What Happens to Stocks After the VIX Spikes Above 30?

Most of the time the S&P 500 was higher a year later, but not reliably: in the 21 episodes since 1990 with a full year of data, the index rose a median 15.1% over the next 12 months and 14 were positive (daily index closes through Oct 29, 2026). The other seven fell, by as much as 37.5%. A 22nd episode, in March 2026, is too recent to have a 12‑month result.

For comparison, the median 12‑month price return across all trading days from 1990 was 11.9%, and 80.8% of those 12‑month windows were positive. So the median after a spike was higher than on an ordinary day, but the mean (7.1% vs 10.2%) and the share of positive years were lower. A VIX spike came before some of the best years in the sample and some of the worst.

The VIX is Cboe’s index of the S&P 500’s expected volatility over the next 30 days, implied by option prices. A close above 30 is uncommon: it happened on 7.9% of trading days since 1990.

The table lists the 21 episodes with a complete 12‑month outcome. Each row is one episode: the first VIX close above 30 after at least 60 trading days (about three months) with every close at or below 30. Returns run from the S&P 500 close on the spike day. “Episode peak” is the highest VIX close before the VIX settled back for 60 days.

S&P 500 index price return, dividends not included. “Lowest close within 12 months” is the deepest S&P 500 close in the 252 trading days after the spike, measured against the spike‑day close. This measures the lowest close relative to the starting close, not maximum peak‑to‑trough drawdown. “Never below” means no later close fell below the starting level. One additional qualifying episode occurred in 2026. Its 252‑trading‑day outcome is incomplete, so it is omitted from the episode table.

By the median, yes, at every horizon; by the mean and the share of positive outcomes, mostly not. The table compares the same 21 completed episodes with every trading day since 1990.

The means tell a less flattering story than the medians: the mean 12‑month return after a spike was 7.1%, below the all‑day mean of 10.2%, because the losing years were large. With 21 completed cases, one or two episodes move these figures a lot, and a higher median does not mean the spike caused the gains that followed.

The tested quiet‑period rules produced similar 12‑month medians, but different sample sizes and positive‑outcome rates. Requiring only 20 quiet days gives 34 episodes and a 12‑month median of 15.5%; requiring 120 gives 18 episodes and 15.5%; requiring a full year gives 8 episodes and 15.7%. The share positive ranged from 64% to 75%.

The losses clustered in a few bear‑market periods. Six of the seven 12‑month losses began in 2000–2001 or 2007–2008, around the recessions associated with business‑cycle peaks in March 2001 and December 2007, as dated later by the National Bureau of Economic Research. The seventh was Dec 1, 2021, ahead of the 2022 bear market. A first qualifying close above 30 did not identify the eventual market low, and the VIX often spiked again later.

The two deepest paths: after the Nov 12, 2007 spike, the S&P 500 closed as much as 41.0% lower within a year; after Sep 15, 2008, as much as 43.3% lower, at the March 9, 2009 low. Neither was visible from the VIX level on the day. Both spikes closed near 31, close to the median spike of 31.66. How past declines unfolded is covered in the history of stock market crashes.

A related pattern: across every day the VIX closed above 30 (731 days with a full year ahead), the median 12‑month return was 22.6%. This daily‑observation sample has a different weighting: long episodes contribute many overlapping observations. Its 22.6% median is not directly comparable to an equally weighted sample of first spikes, and the 731 days are not independent results. No one could know at the time which of those days was near the low.

Usually well below the spike‑day close: the median lowest close within 12 months was 11.2% below it. In 12 of 21 completed cases the index closed more than 10% lower at some point, and in 6 it closed more than 20% lower. Only once, after Oct 27, 1997, did it never close below the spike‑day level.

The spike itself usually came after a decline was already under way. On the median spike day the S&P 500 was 11.2% below its high of the previous 12 months. For what a typical decline looks like, see how big a normal drawdown is and maximum drawdown.

Usually only days: in the median of the 21 episodes the VIX closed back at 30 or lower within 2 trading days, and 13 of the 21 did so within 5. Within an episode, the median number of closes above 30 was 16, because many spikes faded and returned.

The long ones were the crises. After Sep 15, 2008 the VIX stayed above 30 for 170 straight trading days and closed above 30 on 186 days in that episode, peaking at 80.86 on Nov 20, 2008. In 2020 it took 50 trading days to close back at 30 or lower after Feb 27, with the record close of 82.69 on Mar 16. Nine of the 21 episodes peaked above 40 and three above 50.

The VIX closed at 28.18 on Nov 5, 2026, up from 16.61 on Nov 4, the day the S&P 500 fell 4.36% in the November 2026 stock market selloff. Its high of the day was 28.36, so it did not reach 30 and no new episode has started under this page’s rule. Its highest close of 2026 is still 31.05 on Mar 27, a qualifying spike whose 12‑month outcome is not complete yet; it closed above 30 on two days that month, the last on Mar 30. Its lowest close of the year so far was 14.21 on Sep 22.

More than 60 trading days have passed since Mar 30, so the next close above 30 would start a new episode. Its 12‑month result would not enter the table above until it is complete.

These are dated readings, not a forecast. Other measures of market risk, from valuations to concentration, are collected in how risky the stock market is right now. Ways investors protect a portfolio before or during high volatility are in how to hedge a portfolio and VIX hedging; what tends to go wrong in a sell‑off is in what to do when the market crashes.