Week commencing Monday 10 August 2026

A quiet hour, and the table said so

• IB range181.00 pt · avg5 316.65 · 💰 Accumulation
• Split▲ 19% · ▼ 14% · ∅ 67% (n 470)
• Shorter horizons▲ 2h 7% · 4h 12%
• Target distance▲ 181 pt (×1 auto)

181 points against a five-session average of 317 earns the 💰 ACCUMULATION tag, and a compressed hour is exactly where the indicator has least to say. Neither rail comes close to the threshold, both stay grey, and the badge reads abstain.

Look at where the weight sits: ∅ 67% is more than three times either direction. The single most likely outcome, by a wide margin, was that price would reach neither target before the close. That is what happened. The session drifted through the lower half of the window's range and spent the afternoon going nowhere in particular.

Nearly three hundred points of travel, no qualifying run

• IB range206.75 pt · avg5 278.50 · ⚖️ Normal
• Split▲ 20% · ▼ 14% · ∅ 66% (n 369)
• Shorter horizons▲ 2h 4% · 4h 13%
• Target distance▲ 206.8 pt (×1 auto)

Another abstain, and another ∅ in the sixties. But this session was not quiet. The day ranged across nearly three hundred points top to bottom, most of them a long afternoon slide, with a partial recovery into the close.

That is the distinction the ∅ column is drawing, and it is easy to miss. It is not a forecast of a flat tape. It is a statement about reach: whether price is likely to travel a specific distance in one direction before the session ends. Tuesday moved plenty and still reached neither target. A trader holding for the directional destination got nothing out of a day with a great deal of movement in it.

The narrowest hour of the week

• IB range153.25 pt · avg5 248.60 · 💰 Accumulation
• Split▲ 14% · ▼ 19% · ∅ 67% (n 376)
• Shorter horizons▼ 2h 8% · 4h 13%
• Target distance▼ 153.3 pt (×1 auto)

153 points against a five-session average of 249, the second accumulation tag of the week and the third window running to abstain. The shape is by now familiar: ∅ 67%, both rails grey, neither direction anywhere near a lean.

Worth noting how consistent these compressed mornings are. Monday and Wednesday produced almost the same table, 19/14/67 and 14/19/67, from different ranges on different days. The ∅ share is not reacting to the day's news, it is reacting to how little ground the first hour covered.

The week's one lean on the hour, and it did not reach

• IB range375.00 pt · avg5 244.80 · 🚀 Expansion
• Split▲ 69% · ▼ 13% · ∅ 19% (n 620)
• Shorter horizons▲ 2h 54% · 4h 63%
• Target distance▲ 93.8 pt (×0.25 auto)

375 points against a five-session average of 245. This is the one genuinely expansive hour of the week, and the only 60-minute window that produced a lean: ▲ 69%, clear of the 62% threshold, magnet on the upper rail, box tinted.

It did not get there. Price spent the session pressing against the window's high from inside, poked about twenty points beyond it on the mid-afternoon push, and settled back. The target sat 94 points above that high, and the run to it never came. The race closed unresolved and the tick prints against ∅ 19%, the least likely of the three branches.

The half-hour window on the same morning did reach its target, which is the comparison worth making, and it is further down this page. Same direction, same conviction, different destination.

A clear direction and no reach at all

• IB range146.50 pt · avg5 234.70 · 💰 Accumulation
• Split▲ 6% · ▼ 24% · ∅ 70% (n 435)
• Shorter horizons▼ 2h 12% · 4h 19%
• Target distance▼ 146.5 pt (×1 auto)

The most lopsided table of the week, and still an abstain. Of the windows that resolved at all, the downside took it four times out of five. But only 30% resolved in the first place, and ∅ 70% carries the rest.

This is the cleanest illustration on the page of the two separate questions the indicator asks. Which way, if it moves? Down, decisively. Will it move far enough to matter? Almost certainly not. A lean needs both answers, so the rails stay grey.

Price did sell off through the morning and then ground back up into the close without reaching the target. One presentational note: a window that reaches neither target is only marked as settled when the next session opens, and on a Friday the next session is Sunday evening. Friday's close is the effective resolution point, so this window shows as still open on a chart loaded to the end of the week.

