Does London Sweep the Asian Range? Measured

The short answer

The Asian range is the high and low of 20:00 to 00:00 New York, the range most ICT and session traders mark before London. We measured what happened to it on every weekday for gold, silver, oil, the Nasdaq, the Dow, the S&P 500, the DAX, four forex pairs, Bitcoin and Ethereum: 13,456 days, 1 Jul 2022 – 18 Sept 2026.

Whether it breaks is not the question: one side went by 17:00 on 99.5% of days. It usually went before London had even opened: by 02:00 New York on 62% of days, and by the end of the London kill zone at 05:00 on 92%. The high went first 51% of the time and the low 48%, so the first break tells you nothing about direction on its own.

What happens next is the useful part. Once one side broke, price took the other side too before 17:00 on 57% of days. But the day closed past the side that broke first more often (51%) than it reversed all the way through the other side (27%). And the textbook Asian range sweep trade, taken 12,221 times, averaged −0.04R a trade before costs. The range is a map of where liquidity sits, not a signal.

When the Asian range breaks

The range closes at midnight New York, and price is usually still near one of its edges. So the first break tends to come quickly: 40% of days had broken it within the first hour, and 62% broke it before 02:00. London took the first side on only 33% of days, and the New York session on 4.8%.

The split in the last column is the one to know for your market. The indices took both sides on 70–76% of days, because their Asian range is only 17–21% of the whole day's range: the New York cash open travels through it easily. On USD/JPY and AUD/USD Asia is a working session, the range is 38–42% of the day, and both sides went on only 41–43%. Bitcoin and Ethereum broke it later than anything else (79% by 05:00), and gold sat in between at 47%.

Which side goes first, and what comes next

Across every market the first break was close to a coin flip: the high 51.3%, the low 48.1%. Direction has to come from somewhere else, such as the higher timeframe or where the day opened in yesterday's range, and that belongs in your rules before you look at a chart.

The first break usually has room to run. Price travelled at least half a range past the first side on 79% of days and a full range on 61%. Then, on 57% of days, it came back and took the other side too. Where the day finally closed, at 17:00 New York:

Read the last two columns together. On every market the day closed past the side that broke first about half the time, and through the other side far less often. Taking the other side and closing through it are different things: on the Nasdaq price took both sides on 70% of days but closed through the second one on 34%.

The London sweep, and the Judas swing

The classic story is that London runs one side of the Asian range to trap early traders, then reverses through the other side for the real move of the day: the Judas swing. London took the first side on 32% of days. On those days the other side went during the New York session on 34%, and the day closed through the other side on 25%. About one day in four.

That is a real pattern and a minority one. The indices showed it most (31–33% of their London-first days) and crypto least (16–18%). If your plan needs the London move to be fake, it needs a filter that finds those days, and the filter is what has to be tested.

The Asian range sweep trade

We fixed the trade before looking at any result, in the form it is usually taught. After one side of the range breaks, wait for the first 15-minute candle to close back inside the range. Enter at that close, as long as it comes before 12:00 New York and before the other side has been taken. The stop goes at the furthest price the sweep reached; the target is the far side of the range, or its 50%. Anything still open at 17:00 is closed there. The setup triggered on 91% of days.

The far-side trade won 24.8% of the time. That looks poor until you see the payoff: the stop sits just past a sweep that is often small, so a win paid a median 3.5 : 1. The two cancel. The result was −0.04R a trade before costs, and a price that moved at random would have scored zero. Targeting the 50% won 40% of the time at 1.6 : 1, and came to −0.04R. A nearer target buys a higher win rate with a smaller payoff, and nothing else.

None of the usual filters moved it clearly off zero. Sweeps in London (−0.01R) did no better than sweeps before it (−0.07R). Selling a swept high did worse than buying a swept low (−0.09R against 0.00R), which is mostly four years of rising gold, indices and Bitcoin. The ± column is the uncertainty, two standard errors: a figure inside it cannot be told apart from zero.

Every market

The same far-side sweep trade, market by market.

No market showed an expectancy clearly above zero. S&P 500 came out clearly below it. Notice the width of the uncertainty: about ±0.15R on gold from 982 trades. If four years of every day cannot pin the result down tighter than that, a month of screenshots cannot either. The sample size calculator shows how many trades you need.

What this means if you trade the Asian range

• Mark it, and expect it to break. One side goes almost every day, usually before London. The break is information about where price has been, not about where it goes.
• Know your market's personality. On the indices, plan for both sides to go (70–76% of days). On USD/JPY and AUD/USD, the range is a larger share of the day and holds more often.
• Don't assume the first move is fake. The day closed past the first side about twice as often as it reversed through the other. A Judas-swing plan needs a filter that finds its days.
• Test your version, not the textbook one. The bare sweep entry is worth about zero. The edge, if there is one, lives in what you add: a higher-timeframe level at the sweep, a shift in structure on a lower timeframe, a time window, a fair value gap entry. Measure each against this baseline.

That is what the Asian session backtester is for: the range is shaded on the chart as it forms, and you replay each London open one candle at a time with the future hidden, apply your own rules and log the results. The CRT statistics page runs the same test on the candle-range version of the idea.

How we measured it

CRTLAB's own 15-minute price history, 1 Jul 2022 – 18 Sept 2026, weekdays only, for every market except the DAX (from 11 Mar 2025) and crypto (to 21 Sept 2026). The Asian range is the highest high and lowest low of the 15-minute bars from 20:00 to 00:00 New York, the window the CRTLAB chart shades and the kill zone reference defines. The day after it runs from 00:00 to 17:00, the close of the New York trading day. A side counts as taken when a 15-minute bar trades beyond it; the rare bar that took both sides at once (0.1% of days) counts as both. The range sizes are the last year only, because gold and the indices have risen so far since 2022 that a four-year average in dollars or points would describe neither end of it; every percentage uses all four years.

The trade is walked forward on 15-minute bars. A bar that touched both the stop and the target is scored as a loss, because we cannot see inside it, which is the cautious choice. No spread or commission is included: this measures the market's behaviour, not a broker. The filters were chosen before the results were seen; the small or large range compares each night with the median of the 20 before it.

Frequently asked questions