The short answer
An NR7 day is the day with the narrowest range of the last seven (range = high minus low). On gold it is common: 513 of 3,260 tested days from January 2012 to 17 November 2026 were NR7 days, about one day in six.
Here is what the next day did, from our own daily data:
- It traded beyond the NR7 high or low in 501 of 513 cases (97.7%). It broke only the high in 207 (40.4%), only the low in 167 (32.6%) and both sides in 127 (24.8%).
- It closed outside the NR7 range in 363 cases (71%), against 52% for an ordinary day measured against the day before it.
- It was about 1.8 times bigger than the NR7 day itself, but it was not bigger than a normal day: its median range was 0.86 times the 20-day average range, against 0.89 for all days.
So the NR7 day is best read as a quiet day that gives you two tight levels, not as a promise of a big move. What happens after the break is close to what gold does on any other day.
What an NR7 day is, in simple words
Every daily candle has a range: the distance from its high to its low. Some days gold travels $150, some days $40. An NR7 day is simply the day whose range is smaller than each of the six days before it.
Why do traders care? Markets swing between quiet and busy. After a run of quiet days, many traders expect a busy one. The NR7 day hands you two clear lines, its high and its low, very close together. If price breaks one of them the next day, a breakout trader has a reason to act and a nearby level for the stop on the other side.
The idea is usually credited to Toby Crabel’s 1990 book on short-term price patterns, and was later made popular by Linda Bradford Raschke and Laurence Connors in Street Smarts (1995). It is popular because it is exact: no drawing, no opinion. Either today’s range is the smallest of seven, or it is not.
How to spot one on your chart
- Open the daily XAUUSD chart (the 4-hour version exists, but this page tests daily bars).
- For each of the last seven candles, take high minus low. Most platforms show this in the data window, or you can add an ATR(1) indicator, which plots each day’s range.
- If today’s range is smaller than all six before it, today is an NR7 day. Mark its high and low.
- Tomorrow, watch which line breaks first, and whether price closes beyond it.
How we tested it
- Data: COMEX gold futures (GC=F) daily bars from Yahoo Finance, the same
gold-daily.jsonfile our charts use, refreshed on 18 November 2026. We used 3 January 2012 to 17 November 2026. Before 2012 our file is missing too many highs and lows, and the unfinished 18 November bar is left out. - NR7 rule: a day whose range is strictly smaller than each of the six previous days, counted only when all seven days and the next day have a full open, high, low and close.
- What we counted: the next day’s break (high only, low only, both, or none), whether it closed beyond the NR7 range, the next day’s range compared with the NR7 day and with the 20-day average range, and the close five trading days after a breakout close.
- What we did not do: no entries, stops, targets, spreads or swap. This is a study of how price behaves after an NR7 day, not a backtest of a trading system.
What happened the next day
| Next day after… | NR7 days (513) | All tested days (3,260) |
|---|---|---|
| Broke the high only | 207 (40.4%) | 1,289 (39.5%) |
| Broke the low only | 167 (32.6%) | 1,125 (34.5%) |
| Broke both sides | 127 (24.8%) | 339 (10.4%) |
| Stayed inside | 12 (2.3%) | 507 (15.6%) |
| Closed outside the range | 363 (71%) | 1,693 (52%) |
For “all tested days”, each day is compared with the day before it, so the two columns measure the same thing.
Two lessons hide in this table:
- A break is easy, a clean break is not. Because the NR7 range is small (its median was 0.51 times the 20-day average range), the next day almost always pokes through one side. But in one case in four it went through both sides, which means a breakout order on one side would have been triggered and then seen price run to the other side. On daily bars we cannot tell which side broke first on those 127 days.
- Up-breaks slightly outnumbered down-breaks (207 against 167). Gold rose a lot over these years, and the same tilt shows for all days (1,289 against 1,125), so this is the market’s drift, not a special NR7 effect.
Is the next day really bigger?
Compared with the NR7 day, yes: the next day’s range was a median 1.77 times the NR7 day’s range. That sounds exciting until you compare it with a normal day. Measured against the 20-day average range, the day after an NR7 had a median range of 0.86, and an ordinary day had 0.89.
In plain words: the NR7 day is unusually small, the next day is usually normal-sized again, and normal-sized looks big next to a very small day. The last 12 months show the same picture in dollars: the 38 NR7 days had a median range of $59.6, the days after them $98.2, and all 246 days in that period $95.7.
Follow-through: did the move keep going?
