Week commencing Monday 21 September 2026

At a glance

Every window that appeared, in the order the indicator produced them.

▲ / ▼ / ∅ are the measured shares of comparable historical windows in which the upper target was reached first, the lower target was reached first, or neither was reached by the session close; n is how many such windows sit behind them. They are a historical null and not a prediction of the day being read. Every reading on this page was taken with the target multiple set to ×0.5, the setting we recommend for NQ and the default in the current build, unchanged since w/c 17 August. These captures are 5-minute charts, so outcome times are the bar in which the rail was crossed and are good to the bar; every time on this page was read back off the 5-minute bar data rather than off the pictures. The session close is the 16:55 ET bar, and the three windows that reached neither rail were scored to it.

Two comparisons appear in every capture and they answer different questions. The box tag grades the window against its own last five sessions (💰 accumulation, ⚖️ normal, 🚀 expansion), while the status table's regime row grades it against the full seventeen-year history, and the two are allowed to disagree. They read the same way on seven of the ten windows this week. The three that split were Thursday's hour, 💰 Accumulation on the box against compressed in the table, Friday's hour, ⚖️ Normal on the box against expanded, and Wednesday's half hour, ⚖️ Normal on the box against expanded. With the multiple fixed at ×0.5 the regime no longer moves the rails, so it is reported here but it is not doing anything to the targets. The Window strength row read 68% favoured-rail all week, #11 of 91 for the hour and #14 of 91 for the half hour; it describes the window rather than the day.

The red folder USD calendar was empty all week: no high-impact release on any of the five days, so no result on this page sits inside a news window.

The 60-minute window · 09:30 to 10:30

The longer window waits for the whole first hour to complete before it commits. It gets a bigger, better-defined range to work from and, at a fixed multiple, a target that sits further away in points.

Mon 21 Sep 2026 · 09:30 +60m · NQ 5-minute

Monday 21 September

A box twice its average, a long lean, and a rail cleared by five eighths of a point

🧲✅ favoured side reached first · 12:20

• IB range303.25 pt · avg5 151.00 · 🚀 Expansion
• Split▲ 49% · ▼ 11% · ∅ 40% (n 620)
• Shorter horizons▲ 2h 27% · 4h 40% · ▼ 2h 3% · 4h 7%
• Target distance▲ 151.6 pt (×0.5)

The first hour covered 303.25 points against a five-day average of 151, 🚀 EXPANSION on the box and expanded in the table. The low, 30,216.50, was the 09:30 bar itself and the high, 30,519.75, came on the 10:25 bar, the last one of the window, so the box closed at its own top. The read was 49 / 11 / 40 from 620 sessions, directional row 82%, magnet on the upper rail, target 151.6 points above the box at 30,671.38.

Price held around the box high for half an hour, dipping back inside it to 30,500.50 at 10:40, then climbed. The 12:15 bar reached 30,664.25 and stopped 7.1 points short; the 12:20 bar reached 30,672.00 and cleared the rail by 0.625 points. Magnet and tick, one hour and fifty minutes after the window closed, and the thinnest margin of the week's seven crossings. The move then kept going to 30,863.00 at 15:35, 191.6 points beyond the rail, and the 16:55 bar closed near 30,770. The lower rail at 30,064.88 was never within 435 points of the day after the box closed.

Fifth draw of 49 / 11 / 40 on the record and the fourth in Atlas mode: unreached on 27 August, paid on 1 September, unreached on 2 September, paid today, and paid again on Thursday. The 40% ∅ share is the largest any lean this week carried, shared with Thursday's draw of the same cell, and it is a frequency over 620 sessions, not a comment on this one.

Tue 22 Sep 2026 · 09:30 +60m · NQ 5-minute

Tuesday 22 September

A long lean that finished 33.6 points short at ten to four

∅ neither target reached · the 38% branch

• IB range224.25 pt · avg5 170.05 · ⚖️ Normal
• Split▲ 47% · ▼ 15% · ∅ 38% (n 581)
• Shorter horizons▲ 2h 25% · 4h 36% · ▼ 2h 5% · 4h 11%
• Target distance▲ 112.1 pt (×0.5)

224.25 points against a 170.05 average, ⚖️ NORMAL on both rulers. The low, 30,763.25, was the 09:30 bar and the session low as well; the high, 30,987.50, came at 10:10. The read was 47 / 15 / 38 from 581 sessions, directional row 76%, magnet on the upper rail, target 112.1 points above the box at 31,099.62.

