$$$ INTRADAY MOMENTUM DIGGA $$$ - Noise Area Breakout — Strategy by PINEDIGGA
By PINEDIGGA
Performance Metrics
- Author: PINEDIGGA
- Symbol: CME_MINI:MNQ1!
- Timeframe: 5 minutes
- Profit Factor: 1.27
Description
$$$ INTRADAY MOMENTUM DIGGA $$$ v1.0 - Noise Area BreakoutMost of the day the Nasdaq just makes noise. This strategy waits until price leaves the normal noise range for that time of day, then follows it until the close. It is built and measured on NQ futures, and it shows every number, the bad years included.The idea comes from Zarattini, Aziz & Barbon (2024), "Beat the Market: An Effective Intraday Momentum Strategy". This is an independent build of that rule. Before publishing, its trades were checked one by one against an offline reference on CME NQ 1-minute data.How it worksNoise band. For every half hour of the session, the script averages how far price has typically moved away from the open by that time, over the last 14 days. That gives an upper and a lower band. They start narrow at 09:30 and widen through the day, like a funnel.Gap anchor. The upper band is measured from the higher of today's open and yesterday's close, and the lower band from the lower of the two. So a gap has to be overcome first.Checks every 30 minutes (10:00, 10:30 … 15:30 New York). The trade is long if the close is above both the upper band and VWAP, short if it is below both the lower band and VWAP, and flat otherwise. Orders are filled at the next bar's open.Trailing stop. For a long, the stop is the higher of the upper band and VWAP; for a short, the lower of the lower band and VWAP. It is checked only at the 30-minute marks.Session close. Everything is closed at 16:00 New York. Nothing is held overnight.Default: a 15-minute Trend Magic filter (forward test)By default, a trade is taken only in the direction of the last closed 15-minute Trend Magic (CCI 20, ATR 5, ×1). Trend Magic is the indicator published on TradingView by KivancOzbilgic; full credit to its author. It is rebuilt here from the chart's own bars so it can run on 15-minute buckets.A trade is also closed when a 5-minute candle closes on the wrong side of that Trend Magic.Results on NQ, 1.5 pt cost per trade:2025-26: +18.1 pt/trade, against +9.1 pt without the filter. Drawdown 1,216 pt against 1,495.2022-25: +8.6 pt/trade against +9.3 (slightly worse).2010-21: +1.1 pt/trade against +1.5 (worse).2010-26: profit factor 1.27 against 1.25. Max drawdown 1,540 pt against 2,116 pt.It keeps the big trend days.It did not pass our pre-registered test, because 2010-21 is worse. It ships switched on as a forward test: we will publish how it does on new data. To use the original rule, set "Trend Magic 15m" to Off.The tooltip lists all five choices with their measured numbers.What it measured (CME NQ, 1 contract, 1.5 pt cost per trade)TradingView, 5m chart, July 2020 – October 2026, defaults: 1,468 trades, +12,986 pt, profit factor 1.35, win rate 41%, max drawdown 1,566 pt. 1,433 of 1,449 reference trades match on day, side and entry minute.Reference engine, 2010-26, defaults: 3,709 trades, +14,156 pt, profit factor 1.27, win rate 36%. The average win is 49 pt and the average loss 22 pt.Year by year, defaults: 2010-2017 flat to negative, only 2015 positive (+76 pt, the worst −286 pt), 2018 +1,695, 2019 −209, then 2020 +1,084, 2021 +1,208, 2022 +2,481, 2023 +1,886, 2024 +1,922, 2025 +1,435, 2026 (to mid-October) +3,496.Read this before you use itThe profit comes from a few trend days a year. Without the 5 best trades of each year, every tested period is negative. Skipping signals after a losing streak is how you miss those days.You will see clusters of quick trades that get stopped out. They are the cost of the rule. Trades shorter than 30 minutes lose about 20 pt on average, but you only know a trade was short after it has closed. Re-entries pay overall, net of cost, 2010-26: the 2nd trade of the day +3.1 pt, a re-entry after a stop +2.6, after a Trend Magic exit +2.9. A limit of 2 trades a day would have cost +596 pt.Long flat stretches are normal. 2010-2017 was flat to negative after costs (NQ was also much cheaper then, so points were smaller).Size small, for example 1 MNQ. The size is fixed: it does not grow with the account or change with volatility.How to run itUse a 5-minute chart. It gives the same trades as 1-minute, with years more history. On 1-minute, a normal account loads about 2.5 weeks, less than the 14-day warm-up. The Trend Magic default needs a 1m or 5m chart; on other timeframes the table says so and no trades are taken.Symbols: NQ1! / MNQ1!, or a Nasdaq-100 CFD. The session is 09:30-16:00 New York. US holidays and half days are skipped. Other markets need their own session and timezone and have not been measured.CFD users: Size 1 on a CFD usually means 1 index unit, $1 per point. Size 20 = 1 NQ, size 2 = 1 MNQ; check your broker's contract size. The script charges 3 ticks of slippage per side, which is 1.5 pt per trade on NQ but almost nothing on a CFD with a 0.01 tick. Set Settings → Properties → Slippage to 75 ticks (0.75 pt per side), or the backtest will show the CFD without its spread.Chart markers. The arrows and +1 / −1 labels are drawn by TradingView. To hide them, untick "Trade signals" or just "Quantity" under Settings → Style.On the chartThe noise band is drawn as a smooth curve. At each check time it is exactly the level that check uses, and between two checks it runs straight from one to the next. Both ends are known in advance, so nothing is drawn from the future.A soft glow sits along the band edges.The 15m Trend Magic line has a shade between it and the price.A status table shows the state, the next check time, both bands, VWAP and the Trend Magic side.The colours follow a dark or light chart automatically.AlertsCreate one "Any alert() function call" alert. It fires on every long, short, flat, Trend Magic exit and session close, and each message includes the price, the bands and VWAP. Signals fire on the close of the check bar, and the order is for the next bar's open.SettingsTrade: size, and direction (both, long only or short only).Rule: session, timezone, lookback days, band width, check interval, the VWAP filter, the gap anchor and holiday skipping. Every setting's tooltip gives the value that was measured.Trend Magic 15m: the five choices and the line and shade switches.Visuals and alerts.This script is for education and research. Past results, measured or backtested, do not promise future results. Futures and CFDs carry a high risk of loss.© PINE DIGGA | November 2026