL, the Wednesday EIA inventory at 10:30. The Silver Bullet's 10:00 to 11:00 window starts right at the 10:00 release. Many prop firms restrict trading around news such as CPI or NFP, especially on funded accounts; check your firm's rules before trading that window.
Daily loss. Trading every sweep in all four killzones multiplies the trades and the probability of hitting the daily limit. A simple rule: one setup per window and an internal daily loss cap below the firm's.
Stop size. The stop goes beyond the swept extreme, so it changes every day. Calculate it before entering:
| Stop beyond the wick | NQ ($20/point) | MNQ ($2/point) | ES ($50/point) | MES ($5/point) |
|---|---|---|---|---|
| 5 points | $100 | $10 | $250 | $25 |
| 15 points | $300 | $30 | $750 | $75 |
| 30 points | $600 | $60 | $1,500 | $150 |
If the sweep leaves a long wick, drop down to micros instead of trimming the stop.
Consistency. Systems with a distant target (the opposite liquidity, or the ORB's 10R) and few wins string together losing streaks and concentrate profit in a few big days. That clashes with the trailing drawdown and consistency rules. Check how much a single day can weigh at your firm.
Compare news, daily loss and consistency rules in the comparator.
Free NinjaTrader killzones indicator
At El Trader Financiado we built a free indicator for NinjaTrader 8 that draws the killzones in New York time (and converts to your time zone, handling daylight saving changes), the PDH/PDL, the week and the overnight, the opens and a mark on every sweep. We made it for the community and it is available in English and Spanish. Download and settings on its page: session levels indicator for NinjaTrader.
Frequently asked questions
Do killzone times change with daylight saving time?
No. They are fixed in New York time, and Chicago (CT) is always one hour behind New York. If you trade from Europe, the gap with New York shrinks by one hour for a few weeks in March and in October–November, because the US and Europe change clocks on different dates.
Are killzones in New York time or Chicago time?
In New York time, which is how ICT defines them. Chicago time, the one CME uses, runs one hour behind.
What is the midnight open in ICT?
The 00:00 New York price. ICT uses it as the start of the day and as a bias reference: on a bullish day it expects the low to form below that open.
What is the difference between a sweep and a breakout?
In a sweep, the wick exceeds the level and the candle closes back inside. In a breakout, the candle closes outside and price continues.
Does the Silver Bullet work in NQ?
There are no audited public statistics proving it. What it does bring is discipline: one window, one setup or nothing.
Can I trade the NY killzone on CPI days with a prop firm?
It depends on your firm and on whether you are in an evaluation or funded. Many restrict trading for a few minutes before and after news such as CPI or NFP. Check their rules before the session.