Price never left the box on the downside: the low after the window was 30,883.50 at 11:40, 120.25 points inside it. It left on the upside and did not get far enough. The session high of 31,066.00 came on the 15:50 bar, 33.6 points under the rail, twenty minutes after the same rally had paid the half hour, whose rail sat 60.4 points lower. Price fell back and the 16:55 bar closed near 31,000. The lower rail at 30,651.12 was never within 232 points.

Third draw of 47 / 15 / 38: paid on 3 September, unreached on 11 September, unreached today. A 38% ∅ share describes a window that ends inside both rails more than a third of the time, and two of three draws have.

Wed 23 Sep 2026 · 09:30 +60m · NQ 5-minute

Wednesday 23 September

The widest hour since the multiple was fixed, and the rail it put 186 points away was never approached

∅ neither target reached · the 38% branch

• IB range372.75 pt · avg5 184.80 · 🚀 Expansion
• Split▲ 13% · ▼ 49% · ∅ 38% (n 435)
• Shorter horizons▲ 2h 5% · 4h 9% · ▼ 2h 31% · 4h 41%
• Target distance▼ 186.4 pt (×0.5)

372.75 points against a 184.8 average, 🚀 EXPANSION on the box and expanded in the table. The high, 31,013.00, was the 09:30 bar and the session high; the low, 30,640.25, came on the 10:25 bar, the last one of the window, and was the session low. The hour fell from its first bar to its last. That is the widest first hour on the record since the multiple was fixed at ×0.5 on 17 August, and the third widest of all, behind 425.5 on 6 August and 375 on 13 August.

The read was 13 / 49 / 38 from 435 sessions, directional row 79%, magnet on the lower rail. At a fixed multiple a box this size puts the target 186.4 points away, at 30,453.88. Price bounced to 30,827.75 by 10:55 and then held between the box low and its midpoint all afternoon. The closest it came was 30,646.50 on the 13:00 bar, 192.6 points short of the rail and 6.25 above the box low. The 16:55 bar closed near 30,775.

Fourth draw of 13 / 49 / 38: crossed against on 26 August, unreached on 31 August, paid on 15 September, unreached today and unreached again on Friday. The half hour read the same morning from a 180-point box, leaned the same way, and was paid at 10:10.

Thu 24 Sep 2026 · 09:30 +60m · NQ 5-minute

Thursday 24 September

A long lean from the narrowest hour of the week, paid by one 192-point bar

🧲✅ favoured side reached first · 12:15

• IB range148.00 pt · avg5 237.40 · 💰 Accumulation
• Split▲ 49% · ▼ 11% · ∅ 40% (n 620)
• Shorter horizons▲ 2h 27% · 4h 40% · ▼ 2h 3% · 4h 7%
• Target distance▲ 74.0 pt (×0.5)

148 points against a 237.4 average, 💰 ACCUMULATION on the box and compressed in the table, the narrowest first hour of the week. The low, 30,490.25, was the 09:30 bar and the session low; the high, 30,638.25, came on the 10:25 bar, the last one of the window. The read was the same cell as Monday's, 49 / 11 / 40 from 620 sessions, directional row 82%, magnet on the upper rail, target 74 points above the box at 30,712.25.

Price sagged for an hour, from 30,644.25 at 10:35 to 30,501.75 at 11:30, 11.5 points above the box low and 85.5 above the lower rail. The 12:10 bar was still inside the box. Then the 12:15 bar ran from 30,556.00 to 30,748.25, 192.25 points in five minutes, through the half hour's rail at 30,689.00 and through this one, finishing 36 points beyond it. Magnet and tick, one hour and forty-five minutes after the window closed. The high of the day, 30,827.50, came at 13:05.

Two windows, one bar. The scorer counts Thursday as two hits, as it must, and a reader weighing the week should count the morning once. Sixth draw of this cell and its third payment in Atlas mode, twice this week.

Fri 25 Sep 2026 · 09:30 +60m · NQ 5-minute

Friday 25 September

A short lean from a box the half hour's cross had just deepened, and the day went nowhere

∅ neither target reached · the 38% branch

• IB range242.25 pt · avg5 239.90 · ⚖️ Normal
• Split▲ 13% · ▼ 49% · ∅ 38% (n 435)
• Shorter horizons▲ 2h 5% · 4h 9% · ▼ 2h 31% · 4h 41%
• Target distance▼ 121.1 pt (×0.5)

242.25 points against a 239.9 average, ⚖️ NORMAL on the box and expanded in the table. The high, 30,926.25, came at 09:55, the last bar of the half hour. The low, 30,684.00, came at 10:15, on the bar that took the half hour through its lower rail against a long lean. The hour read 13 / 49 / 38 from 435 sessions, the same cell as Wednesday, directional row 79%, magnet on the lower rail, target 121.1 points below the box at 30,562.88.

