Noise Area Breakout Backtest: Gold (XAUUSD), Asia, London and New York

00The verdict in three lines

• The "noise area" breakout comes from a 2024 paper by Zarattini, Aziz and Barbon, tested there on the S&P 500 ETF: for every minute of the session, average how far price usually moves from the open at that time of day over the last 14 days; that range around the open is the noise area. Buy when price closes above it, sell short when it closes below, check every 30 minutes, trail the stop at the band or the VWAP, and close everything at the end of the session. We tested the entry and exit unchanged, one ounce at a time (no leverage, no position sizing), on spot gold (XAUUSD, a broker CFD) in the Asia, London and New York sessions, with the rules, costs, gates and a control fixed in advance.
• None of the six variants passed. On gold every variant ended close to zero after costs, between -$0.34 and +$0.12 an ounce a trade: before costs they made about what the $0.30 round trip costs. The Asia and New York versions ranked above the moved-entry control, because the same entries placed on other days lost more, but beating moved entries is not the same as making money.
• The rule passed on NQ futures in the New York session (read the NQ post); it did not carry over to gold.

01The rule

Noise area: for each checkpoint m minutes after the session open, sigma is the average of |price at m ÷ open - 1| over the previous 14 sessions. The upper band is the higher of today's open and the previous session's close, times (1 + sigma); the lower band is the lower of the two, times (1 - sigma). Checkpoints every 30 minutes from 30 minutes after the open, up to 30 minutes before the close; decisions use the close of the checkpoint minute and orders fill at the open of the next minute. Entry: above the upper band buy, below the lower band sell short (no shorts in the long-only versions). Exit: a long is closed when price closes below the higher of the upper band and the VWAP of the session, a short when it closes above the lower of the lower band and the VWAP; after such an exit the same checkpoint may open the opposite position; everything is closed at the session close. Sessions: Asia (09:00 to 15:00 Tokyo); London (08:00 to 16:30 London); New York (09:30 to 16:00), each tested on its own; New York also follows the exchange's early closes.

Six variants: NA-XAU-AS-LS, NA-XAU-AS-L, NA-XAU-LD-LS, NA-XAU-LD-L, NA-XAU-NY-LS, NA-XAU-NY-L. With the same six on the two other markets, eighteen variants were tried, so the protocol requires a development t-statistic of 2.5.

02Data and costs

XAUUSD, the 1-minute bid history of an IC Markets MetaTrader 4 demo server, June 2004 to 7 October 2026, with the corrected clock of our clock audit (in many weeks of 2004 to 2010 the time stamps were an hour early; for some weeks the correction is an estimate). The VWAP uses tick volume, the number of price changes a minute, because a CFD has no contract volume. Periods: development 2004 to 2012, validation 2013 to 2019, holdout 2020 to 2026. These periods are not new: the same file and years were used by earlier tests, so the holdout is reported as already used. Costs: $0.30 per ounce a round trip; a flip pays two. Engine check: na_selftest.js checked the Asia and London session times against the time zone database on every weekday from 2004 to 2026, hand-built sessions (sigma, both bands, a long, a short, the stop, a flip, the session close, an early close), and all eighteen variants trade by trade against an independent re-implementation (na_selftest_result.txt).

03Results

Results in dollars per ounce. The 95% range is the Wilson interval for the win rate. The drawdown is on closed trades; every position is closed by the end of its session.

Cumulative net result by month, long-and-short versions. * Holdout already used before this test.

By period

04New York by year

05Longs, shorts and the best trades

06Is it the breakout or the market?

The benchmark is a moved-entry control: each session's first entry (its checkpoint and direction) is moved, with all sessions together, by one random offset of at least 5% of the sessions; on the new day the position opens at the same checkpoint in the same direction and is then managed by the same exit rule and costs. 500 runs with fixed seeds; the bar is the 95th percentile of the average net result per trade. It asks whether leaving the noise area picks better days than the same entries placed on other days.

07The gates

08Verdict by variant

Not tested: other markets, position sizing and leverage, other checkpoint intervals, live fills.

TWhat this shows, and what to take into your own tests

What the results show

• None of the six variants passed: every one ended close to zero after costs, between -$0.34 and +$0.12 an ounce a trade.
• Asia and New York ranked above the moved-entry control, but only because the same entries placed on other days lost more.
• The rule that passed on NQ in New York did not carry over to gold.

What remains unknown

• Other markets and the paper's own market, the S&P 500 ETF.
• Position sizing and leverage, which we did not use.
• Real contract volume for the VWAP: gold used tick volume.

Why it failed

Gold's breakouts did not run far enough past the noise area to pay for the false ones and the costs: before costs the variants made between -$0.04 and +$0.42 an ounce a trade, about the $0.30 a round trip costs, so nothing was left after costs.

What you can apply to your own tests

• Test each session on its own: the same rule can earn in one and lose in another.
• Beating a random control is not enough; the result has to clear its costs too.
• Try a promising rule on a second market before trusting it.

A note on the numbers: drawdown figures are measured on closed trades unless a table says it includes open losses. Open losses inside a trade are not counted in a closed-trade drawdown, so real account risk is higher, especially for systems with no stop. Backtests are not live trading.

FAQCommon questions

Does the noise area breakout work on gold (XAUUSD)?

Not in our test. In the New York session the long-and-short version ended at -$78 per ounce after costs and failed the protocol's gates; the Asia and London sessions did not do better.

What is the noise area?

For every time of day, the average distance price has moved from the session open over the last 14 sessions. Around today's open (or yesterday's close, whichever is further) that distance draws an upper and a lower band; a breakout beyond a band is the entry signal, as published by Zarattini, Aziz and Barbon in 2024.

How was the noise area breakout tested?

On gold (XAUUSD) 1-minute data in three sessions, one ounce per trade, with rules written and dated before the data was run, costs included on every trade, results split into development, validation and holdout periods, a random-entry control as the benchmark, and every variant reported. Read the protocol.

09Check our arithmetic

Code provenance: our own code, written from the rule as published in the paper. No third-party source code is used or published.

10Data notes

Holes. Removing the sessions that touch a known hole in the data changes no verdict:

Clock. This test depends on the time of day, so it ran on the corrected clock. With the time stamps as the file has them, the numbers move a little and no verdict changes:

11Corrections

18 Oct 2026, navigation: added links to the new guide that compares all 11 gold strategy tests. No result changed. Before publication, one bug was fixed after the first run: the engine stored the session's open price under the same name as its open time, so the check for holes in the data marked almost every session. Trades, bands and gates were not affected (the independent re-implementation matched trade by trade); the field was split, the self-test passed again, and every market was run again. See the run log in the specification. 22 Oct 2026, navigation: added a link to the new guide comparing trading costs across all our tests. No result changed. 22 Oct 2026, navigation: added a link to the new cross-market scoreboard, which includes this test. No result changed. 22 Oct 2026, navigation: added a link to the new guide comparing trend following and mean reversion across all our tests. No result changed.

+More tests and guides

• Best gold trading strategy
• Noise area breakout nq
• Noise area breakout es
• Gold opening range breakout backtest
• Turtle trading strategy backtest
• Trend following vs mean reversion: this test next to the others of its style
• Free prop firm challenge calculator
• What all the backtests found so far
• The backtesting protocol
• How to backtest a trading strategy without fooling yourself
• Backtesting glossary