A lean built on a thin majority

• IB range122.25 pt · avg5 252.80 · 💰 Accumulation
• Split▲ 35% · ▼ 21% · ∅ 44% (n 459)
• Shorter horizons▲ 2h 17% · 4h 25%
• Target distance▲ 122.3 pt (×1 auto)

The half hour reads Monday differently from the hour, and it is worth understanding why. Atlas mode does not ask whether the leading share is large in absolute terms. It asks two things: of the windows that resolved, did one side take enough of them, and did enough of them resolve at all. Here 35% against 21% is a 62 to 38 split of the decided races, which just clears the bar, and 56% resolved, which clears the other. So the magnet appears on a 35% rail.

It is a lean, but a thin one, and the chart is honest about that: ∅ 44% is still the largest single number on it. Price pushed above the box shortly after the window closed, stalled well short of the target, and spent the rest of the day drifting back down. The 44% branch is the one that landed.

Nearly, which counts as neither

• IB range206.75 pt · avg5 214.00 · ⚖️ Normal
• Split▲ 38% · ▼ 48% · ∅ 15% (n 371)
• Shorter horizons▼ 2h 38% · 4h 44%
• Target distance▼ 103.4 pt (×0.5 auto)

The only ⚖️ NORMAL window of the week, and the only reading where the vast majority of comparable sessions resolved: ∅ is just 15%. Something was expected to happen. The problem is the split. 38% up against 48% down is close to a coin flip once you set the undecided cases aside, nowhere near the threshold, so both rails stay grey.

Then the 15% landed anyway. Price fell hard through the afternoon and stopped just short of the lower target before turning back up. Six sessions out of seven in this table would have reached one side or the other. This one missed by a small margin, which for a race scored to the close is the same as not getting there.

The lesson is not that the table was wrong. A 15% branch lands roughly one time in seven, and a set of frequencies that never produced its unlikely outcomes would be the suspicious one. The lesson is that this is what a low ∅ buys you: better odds, not a guarantee.

One point short of a lean

• IB range148.25 pt · avg5 196.05 · ⚖️ Normal
• Split▲ 23% · ▼ 36% · ∅ 41% (n 347)
• Shorter horizons▼ 2h 20% · 4h 27%
• Target distance▼ 148.3 pt (×1 auto)

This is the closest the week came to a fourth lean. 36% down against 23% up is a 61 to 39 split of the decided races, and the threshold is 62. One percentage point separates this window from a magnet on the lower rail.

We are showing it precisely because it is marginal. A threshold has to sit somewhere, and sitting somewhere means occasionally declining a call that only just fails. The alternative, moving the bar to fit the case in front of you, is how a measurement stops being one.

As it happens the day finished in the ∅ 41% branch, the largest of its three shares, so the abstain cost nothing here. That is not the justification for the rule. The rule is worth keeping whether or not any particular near miss turns out well.

The week's one clean result

• IB range318.25 pt · avg5 192.95 · 🚀 Expansion
• Split▲ 80% · ▼ 14% · ∅ 6% (n 635)
• Shorter horizons▲ 2h 71% · 4h 76%
• Target distance▲ 79.6 pt (×0.25 auto)

318 points in thirty minutes, against a five-session average of 193 for the same window. The opening half hour ran almost vertically and the 🚀 EXPANSION tag follows. The table that comes back is the strongest of the week by some distance: ▲ 80%, ∅ down to 6%, and 71% of comparable windows resolving up inside two hours.

Price cleared the target at about 10:30, half an hour after the window closed, and the magnet-and-tick mark prints on the chart.

Now hold this beside the same morning's 60-minute window. Both leaned up. Both were right about direction. Only this one was paid, because the hour's extra range put its target higher up, and price stopped in between the two. Direction and reach are separate questions on the same chart, and the week's single result turns on the second one.

The week's only cross, on a window that offered nothing

• IB range110.75 pt · avg5 194.60 · 💰 Accumulation
• Split▲ 32% · ▼ 31% · ∅ 37% (n 367)
• Shorter horizons▲ 2h 14% · 4h 23% · ▼ 2h 16% · 4h 23%
• Target distance▲ 110.8 pt (×1 auto)

32% against 31%. It is difficult to construct a flatter table than this one. The two directions are separated by a single percentage point, ∅ is larger than either of them, and the shorter-horizon figures are almost identical on both sides as well.

Price took the lower target at about 11:20, and because the upper rail held the marginally larger share the mark prints as a ❌. That is the week's only cross, and it is worth being precise about what it means: the indicator scores the larger of the two shares whether or not that share cleared the lean threshold, so an abstained window can still be marked wrong. No lean was offered here, so nothing was risked on it.

Reading a one-point gap as a directional signal is exactly the mistake the abstain state exists to prevent.