Every time the Nasdaq‑100 closed 10% or more below its record since 1985, dated: how deep each fall went, how long it took to recover, and which ones became bear markets. Plus how the S&P 500 compares and where the index stood on Nov 2, 2026.
Market risk reading, cross‑asset context and an events calendar: see Exent →
The Nasdaq‑100 has had 36 declines of 10% or more from a record close since Jan 31, 1985, about 0.86 a year, or one every 14 months on average. The latest started on Nov 2, 2026, when the index closed 10.4% below its Oct 6 record. It was the third such episode of 2026; the first two crossed the 10% line on Mar 27 and Jul 29.
The average hides long gaps. From 1985 to early 2000 there were 22 episodes. Then the index fell 82.9% in the dot‑com bust and did not close above its March 2000 record again until Nov 3, 2015 (QQQ, the fund, took until 2016 on price). No new episode could start in those 15 years, because the index never made a new record to fall from. Since November 2015 there have been 14.
A correction here is a fall of 10% to 19.9% from a record close, and a bear market a fall of 20% or more. The line between them is a convention, explained in correction vs bear market.
Each row is one episode. Day counts are trading days. “Days to new record” runs from the trough to the first close above the old record.
| Prior record (close) | Crossed −10% | Trough | Depth | Days record to trough | Days trough to new record | New record | Bear market (20%+)? |
|---|---|---|---|---|---|---|---|
| Feb 8, 1985 (128.65) | Mar 25, 1985 | Oct 8, 1985 | −16.7% | 166 | 45 | Dec 11, 1985 | No |
| Jul 3, 1986 (164.51) | Jul 15, 1986 | Sep 15, 1986 | −18.0% | 50 | 87 | Jan 19, 1987 | No |
| Oct 5, 1987 (213.81) | Oct 16, 1987 | Oct 26, 1987 | −39.9% | 15 | 401 | May 26, 1989 | Yes |
| Oct 9, 1989 (236.79) | Dec 18, 1989 | Jan 30, 1990 | −16.5% | 78 | 80 | May 24, 1990 | No |
| Jul 16, 1990 (246.18) | Aug 2, 1990 | Oct 11, 1990 | −32.9% | 62 | 81 | Feb 6, 1991 | Yes |
| Apr 17, 1991 (283.38) | Jun 24, 1991 | Jun 24, 1991 | −11.5% | 47 | 36 | Aug 14, 1991 | No |
| Feb 12, 1992 (355.97) | Apr 2, 1992 | Jun 26, 1992 | −18.5% | 94 | 112 | Dec 4, 1992 | No |
| Jan 14, 1993 (381.34) | Feb 22, 1993 | Apr 26, 1993 | −14.2% | 70 | 107 | Sep 27, 1993 | No |
| Mar 17, 1994 (418.45) | Apr 4, 1994 | Jun 24, 1994 | −15.9% | 68 | 143 | Jan 18, 1995 | No |
| Sep 11, 1995 (609.75) | Oct 9, 1995 | Oct 9, 1995 | −10.3% | 20 | 18 | Nov 2, 1995 | No |
| Nov 3, 1995 (621.71) | Dec 18, 1995 | Jan 9, 1996 | −14.0% | 44 | 22 | Feb 8, 1996 | No |
| Jun 5, 1996 (699.35) | Jul 15, 1996 | Jul 23, 1996 | −14.4% | 33 | 37 | Sep 13, 1996 | No |
| Jan 22, 1997 (925.52) | Mar 17, 1997 | Apr 2, 1997 | −15.3% | 48 | 23 | May 5, 1997 | No |
| Oct 9, 1997 (1,148.21) | Oct 27, 1997 | Dec 24, 1997 | −18.2% | 53 | 37 | Feb 19, 1998 | No |
| Jul 20, 1998 (1,465.89) | Aug 4, 1998 | Oct 8, 1998 | −23.0% | 57 | 22 | Nov 9, 1998 | Yes |
| Feb 1, 1999 (2,130.93) | Feb 17, 1999 | Mar 2, 1999 | −11.4% | 20 | 19 | Mar 29, 1999 | No |
| Apr 9, 1999 (2,232.47) | Apr 19, 1999 | Apr 19, 1999 | −11.9% | 6 | 5 | Apr 26, 1999 | No |
| Apr 26, 1999 (2,260.66) | May 25, 1999 | May 25, 1999 | −11.6% | 21 | 18 | Jun 21, 1999 | No |
| Jul 16, 1999 (2,459.48) | Aug 4, 1999 | Aug 10, 1999 | −12.0% | 17 | 11 | Aug 25, 1999 | No |
| Jan 3, 2000 (3,790.55) | Jan 6, 2000 | Jan 6, 2000 | −11.9% | 3 | 8 | Jan 19, 2000 | No |
| Jan 21, 2000 (3,849.96) | Jan 28, 2000 | Jan 28, 2000 | −10.5% | 5 | 4 | Feb 3, 2000 | No |