We took every case where the next day closed outside the NR7 range and looked at the close five trading days later.
| Breakout close | Cases | Close 5 days later further in the same direction | Closed back inside the NR7 range within 5 days |
|---|---|---|---|
| Above the NR7 high | 198 | 115 (58.1%) | 68 (34.3%) |
| Below the NR7 low | 165 | 71 (43.0%) | 59 (35.8%) |
For comparison, on all 3,260 days the close five days later was higher than the starting close 54.1% of the time. So upside breakouts kept going a little more often than gold’s normal upward drift, and downside breakouts a little less often than the matching 45.9%. Both differences are small for samples of 165 to 198, and about one breakout close in three was back inside the NR7 range within a week.
Year by year
| Year | NR7 days | Next day broke high only | Low only | Both | Neither | Next day closed outside |
|---|---|---|---|---|---|---|
| 2012 | 39 | 12 | 18 | 8 | 1 | 32 (82%) |
| 2013 | 31 | 13 | 13 | 5 | 0 | 25 (81%) |
| 2014 | 33 | 13 | 9 | 9 | 2 | 28 (85%) |
| 2015 | 39 | 15 | 12 | 11 | 1 | 32 (82%) |
| 2016 | 32 | 15 | 11 | 6 | 0 | 26 (81%) |
| 2017 | 32 | 15 | 8 | 9 | 0 | 24 (75%) |
| 2018 | 24 | 12 | 7 | 4 | 1 | 19 (79%) |
| 2019 | 30 | 13 | 4 | 10 | 3 | 23 (77%) |
| 2020 | 38 | 14 | 16 | 7 | 1 | 29 (76%) |
| 2021 | 34 | 12 | 12 | 10 | 0 | 17 (50%) |
| 2022 | 39 | 16 | 14 | 9 | 0 | 28 (72%) |
| 2023 | 34 | 13 | 9 | 11 | 1 | 21 (62%) |
| 2024 | 34 | 16 | 9 | 9 | 0 | 19 (56%) |
| 2025 | 42 | 19 | 12 | 10 | 1 | 22 (52%) |
| 2026 (to 17 Nov) | 32 | 9 | 13 | 9 | 1 | 18 (56%) |
| All | 513 | 207 | 167 | 127 | 12 | 363 (71%) |
The count is steady, about 30 to 40 a year (2018 and 2019 are lower partly because our file is missing some days in those years). The interesting column is the last one. From 2012 to 2018, the day after an NR7 closed outside its range 81% of the time (186 of 230), against 57% for all days. From 2019 to 2026 that fell to 63% (177 of 283), against 48% for all days. Gold has become more two-sided inside the day in recent years, with more days that run one way and close back in the middle. Treat any rule built on the early years with care.
A real example: 21 and 22 October 2026
On 21 October 2026, gold’s daily range was $38.3 (high $4,160.6, low $4,122.3), smaller than each of the six days before it. That made it an NR7 day.
- In the very first hour of 22 October (22:00 UTC on the 21st, the start of the futures day), price poked above the NR7 high to $4,164.3, only $3.7 above the line, and fell back. A trader who bought that first poke was quickly in trouble.
- By 03:00 UTC price had broken below the NR7 low of $4,122.3, and the selling kept coming. The day’s low was $4,036.8 in the last hour, and gold closed at $4,060.4, $61.9 below the NR7 low.
- The next day, 23 October, it fell further to $3,989.1, and five trading days after the breakout close (29 October) it closed at $4,031.4, still below the 22 October close.
This was one of the 127 “both sides” days, and it shows why many traders wait for a close beyond the NR7 level, or at least for the first hour to settle, before acting.
Not every clean break runs. On 8 October (NR7 range $42.6, high $4,170.7, low $4,128.1), the next day broke only the high and ran to $4,233.7, closing at $4,216.3. Three trading days later, on 14 October, gold closed at $4,107.9, back below the NR7 low. That is the one-in-three “back inside within a week” case from the table above.
The newest NR7 day in our data is 17 November 2026 (range $54.8: high $3,894.0, low $3,839.2). Early on 18 November gold broke below that low and traded under $3,800; that day was not finished when we wrote this, so it is not in the counts.
What the numbers say, in plain English
- NR7 days give you tight, clear levels. That is their real value: a nearby line for an entry trigger and a nearby line on the other side for the stop.
- A break of one side is almost guaranteed and tells you little on its own. A close beyond the level is a stronger fact, and even that came back inside about a third of the time within a week.
- The day after is usually a normal day, not a giant one. Plan targets from the normal daily range (the ATR), not from the hope of a huge expansion.
- Recent years are noisier. The share of next days that closed outside the NR7 range fell from 81% (2012-2018) to 63% (2019-2026).
How traders usually frame an NR7 setup
This is how the pattern is commonly described, not a recommendation:
- Mark the NR7 high and low at the end of the day.
