Noise Area Breakout Backtest: NQ Futures, 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 contract at a time (no leverage, no position sizing), on E-mini Nasdaq-100 futures (NQ) in the Asia, London and New York sessions, with the rules, costs, gates and a control fixed in advance.
• The New York session passed every gate, long and short (+$231,557 over 1,504 trades, +$153.96 a trade after costs) and long only (+$152,227 over 747): development t-statistics of 2.82 and 2.98 against the 2.5 needed, positive in every period and every calendar year from 2020 to 2026, on both the long and the short side, and ranked at the 100th and 99.6th percentile of the control (95 needed). They are the first variants on this site to pass every gate and keep that verdict after the check for holes in the data.
• What that does not mean: the holdout had been used by earlier tests on this NQ file, eighteen variants were tried, and the same rule on ES did not pass (positive, but a development t-statistic of 1.91 and a losing holdout). A backtest that passes is a candidate. The only test that counts now is new data, with the rule unchanged; the Asia and London sessions did not show an edge.

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-NQ-AS-LS, NA-NQ-AS-L, NA-NQ-LD-LS, NA-NQ-LD-L, NA-NQ-NY-LS, NA-NQ-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

E-mini Nasdaq-100 futures (NQ), NinjaTrader's back-adjusted 1-minute history with contract volume, January 2020 to 25 September 2026. The bands are ratios to price, and back-adjustment shifts older prices by constants, which changes the band width by a few percent at most. Periods: development 2020 to 2023, validation January 2024 to March 2025, holdout April 2025 to September 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: $15.76 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 contract. 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

Both New York versions made money in every calendar year, 2026 counted to September.

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

What would change our mind: a record on new NQ data, with the rule unchanged, that contradicts the backtest. That test is pre-registered: the protocol, frozen on 11 October 2026 before any new data, and the monthly journal are on the forward-test page.

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

What the results show

• In the New York session both versions passed all six gates: positive in development, validation and the holdout, in every calendar year, long and short, and above the control.
• The Asia and London sessions did not show an edge; London lost heavily.
• Costs were small next to the result: about $15.76 a trade against a gross of +$169.72.

What remains unknown

• Whether it holds on new data: the holdout had been used by earlier tests on this file, so only a forward test can confirm it.
• Why it works on NQ and not, by our gates, on ES.
• Live fills in fast markets, and the position sizing and leverage of the original paper, which we did not use.

Why it passed

In the New York session NQ moved far enough, often enough, after leaving its noise area for the trailing stop to keep the trend days and cut the others. Each trade made +$153.96 on average after $15.76 of costs, and it did so in every year from 2020 to 2026. That is a candidate, not a proof: eighteen variants were tried, and the data had been seen by earlier tests.

What you can apply to your own tests

• When something passes, try it on the closest market you have not used. Here ES did not confirm it.
• Before trusting a pass, write down how you would test it on new data: the unchanged rule, the start date and what would count as a failure.
• Check a result session by session: the same rule earned in New York and lost in London.

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 NQ futures?

In our backtest it did in the New York session: +$231,557 per contract over 1,504 trades from 2020 to 2026 for the long-and-short version, after costs, passing every gate. It is a candidate, not a proof: the data had been used by earlier tests, so a forward test on new data comes next.

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 NQ futures 1-minute data in three sessions, one contract 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:

Back-adjustment. The bands are ratios to price; back-adjusted older prices shift those ratios by a few percent at most. The trades, entries and exits are on the real minute-to-minute moves, which back-adjustment does not change.

11Corrections

11 Oct 2026: a link to the forward-test protocol, frozen that day, was added; no number 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

• Noise area breakout es
• Noise area breakout gold
• Intraday momentum strategy nq
• Ema ribbon 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