The first bar after the window, 10:30, made the low of the rest of the day at 30,718.50, 155.6 points short of the rail. Price then climbed back to the top of the box and spent the afternoon in its top half, nine bars poking above its high: the session high of 30,952.50 on the 12:15 bar was 26.25 points above the box high and 94.9 points under the upper rail at 31,047.38, and the 16:55 bar closed near 30,920. Neither target was approached.

Friday is the only morning of the five where the two durations leaned opposite ways. The half hour was long and was crossed in fifteen minutes; the hour, reading the box that cross had just made deeper, was short and reached nothing. Fifth draw of 13 / 49 / 38, its third blank.

The 30-minute window · 09:30 to 10:00

The shorter window commits half an hour earlier, from a smaller range, and its target is therefore usually nearer. It is a different measurement of the same morning, not a second opinion on it.

Mon 21 Sep 2026 · 09:30 +30m · NQ 5-minute

Monday 21 September

A long lean off a box nearly twice its average, paid thirty minutes after it closed

🧲✅ favoured side reached first · 10:30

• IB range206.25 pt · avg5 113.35 · 🚀 Expansion
• Split▲ 65% · ▼ 16% · ∅ 19% (n 635)
• Shorter horizons▲ 2h 48% · 4h 57% · ▼ 2h 8% · 4h 12%
• Target distance▲ 103.1 pt (×0.5)

206.25 points in the first half hour, 🚀 EXPANSION against a 113.35 average and expanded in the table. Low 30,216.50 on the opening bar, high 30,422.75 on the 09:55 bar, the last one of the window. The read was 65 / 16 / 19 from 635 sessions, directional row 81%, decisive row 81%, magnet on the upper rail, target 103.1 points above the box at 30,525.88.

Price dipped to 30,396.75 at 10:05, inside the box, and climbed. The 10:30 bar, the first one after the hour's box had closed, reached 30,534.25 and cleared the rail by 8.4 points. Magnet and tick.

Read this mark and the hour's together. The two boxes share the 30,216.50 low, the second half hour added 97 points at the top and nothing at the bottom, and one rally cleared both rails one hour and fifty minutes apart. They are separate windows and the scorer counts them separately, as it must, but a reader judging the week should count the morning once. Sixth draw of 65 / 16 / 19 and its third payment.

Tue 22 Sep 2026 · 09:30 +30m · NQ 5-minute

Tuesday 22 September

The same lean as Monday, paid five and a half hours after the box closed

🧲✅ favoured side reached first · 15:30

• IB range184.00 pt · avg5 121.45 · 🚀 Expansion
• Split▲ 65% · ▼ 16% · ∅ 19% (n 635)
• Shorter horizons▲ 2h 48% · 4h 57% · ▼ 2h 8% · 4h 12%
• Target distance▲ 92.0 pt (×0.5)

184 points against a 121.45 average, 🚀 EXPANSION on the box and expanded in the table. Low 30,763.25 on the opening bar, high 30,947.25 at 09:50. It drew the same cell as Monday, 65 / 16 / 19 from 635 sessions at 81% directional, magnet on the upper rail, target 92 points above the box at 31,039.25.

Nothing was settled for five hours. Price reached 30,987.50 at 10:10, 51.75 points short, fell back to 30,883.50 at 11:40, and only started climbing after two o'clock. The 15:25 bar stopped 12.25 points short; the 15:30 bar reached 31,048.00 and cleared the rail by 8.75. Magnet and tick, five hours and thirty minutes after the window closed, the longest wait of the week's seven crossings. The high of the day, 31,066.00 at 15:50, left the hour's rail 33.6 points out of reach.

Seventh draw of 65 / 16 / 19, the second most drawn cell on the record, one behind 28 / 53 / 19: paid on 26 August, crossed against on 4 September, unreached on 9 and 11 September, paid on 16 September, paid on Monday, paid today. Four payments, one cross and two blanks from seven draws of a 65% share is roughly what 65% describes, and seven draws is far too few to say more.