| Mar 27, 2000 (4,704.73) | Apr 3, 2000 | Oct 7, 2002 | −82.9% | 634 | 3,292 | Nov 3, 2015 | Yes |
| Nov 3, 2015 (4,719.05) | Jan 13, 2016 | Feb 9, 2016 | −16.3% | 66 | 118 | Jul 28, 2016 | No |
| Jan 26, 2018 (7,022.97) | Feb 8, 2018 | Feb 8, 2018 | −10.2% | 9 | 20 | Mar 9, 2018 | No |
| Mar 12, 2018 (7,131.12) | Apr 2, 2018 | Apr 2, 2018 | −10.4% | 14 | 44 | Jun 4, 2018 | No |
| Aug 29, 2018 (7,660.18) | Oct 24, 2018 | Dec 24, 2018 | −23.0% | 80 | 78 | Apr 17, 2019 | Yes |
| May 3, 2019 (7,845.73) | Jun 3, 2019 | Jun 3, 2019 | −11.1% | 20 | 22 | Jul 3, 2019 | No |
| Feb 19, 2020 (9,718.73) | Feb 27, 2020 | Mar 20, 2020 | −28.0% | 22 | 53 | Jun 5, 2020 | Yes |
| Sep 2, 2020 (12,420.54) | Sep 8, 2020 | Sep 23, 2020 | −12.8% | 14 | 48 | Dec 1, 2020 | No |
| Feb 12, 2021 (13,807.70) | Mar 8, 2021 | Mar 8, 2021 | −10.9% | 15 | 23 | Apr 9, 2021 | No |
| Nov 19, 2021 (16,573.34) | Jan 20, 2022 | Dec 28, 2022 | −35.6% | 277 | 243 | Dec 15, 2023 | Yes |
| Jul 10, 2024 (20,675.38) | Aug 2, 2024 | Aug 7, 2024 | −13.6% | 20 | 64 | Nov 6, 2024 | No |
| Feb 19, 2025 (22,175.60) | Mar 10, 2025 | Apr 8, 2025 | −22.9% | 34 | 52 | Jun 24, 2025 | Yes |
| Oct 29, 2025 (26,119.85) | Mar 27, 2026 | Mar 30, 2026 | −12.1% | 103 | 11 | Apr 15, 2026 | No |
| Jun 2, 2026 (30,660.60) | Jul 29, 2026 | Jul 29, 2026 | −11.3% | 39 | 38 | Sep 22, 2026 | No |
| Oct 6, 2026 (31,224.69) | Nov 2, 2026 | Nov 2, 2026 (so far) | −10.4% (so far) | 19 (so far) | not yet | not yet | Not so far |
Nasdaq‑100 price index closes, Exent’s data, Jan 31, 1985 to Nov 2, 2026. The last row is open: its trough and depth can still change. The 2007–09 financial crisis has no row of its own. The index fell 53.7% from its Oct 31, 2007 close to Nov 20, 2008, but that fall happened while it was still below its March 2000 record, so it sits inside the 2000 episode.
The ones that stayed under 20% were usually over in a few months. Across the 27 finished corrections, the median fall was 12.1%. The index took a median 33 trading days to go from record to trough and 36 more to close at a new record: 70 trading days in all, about three months. The slowest recovery from the trough was 143 trading days, in 1994.
The 8 bear markets were a different story. Their median depth was 30.5%. Some still recovered fast: the 1998 fall of 23.0% was back at a record 22 trading days after its trough, and the 2025 fall of 22.9% took 52. Others took years. The 1987 crash needed 401 trading days, the 2022 bear market 243, and the dot‑com bust 3,292, more than 13 years from the October 2002 trough. A fall that starts as a 10% correction does not tell you in advance which kind it will be.
Recovery times for the wider market, including dividends, are in how long markets take to recover from a crash.
It is uncommon. In the 42 calendar years from 1985 to 2026, the Nasdaq‑100 crossed the 10% line at least once in 24 of them. It did so three or more times in only four: 1999 (four times), 2000, 2018 and 2026. Two or more happened in seven years. Each 2026 episode was a separate fall, because the index made new records on Apr 15 and Sep 22 in between. The earlier years ended very differently (a bubble peak in 2000, a record high in 2019), so the count says nothing about what follows.
| Decade | Nasdaq‑100 episodes | S&P 500 episodes |
|---|---|---|
| 1980s (from Jan 31, 1985) | 4 | 1 |
| 1990s | 15 | 5 |
| 2000s | 3 | 2 |
| 2010s | 5 | 3 |
| 2020s (to Nov 2, 2026) | 9 | 3 |
Counted by the year of the −10% crossing. The low count in the 2000s reflects the long wait for a new record after 2000, not a calm decade.
