QQQ Fell 6.6%, Then Traded Inside the Range. What Followed the Five Earlier Times?

Research note: QQQ split-adjusted daily bars March 10, 1999 to November 6, 2026; VXN closes from 2001. Frozen November 7, 2026. How we checked it · How we use AI.

• The first week was no different. A median +0.8% over five sessions after the inside day, against +0.5% after the 26 other 6% drops.
• The first three days were worse. A median −6.3% against +0.6%, on five cases.
• Every earlier case sat deep in a decline. On the drop day QQQ was 28.6% to 64.4% below its 52-week high. On November 5, 2026 it was 14.8% below.
• 6% drops this close to the high are rare. Four earlier episodes since 1999. Three found their low within 8 sessions; April 2000 opened the dot-com bear market.

QQQ lost 6.6% on November 5, 2026, closing at its low of the day, $647.05. The next session it opened at $658.35, traded between $653.99 and $664.52, and closed at $656.10, +1.40%. It never reached either end of the previous day’s range. The Nasdaq-100 volatility index, VXN, closed at 34.02.

My read is that the inside day itself says little. Five earlier cases look ominous on a table, but they look ominous because of when they happened, and the other 6% drops of the same years look the same. What makes this one unusual is how close to its high QQQ was when it fell, and on that question the record is short and split.

QQQ is 13.6% below its record close of $759.66 from October 6, 2026.

Five times since 1999

The rule: a QQQ close 6% or more below the prior close, then a session with a lower high and a higher low. A touch of either end fails it. That leaves five earlier sessions: May 2000, March 2001, October 2001, September 2008, March 2020. Each is followed to its low, the lowest close before QQQ’s next record close, so a crash and a bear market each run to their own bottom.

The first week was mixed. QQQ was higher five sessions later in three of the five, with a median +0.8%, and its worst intraday low inside that week was a median 9.5% below the inside day’s close. The first three sessions were weaker: higher in two, with a median −6.3%.

The longer view is grim. Four of the five fell at least 10% further before the low, and the median was −40.5%. Only March 2020 found its low within 4 sessions, 6.4% lower.

Every episode to its low

Three of the paths end at the same place: October 9, 2002, the bottom of the dot-com bear market. The May 2000 path took 604 sessions to get there, 76.8% lower. September 2008 ran 34.8% further in 37 sessions, to the November 2008 low. March 2020 was the exception: the low came 4 sessions later, 6.4% down, and QQQ set a record on June 5, 2020.

The inside day was not what mattered

The same measures on every other 6% down day tell the story. QQQ had 26 of them without an inside day the session after. Measured from that next session, the median low before the next record was −51.6%, deeper than after the inside days, and 21 of 26 fell at least 10% further. Over a week the medians were +0.5% and +0.8%. Those days cluster in 2000–2002 and 2008, so they are not independent, but they make the point: the deep lows belong to the bear markets, not to the shape of the next session.

The one place the inside days looked different was the first three sessions: a median −6.3% against +0.6%. On five cases I would not lean on that.

What makes this one different

Every earlier inside-day case came after QQQ had already fallen hard: 28.6% to 64.4% below its 52-week high on the drop day. On November 5, 2026 it was 14.8% below. Of the 31 6% drops since 1999, only two came closer to the high.

Grouped into episodes, four earlier runs of 6% drops started within 20% of the 52-week high. They went both ways.

Three of them were shocks that passed: January 2000, August 2011, March 2020. QQQ’s low came within 8 sessions and no more than 17.1% below the session after the drop. April 2000 was the other kind. QQQ fell 6% or more two times in that run, 12.6% below its high on the first, and did not see the low for 630 sessions.

How sturdy is the list?

The 6% line is a round number. At 5% the list grows to nine earlier cases, adding February 5, 2001, July 26, 2002, January 21, 2009, September 14, 2022. At 7% it shrinks to four.

The inside-day test is strict, and some sessions missed it by very little. Three sessions after a 6% drop went outside the range by less than 1%: September 22, 2000 (below the low by 0.57%), April 17, 2001 (above the high by 0.65%), December 2, 2008 (below the low by 0.19%). Highs and lows from before 2001, when prices were still quoted in fractions, differ between data sources, so the 2000 cases in particular depend on the record used.

What I take from it

The inside day after a 6% drop has the look of a signal, and the five earlier cases have the look of a warning. Read against every other 6% drop, it is neither. Four of the five came inside the two great Nasdaq bear markets, where nearly every 6% day was followed by a much lower low.

The question that matters is the one the inside day cannot answer: is this a shock near the high, like 2000, 2011, 2020, or the first break of a top, like April 2000? Those four episodes split 3 to 1, and the one that turned into a bear market looked like the others in its first weeks.

What would change my read

The range itself is the first test. A close below $647.05, the November 5, 2026 low, would break the inside day to the downside. A close above $683.65 would take back the whole drop.

The second is depth. QQQ is 13.6% below its October 6, 2026 record. A close 20% below it, near $607.73, would put this episode with the bear-market cases rather than the shocks that passed.

How we checked it

• The setup. A QQQ close 6% or more below the prior close, then a session whose high is below the drop day’s high and whose low is above its low. The rules were set before any outcome was read.
• Outcomes. QQQ returns from the inside day’s close over 1, 3 and 5 sessions and one month, the lowest intraday low of the next five sessions, and the lowest close before QQQ’s next record close.
• Baseline. Every other 6% down day since 1999, measured the same way from the session after it.
• Near the high. Every 6% down day with QQQ within 20% of its 52-week high, grouped into episodes.
• Robustness. Drop thresholds from 5% to 7%, and the sessions that missed the inside-day test by less than 1%.
• Frozen evidence. Everything on this page uses the snapshot of November 7, 2026.