Noise Area Breakout Backtest: YM (Dow) Futures, Asia, London and New York

00The verdict in three lines

• The "noise area" breakout of Zarattini, Aziz and Barbon (2024) is the only day-trading rule on this site that passed all six of our gates: on NQ futures in the New York session, long and short and long only, now being tested on new data. On ES the same rule made money but did not pass. We tested it unchanged on E-mini Dow futures (YM), in the same three sessions, with the rules, costs, gates and a control fixed before the run, and wrote down in advance what would count as a replication: a pass of either New York version.
• None of the six variants passed, and the New York session made nothing even before costs: +$0.03 a trade for the long-and-short version over 1,603 trades, +$0.16 for long only, so after $15.76 of costs a trade they lost $25,218 and $12,510. The same entries placed on other days did as well (percentile 53.0 and 50.8 of the control, 95 needed). Asia and London lost too.
• The NQ result did not carry over. Before costs, the New York long-and-short version made +0.046% of the price a trade on NQ, +0.023% on ES and 0.000% on YM. This does not disprove the NQ result, but it means the rule is not a general edge in US index futures; whether it holds on NQ itself is for its forward test to show.

01The rule, unchanged

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-YM-AS-LS, NA-YM-AS-L, NA-YM-LD-LS, NA-YM-LD-L, NA-YM-NY-LS, NA-YM-NY-L. With these, the noise area rule has been tried in 28 variants on this site (18 on ES, NQ and gold, 4 on Bitcoin, 6 here), so the protocol requires a development t-statistic of 2.5, as on the other markets. Replication check, written before the run: the NQ result counts as replicated on YM only if NA-YM-NY-LS or NA-YM-NY-L passes all six gates.

02Data and costs

E-mini Dow futures (YM, $5 a point), NinjaTrader's back-adjusted 1-minute history with contract volume, January 2020 to 25 September 2026, cut at the same time as our ES and NQ noise-area tests. The bands are ratios to price, and back-adjustment shifts older prices by constants, which changes the band width a little. 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 our trendline and Sneaky Pivot tests on YM, so the holdout is reported as already used; no noise-area number on YM had been computed. Costs: $15.76 a round trip ($5.76 of commission and one tick, $5, of slippage on each fill), as in all our YM tests; a flip pays two. With two ticks of slippage per fill every average below would be $10 a trade lower. Engine check: the engine and the independent re-implementation are those of the NQ test, unchanged (its hash is pinned by the NQ forward test); na_ym_selftest.js checked the YM data, the New York early closes against our record of the file, and all six YM variants trade by trade against the independent re-implementation (na_ym_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

Both New York versions lost in every period. London lost in development and validation and made money in the holdout (+$12,062 long and short), not enough to make up for the years before.

04Three US index futures compared

The New York versions of the same rule on NQ, ES and YM, over the same years. The gross result as a share of the entry price is comparable across the three contracts; points and dollars are not, because the indices trade at different levels.

The edge before costs shrinks from the Nasdaq-100 to the S&P 500 and is gone on the Dow. If the rule caught a general intraday momentum in US stock index futures, YM should have shown at least a small version of it. It showed none, so the NQ result is either specific to NQ or partly luck; the backtest cannot tell which. Its forward test on new NQ data can.

05New York by year

2026 counted to September.

06Longs, shorts and the best trades

07Is 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. On YM it did not, in any session.

08The gates

09Verdict by variant

The causes follow the protocol's definitions: Costs where the result before costs was positive and the result after costs negative. For NA-YM-AS-LS, NA-YM-AS-L, NA-YM-NY-LS, NA-YM-NY-L the gross was close to zero (at most +$6.36 a trade, about one point), so costs did not kill a real edge: there was none to kill. The headline cause is therefore No proven edge.

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. In the New York session, the one that passed on NQ, the rule made nothing even before costs.
• The same entries placed on other days did as well: the New York versions ranked at percentile 53 and 51 of the control, where 95 was needed.
• Across the three US index futures the edge shrank from NQ to ES and was gone on YM: +0.046%, +0.023% and 0.000% of the price a trade before costs.

What remains unknown

• Whether the NQ result holds on new NQ data: only its forward test can say.
• Why the rule earned on NQ and not on the Dow; a stronger intraday trend on the Nasdaq-100 is one explanation, not a tested one.
• Position sizing and leverage, which we did not use, and live fills.

Why it failed

On YM, leaving the noise area did not lead to moves that ran far enough to pay for the false breaks: in New York the long-and-short version made +$0.03 a trade before costs, about zero, and lost the $15.76 of costs on every trade. The control shows the timing added nothing.

What you can apply to your own tests

• When a rule passes on one market, test it unchanged on the closest markets before trusting it. This one did not carry over to ES or YM.
• Compare markets in a unit that does not depend on the contract: here the gross result as a share of the price.
• Write down before the run what would count as a replication; here it was a pass of either New York version.

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 Dow futures?

Not in our test: none of the six versions passed on YM from 2020 to 2026. The New York versions lost $25,218 and $12,510 after costs, and made almost nothing before them.

Why did it pass on NQ and not on YM?

The backtest cannot say. Before costs the New York version earned about 0.05% of the price a trade on NQ, half that on ES and nothing on YM. A stronger intraday trend on the Nasdaq-100 is one explanation; luck is another. The forward test on new NQ data will show which.

What is the noise area in day trading?

The range around the session open where price usually moves at that time of day, from the average move of the last 14 sessions (Zarattini, Aziz and Barbon, 2024). The rule buys when price closes above it and sells short below it, with a stop at the band or the VWAP.

How was the noise area breakout tested on YM?

With the NQ rules unchanged, in the Asia, London and New York sessions, $15.76 of costs a round trip, a control that moves the same entries to other days, and 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.

10Check 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.

11Data 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 a little. The trades, entries and exits are on the real minute-to-minute moves, which back-adjustment does not change.

12Corrections

None since publication.

+More tests and guides

• Noise area breakout nq
• Noise area breakout es
• Previous day high and low strategy
• Day trading strategies backtested: this test next to the other day-trading tests
• 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