- Check the calendar. A CPI, jobs report or Fed day the next day can turn a quiet range into a fast two-sided day.
- Decide in advance: trade the first touch, or wait for a close (on the daily, or on the 1-hour chart) beyond the level.
- Put the stop beyond the other side of the NR7 range, or at a set multiple of ATR, and size the trade so that stop costs no more than your planned risk. The position size calculator and the ATR stop-loss calculator do the maths.
- Have a time exit. If nothing has happened in a few days, the squeeze idea has failed.
Related set-ups on this site: the inside bar breakout (a pause that sits inside yesterday’s candle), the previous day high and low and the breakout and retest.
Limitations
- Daily bars only. When the next day broke both sides, we cannot see which came first. Hourly data would answer that, but our hourly file only starts in mid-2024.
- Futures, not spot. COMEX futures roll from one contract month to the next, which can change a day’s range slightly around roll dates. Spot XAUUSD on your platform will show slightly different highs and lows.
- Short days. Holiday sessions (around Christmas, Thanksgiving and US holidays) are naturally narrow and can show up as NR7 days without meaning much.
- No costs, no system. Nothing here includes spread, slippage or swap, and nothing here says a rule built on NR7 makes money.
Test it yourself
You can check every number on this page. Download daily gold data (our backtesting with free data guide shows how), mark the NR7 days in a spreadsheet with one formula, and count what the next day did. Then write exact rules, add your own costs, and run them in the Gold Strategy Lab or your platform’s tester before risking money.
Learn it with other gold traders
Range squeezes are easier to judge when you can see how other traders read the same chart. Join Gold Community Free to share NR7 days as they form, the calendar risks around them, and how each one played out. No one can promise profits, and nobody there will.
Educational content only, not financial advice. Trading gold with leverage carries a high risk of loss.
Sources
- COMEX gold futures (GC=F) daily bars via Yahoo Finance, from our own data file
gold-daily.json(refreshed 18 November 2026); our calculation. - Toby Crabel, Day Trading with Short Term Price Patterns and Opening Range Breakout (1990), the work the NR7 idea is usually credited to.
- Linda Bradford Raschke and Laurence Connors, Street Smarts: High Probability Short-Term Trading Strategies (1995).
Frequently asked questions
What is an NR7 day in trading?
NR7 means narrowest range of 7. It is a day whose range (high minus low) is smaller than the range of each of the six days before it. The idea, usually credited to Toby Crabel’s work on range contraction, is that a quiet day is often followed by a bigger one.
How often does gold make an NR7 day?
In our daily COMEX gold data from January 2012 to 17 November 2026, 513 of 3,260 tested days were NR7 days, about 15.7%, or roughly one day in six. Most years had between 30 and 42.
Does gold break out after an NR7 day?
Almost always in the simple sense: the next day traded beyond the NR7 high or low in 501 of 513 cases (97.7%), because the range is so small. The next day closed outside the NR7 range in 363 cases (71%). But the next day was not bigger than a normal day: its median range was 0.86 times the 20-day average range, against 0.89 for all days.
Is the NR7 strategy profitable on gold?
We do not claim that it is. This page describes how the pattern behaved in the past; it does not test entries, stops and costs as a trading system, and past behaviour is no promise. If you want to trade it, write exact rules and test them on your own data and costs first.
What is the difference between NR7 and an inside bar?
An inside bar sits completely inside the previous day’s range. An NR7 day only has to be the smallest range of the last seven; it can stick out above or below yesterday. Some traders look for days that are both (often called NR7 inside days) as an even tighter squeeze.
Comments
No comments yet.
More in Strategies & technical analysis
- Order Flow on Gold Futures: Footprint Charts, Delta and Heatmaps Explained for XAUUSD Traders
- London Close Reversal on Gold: Does XAUUSD Really Turn at 4 pm London? We Tested 601 Days
- How to Identify Support and Resistance on a Gold Chart
- ICT Unicorn Model on Gold: Breaker Block Plus Fair Value Gap, Explained With XAUUSD Rules
- ICT Concepts for Gold Trading: Liquidity, FVGs and Market Structure
Latest on the blog
- The Philly Fed Survey for Gold Traders: Diffusion Index, Prices Paid and the Thursday Reaction
- What Is the Fed Beige Book? A Gold Trader's Guide to District Reports, Price Clues and Market Reaction
- Trading Psychology Quotes That Actually Help: 12 Lines From Famous Traders and What They Mean for Gold
- Gold Trading Tax in the UK: Spread Betting vs CFDs vs Futures, Explained From HMRC's Own Guidance