Same rule, same window (Jan 31, 1985 to Nov 2, 2026). The Nasdaq‑100 had about two and a half times as many episodes.
| Nasdaq‑100 | S&P 500 | |
|---|---|---|
| Episodes of 10%+ from a record | 36 | 14 |
| How often, on average | About every 14 months | About every 3 years |
| Stayed under 20% | 28 (incl. the open one) | 9 |
| Became bear markets (20%+) | 8 | 5 |
| Median depth, finished corrections under 20% | −12.1% (27) | −14.2% (9) |
| Median trading days, record to trough | 33 | 62 |
| Median trading days, trough to new record | 36 | 81 |
| Years with 3 or more episodes | 1999, 2000, 2018, 2026 | None |
| Deepest fall | −82.9% (2000–02) | −56.8% (2007–09) |
| Annualized volatility of daily returns | 25.7% | 18.2% |
| Nov 2, 2026 close vs Oct 6, 2026 record | −10.4% | −3.5% |
The S&P 500 sample is small. Three of its nine corrections stopped just short of 20% (1990, 1998 and 2018 at 19.9%, 19.3% and 19.8%), and the 2025 fall closed 18.9% below its record. So its medians move a lot with one episode. Both counts start on Jan 31, 1985, the Nasdaq‑100’s first close. This counts only falls from a record close. Counts that start from any earlier peak are higher: about one 10% fall every 18 months for the S&P 500 from 1954 to 2024, as in how often the stock market crashes.
Because it swings more. The Nasdaq‑100 holds the 100 largest non‑financial companies listed on Nasdaq, and it leans heavily on a handful of big technology and internet names. When those few stocks move together, the whole index moves. Its daily returns were about 40% more volatile than the S&P 500’s over the full period (25.7% vs 18.2% a year), and still more volatile since 2016 (22.5% vs 17.8%). A more volatile index reaches the 10% line more often.
The current episode shows the effect of concentration. From the Oct 6 record to the Nov 2, 2026 close, the Nasdaq‑100 fell 10.4% and the S&P 500 3.5%, while the equal‑weight S&P 500 fund RSP rose 3.6% (price only). The average S&P 500 stock was not falling; the biggest ones were. Why the two ways of weighting can diverge is in equal weight vs cap weight.
More volatility also cut the other way: the Nasdaq‑100’s finished corrections were slightly shallower and shorter in this sample than the S&P 500’s, partly because many were quick shakeouts that the broader index never registered. How far QQQ, the fund that tracks the index, fell in each big selloff since 2000 is in how much QQQ can fall in a crash. The full list of market‑wide falls is in the history of stock market crashes.
The Nasdaq‑100 closed at 27,989.82 on Mon, Nov 2, 2026. That is 10.4% below its record close of 31,224.69 on Oct 6, 2026, 19 trading days (27 calendar days) earlier. By this page’s rule, an episode opened on Nov 2. It ends at the next close above 31,224.69.
The S&P 500 closed at 7,541.40, 3.5% below its own Oct 6 record of 7,818.93. It is not in a correction by this rule.
These are dated facts, not a forecast. The table above shows that past episodes ended anywhere from a few days to many years later.
A 10% drop alone does not say whether risk is spreading. Members get Market intel’s read on whether risk is rising across rates, credit and the 7 cross‑asset ratios this week.
The Nasdaq-100 closed 10% or more below its record 36 times from January 31, 1985 to November 2, 2026, about once every 14 months on average. None could start from 2000 to 2015, while the index was still below its March 2000 record.
For the 27 finished corrections that stayed under 20%, the median time from record to new record was 70 trading days, about three months. The median fall was 12.1%. Episodes that became bear markets took from 22 trading days to more than 13 years to recover from the trough.
As of the November 2, 2026 close, yes by the 10% rule: the Nasdaq-100 closed at 27,989.82, 10.4% below its October 6, 2026 record of 31,224.69. It was the third such episode of 2026. The S&P 500 was 3.5% below its record.
Depth. A correction is a fall of 10% to 19.9% from a record close; a bear market is 20% or more. Of the 35 finished Nasdaq-100 episodes since 1985, 27 stayed corrections and 8 became bear markets.
This page cannot tell you that; it is history, not advice. Many investors write down a plan before a fall: how much of a decline they can hold through, what share of the portfolio sits in the biggest tech stocks, and when they rebalance. Then they check the plan, not the headline.
Market intel provides a Risk ON / Neutral / Risk OFF reading from four pillars, cross-asset analysis, an events calendar and regular written notes. Context for reviewing a plan, not a crash forecast.
Educational content, not investment advice. Examples are illustrative; past performance does not guarantee future results.