Wed 23 Sep 2026 · 09:30 +30m · NQ 5-minute

Wednesday 23 September

A short lean at 62%, paid ten minutes after the box closed

🧲✅ favoured side reached first · 10:10

• IB range180.00 pt · avg5 142.05 · ⚖️ Normal
• Split▲ 17% · ▼ 62% · ∅ 21% (n 396)
• Shorter horizons▲ 2h 7% · 4h 12% · ▼ 2h 48% · 4h 57%
• Target distance▼ 90.0 pt (×0.5)

180 points against a 142.05 average, ⚖️ NORMAL on the box and expanded in the table. The high, 31,013.00, was the opening bar and the low, 30,833.00, was the 09:55 bar, the last one of the window: the half hour fell from its first bar to its last, as the hour would go on to do. The read was 17 / 62 / 21 from 396 sessions, directional row 79%, decisive row 79%, the largest ▼ share of the week's three short leans. Magnet on the lower rail, target 90 points below the box at 30,743.00.

The 10:05 bar fell to 30,766.25, 23.25 points above the rail. The 10:10 bar fell to 30,703.75 and through it by 39.25. Magnet and tick, ten minutes after the window closed, the second earliest resolution on the record after the 10:00 of 18 September. The fall carried on to 30,640.25 at 10:25, 102.75 points beyond the rail, where it made the low of the hour's box and of the whole session.

First appearance of 17 / 62 / 21 anywhere on the record, from a Medium bin of 396 sessions. One draw of a cell is one draw. The hour leaned the same way off the same morning and its rail, 186 points from a box more than twice this size, was never within 192 points.

Thu 24 Sep 2026 · 09:30 +30m · NQ 5-minute

Thursday 24 September

An abstain at 58%, and the larger share was reached by the same bar that paid the hour

✅ larger share reached first · 12:15 · abstained window

• IB range132.50 pt · avg5 156.35 · ⚖️ Normal
• Split▲ 50% · ▼ 36% · ∅ 14% (n 365)
• Shorter horizons▲ 2h 35% · 4h 43% · ▼ 2h 24% · 4h 30%
• Target distance▲ 66.3 pt (×0.5)

132.5 points against a 156.35 average, ⚖️ NORMAL on both rulers. Low 30,490.25 on the opening bar, high 30,622.75 at 09:45. The read was 50 / 36 / 14 from 365 sessions and the directional row printed 58%, four points under the 62% the Atlas mode asks for, so no magnet was placed and both rails printed grey. The only window of the ten that abstained.

The box ran 30,490.25 to 30,622.75, rails at 30,424.00 and 30,689.00. Price made 30,644.25 at 10:35, 44.75 points short of the upper rail, and sagged to 30,501.75 by 11:30, 77.75 above the lower. Then the 12:15 bar ran 30,556.00 to 30,748.25, through this rail by 59.25 points and through the hour's 23.25 points further on. A plain green tick, recording that the larger of the two grey shares got there first, two hours and fifteen minutes after the window closed.

Third draw of 50 / 36 / 14, all three abstains at the same 58%: crossed on 17 August, neither on 17 September, larger share today. Last Thursday's half hour was 132 points wide and read this exact cell; this one was 132.5 and read it again. Half a point of box, the same grey rails, a blank one week and a tick the next.

Fri 25 Sep 2026 · 09:30 +30m · NQ 5-minute

Friday 25 September

A long lean crossed on the wrong side fifteen minutes after the box closed, by 3.75 points

🧲❌ crossed against the lean · 10:15

• IB range159.00 pt · avg5 156.45 · ⚖️ Normal
• Split▲ 59% · ▼ 24% · ∅ 17% (n 562)
• Shorter horizons▲ 2h 43% · 4h 52% · ▼ 2h 12% · 4h 19%
• Target distance▲ 79.5 pt (×0.5)

159 points against a 156.45 average, ⚖️ NORMAL on both rulers. Low 30,767.25 on the opening bar, high 30,926.25 on the 09:55 bar, the last one of the window, so the box closed at its own top. The read was 59 / 24 / 17 from 562 sessions, directional row 71%, decisive row 83%, magnet on the upper rail, target 79.5 points above the box at 31,005.75.

