Correction realities
- S&P 500 entered correction territory last week
- 21.5% of similar corrections since 1983 lasted one day
- Fewer than half became bear markets (but others came close)
It’s been a week since the S&P 500 (SPX) entered correction territory, closing more than 10% below its February 19 close. As of Wednesday, the SPX had bounced 2.8%, likely leading many traders and investors to wonder if the correction had ended the same day it began:
Source: Power E*TRADE. (For illustrative purposes. Not a recommendation.)
So far, all we know about this decline is how long it took—16 trading days, which is SPX’s fifth-quickest trip from record high to correction since 1983. While many recent market discussions have focused on corrections as possible long-term buying opportunities, a deep dive into similar moves offers several potentially useful insights into what may be next for the market.
To mirror current conditions, the table below shows what happened after the past 14 times the SPX initially closed 10% or more below a previous record high. In other words, the table does not include moves such as the -10.3% “correction” in July–October 2023 because the SPX’s July high was not an all-time high. The table details how long it took the index to fall into a correction, how much more it fell (if it did), how long it took to reach that eventual low, and how long it took to rally back to the pre-correction high:1
Source (data): Power E*TRADE and Standard & Poor's. (For illustrative purposes. Not a recommendation. Note: It is not possible to invest directly in an index. Note: “Days” refers to trading days rather than calendar days. Summary statistics for “Add’l days to hit low,” “Add’l decline, “Total decline,” and “Days to return to pre- correction high” are based on the 11 corrections that didn’t begin and end on the same day.)
Some immediate takeaways:
1. Corrections ended the same day they began in only three cases (1990, 1997, 2019).
2. A correction deepened to become a bear market (20% or more below the pre-correction high) only five times, although the SPX came within a percentage point of that threshold two other times (in 1990 and 1998).
While those who think the current correction is already over or will morph into a bear market may be correct, they don’t have the weight of historical evidence on their side. Most of the sell-offs landed between these two extremes.
Aside from one-day corrections, it took the SPX anywhere from 11 to 622 trading days (roughly two weeks to nearly 30 months) to reach the lowest low of the move. The median was 38 trading days (a little less than two months), and the wait time was longer than this benchmark only four times. The SPX’s additional decline after entering a correction ranged from -1.7% to -51.9% (median -10.7%). The total decline, from the pre-correction high to the move’s low, ranged from -12.1% to -56.8% (median -19.9%).
In terms of the journey back to all-time highs, it took the SPX anywhere from 28 to 1,788 trading days (five weeks to more than seven years), with a median wait time of 221 trading days.
Such wide-ranging numbers may not seem particularly useful for traders looking for insights into the current move, but data always benefits from context. For example, the table includes two of the most significant market events of the past 50 years—the 2000 dot-com implosion and the 2008-2009 financial crisis. While no one knows what tomorrow may bring, few market watchers are making the case that there are currently systemic risks comparable to the ones that were in play during those episodes. Similarly, the trading “circuit breakers” that were implemented after the 1987 stock market crash make it impossible for the SPX to fall 20% in one day, as it did that year.
In other words, history suggests the recent correction probably won’t morph into the type of deep, extended bear market that unfolded in 2000-2003 and 2008-2009, although such a move can’t be ruled out. The SPX fell an additional 10% or more after closing in correction territory only six of 14 times (although it came close in two other cases).
Overall, the table suggests better-than-average odds the SPX will trade below last Thursday’s close, and that it will take at least several weeks to get back to its February high. The fastest the index has ever gone from correction to new record high was 28 trading days (nearly six weeks), and that was after the one-day correction in October 1997. The next two shortest wait times were 34 and 70 trading days.
Finally, if the data appears to suggest the SPX could move sideways to lower for a while, that jibes with Morgan Stanley & Co. strategists’ previous forecasts for a choppy, rangebound market in the early part of the year, as well as their more recent observations that the recent bounce probably wouldn’t turn into a more durable rally.2
Today’s numbers include (all times ET): weekly jobless claims (8:30 a.m.), Philadelphia Fed Manufacturing Index (8:30 a.m.), current account (8:30 a.m.), existing home sales (10 a.m.), Leading Economic Indicators Index (10 a.m.), EIA Natural Gas Report (10:30 a.m.).
Today’s earnings include: Dollar General (DG), Duluth (DLTH), DocuSign (DOCU), Ulta Beauty (ULTA).
Click here to log on to your account or learn more about E*TRADE's trading platforms, or follow the Company on Twitter, @ETRADE, for useful trading and investing insights.
1 All figures reflect S&P 500 (SPX) daily closing prices, 1983-2025. Summary statistics for “Add’l days to hit low,” “Add’l decline, “Total decline,” and “Days to return to pre- correction high” are based on the 11 corrections that didn’t begin and end on the same day. Supporting document available upon request.
2 MorganStanley.com. Is the Correction Over Yet? 3/17/25.