The 10:00 bar, the first after the window, fell to 30,798.25, still 31 points inside the box. The 10:05 bar fell to 30,752.25 and out of the bottom of it, and the 10:10 bar to 30,735.00. The 10:15 bar fell to 30,684.00 and through the lower rail at 30,687.75 by 3.75 points. Magnet and cross, fifteen minutes after the window closed, the earliest any window on the record has been marked against. Price turned on that bar: the high of the rest of the day, 30,952.50 at 12:15, was 53.25 points short of the rail the lean had pointed at.

The same 10:15 bar set the low of the hour's box, which then leaned short from it and reached nothing. Friday is the only morning of the five on which the two durations favoured opposite rails. First appearance of 59 / 24 / 17 on the record, and the fifth Atlas-mode lean of 53 to be crossed against.

Five half hours resolved, three hours did not

Seven of the ten windows on this page were settled by a rail. All five half hours were, which no week on the record had managed before, and four of the five went to the favoured or larger side. Two of the five hours were, and the three that were not are the week's three blanks. The same five mornings, read at two lengths, gave the shorter window a clean sheet and the longer one a losing one, and neither says anything about the next five.

Three cells drew twice each this week with identical shares and bins both times, and each pair produced the same mark. 49 / 11 / 40 from 620 read Monday's and Thursday's hours: both paid, from boxes of 303.25 and 148 points. 65 / 16 / 19 from 635 read Monday's and Tuesday's half hours: both paid, at 10:30 and 15:30. 13 / 49 / 38 from 435 read Wednesday's and Friday's hours: both reached nothing. A week ago two repeated cells gave opposite marks each time; this week three gave matching ones. Both are what a frequency measured over hundreds of sessions looks like when it is sampled twice, and neither week is evidence about the cell.

Friday is the only morning of the five where the two durations favoured opposite rails. The half hour leaned long at 71% off a 159-point box and the 10:15 bar, the fourth after it closed, fell through its lower rail by 3.75 points, the earliest cross on the record. That bar set the low of the hour's box; the hour closed fifteen minutes later reading 13 / 49 / 38, leaned short from it, and then watched price climb back to the top of the box and spend the afternoon in its top half, never within 94.9 points of either rail. One morning, one lean crossed and one lean unreached, and the only rail price reached all day was cleared by 3.75 points at the bottom of one candle.

Monday's hour was cleared by 0.625 points: the 12:15 bar stopped 7.1 short and the 12:20 bar reached 30,672.00 against a rail at 30,671.38. It is scored as a payment. A week ago Wednesday's half hour was scored the same way on 0.45 points, and last Thursday's two blanks were 15.8 and 10.5 points short. The record grades rails reached and not approaches, and that convention has to hold in the weeks it helps as well as the weeks it costs.

Across eight weeks the running count now stands at 80 windows: 33 reached the favoured or larger share first, 11 reached the other rail first, and 36 reached neither. Of the 53 that were in Atlas mode, 25 paid, 23 reached nothing and 5 were crossed against. Eighty windows cannot confirm or refute frequencies measured over hundreds of sessions, and nine leans in one week is a count of leans, not a grade.

What this week actually shows

Five things, from the week with the most leans the record has published.

1. Nine of ten windows offered a lean. The previous most was eight, on w/c 24 August and w/c 14 September. Five of the nine were paid, which equals last week's five of eight rather than beating it; one was crossed and three reached nothing. The Atlas-mode count across the whole record is now 53 windows: 25 paid, 23 unreached, 5 against.
2. Every half hour reached a rail, and three of five hours did not. Five of five is the first clean sheet for the shorter window on the record; the previous best was four, four times. The running total across 80 windows is 33, 11 and 36, and the three hours that reached nothing carried three of the four furthest targets of the week.
3. Three repeated cells, three matching pairs. 49 / 11 / 40 paid on Monday and Thursday; 65 / 16 / 19 paid on Monday and Tuesday; 13 / 49 / 38 reached nothing on Wednesday and Friday. A week earlier two repeated cells split their marks. Identical readings agreeing on one week and disagreeing on another is exactly what a share over hundreds of sessions looks like from two draws.
4. Two windows were settled by one bar, and one window by five eighths of a point. Thursday's 12:15 bar crossed both rails in five minutes; Monday's hour cleared its rail by 0.625 points on the 12:20 bar after stopping 7.1 short on the one before. Both are scored as the indicator scores them, and both are worth knowing the size of.
5. Friday's half hour was crossed against at 10:15, by 3.75 points. It is the earliest cross on the record and the fifth Atlas-mode lean of 53 to be marked against. The same bar made the low of the hour's box, the hour leaned the other way from it, and the session then reached neither of the hour's rails.