A golden cross is when the S&P 500's 50-day moving average climbs above its 200-day average; a death cross is the reverse, and on the S&P 500 neither has been much of a crystal ball. In Cboe's index history since 1975, the S&P was higher 12 months after a death cross 76% of the time, against 78% for all sessions. Golden crosses did better, at 84%. That gap is real but modest, and it rests on just 25 signals of each kind. The cross describes the trend you already had. It does not call the next one.
What the two moving averages actually measure
A simple moving average (SMA) is plain arithmetic. Add the last 50 closing prices and divide by 50, and you have the 50-day SMA. Do the same with 200 closes and you have the 200-day. Each session, the newest close joins the window and the oldest one drops out.
As of the October 30, 2026 close, the last 50 S&P 500 closes summed to 384,854.04. Divide by 50 and you get a 50-day SMA of 7,697.08. The last 200 closes summed to 1,461,951.43, for a 200-day SMA of 7,309.76.
In that window, the oldest close in the 50-day average was August 21, 2026's 7,674.37. The next session's close replaces it. If the new close is higher than the one leaving, the average rises. If it's lower, the average falls. The 200-day works the same way, but each close is only 1/200th of the average, which is why it turns like an oil tanker.
- Golden cross: the 50-day SMA moves from below the 200-day to above it.
- Death cross: the 50-day SMA moves from above the 200-day to below it.
Some chart platforms use exponential averages or intraday data, so cross dates can differ from one screen to the next. Everything below uses simple averages of daily closes.
The scorecard: 25 golden crosses, 25 death crosses
We ran the test on Cboe's S&P 500 closing history. The first date with 200 closes to average was October 15, 1975. From then through the October 30, 2026 close, the 50-day crossed the 200-day 50 times: 25 golden, 25 death. For each cross, we measured the S&P's price change over the next 63, 126 and 252 sessions (about 3, 6 and 12 months). Then we compared those results with the same windows starting from every session since October 1975.
| Forward window | After golden cross (median / % up) | After death cross (median / % up) | All sessions (median / % up) |
|---|---|---|---|
| 3 months (63 sessions) | +3.86% / 76% | +5.09% / 68% | +2.96% / 68% |
| 6 months (126 sessions) | +6.03% / 88% | +6.39% / 68% | +5.51% / 73% |
| 12 months (252 sessions) | +13.01% / 84% | +11.74% / 76% | +11.64% / 78% |
Three things stand out.
- Golden crosses had better odds. The S&P was up 12 months later 84% of the time (21 of 25), against a 78% baseline. At six months it was 88% (22 of 25) against 73%.
- Death crosses weren't a reliable sell signal. At 12 months the index was higher 19 of 25 times. The median gain after a death cross, +11.74%, was in line with the all-sessions median of +11.64%. It was above the baseline at three and six months too.
- The damage sits in the tail. The mean 12-month return after a death cross was +9.69%, below its +11.74% median. A few deep losses pull the average down. Those are the cases traders remember.
Two caveats matter. First, these are price returns with dividends excluded. Second, the windows overlap: the baseline counts nearly every session, so neighboring days share most of their future, and some cross windows overlap each other too. With 25 signals per side, one or two outcomes can move a percentage by four points. Read this as a rough historical tendency, not a statistical law.
Why the cross lags, and how it whipsaws
A 200-day average is built to move slowly. By the time the 50-day sinks under it, the market has usually already fallen. At every death cross since 2000, the S&P was already below its highest close of the prior 252 sessions, by anywhere from 5.0% (July 2006) to 22.4% (March 2020).
The two most recent sell-offs show the problem:
- 2020: The S&P closed at 3,386.15 on February 19, 2020 and bottomed at 2,237.40 on March 23, 2020, down 33.9%. The death cross printed on March 30, 2020, five sessions after the low. Twelve months later the index was up 50.7%.
- 2025: The S&P fell from 6,144.15 on February 19, 2025 to 4,982.77 on April 8, 2025, an 18.9% drop. The death cross arrived on April 14, 2025, four sessions after the low. The 12-month return from there was +30.3%.
The lag cuts the other way too. The golden cross on July 1, 2025 came with the S&P at 6,198.01, about 24.4% above that April 8 low. By then the index had already climbed back above its February 19, 2025 close of 6,144.15.
Then there are whipsaws, where a cross reverses almost at once. Four of the 50 crosses were undone within 30 sessions. In November 1986 and November 1999, a death cross flipped back to golden after just five sessions. A golden cross on December 21, 2015 turned into a death cross 13 sessions later, on January 11, 2016. Across all 50 signals, the median gap between crosses was 203 sessions. Quick flips are uncommon, but they come without warning.
Every death cross since 2000
This table settles the "death cross means a crash" argument. "Already off 1-yr high" is the distance from the prior 252-session closing high; "worst close" is the lowest close over the next 252 sessions.
| Death cross | S&P close | Already off 1-yr high | Worst close, next 12 months | 12-month change |
|---|---|---|---|---|
| Oct 30, 2000 | 1,398.66 | -8.4% | -30.9% | -21.2% |
| Aug 18, 2004 | 1,095.17 | -5.4% | -0.4% | +11.4% |
| Jul 19, 2006 | 1,259.81 | -5.0% | -1.5% | +21.8% |
| Dec 21, 2007 | 1,484.46 | -5.2% | -49.3% | -41.3% |
| Jul 2, 2010 | 1,022.58 | -16.0% | 0.0% | +31.0% |
| Aug 12, 2011 | 1,178.81 | -13.6% | -6.8% | +19.1% |
| Aug 28, 2015 | 1,988.87 | -6.7% | -8.0% | +9.6% |
| Jan 11, 2016 | 1,923.67 | -9.7% | -4.9% | +17.9% |
| Dec 7, 2018 | 2,633.08 | -10.2% | -10.7% | +19.1% |
| Mar 30, 2020 | 2,626.65 | -22.4% | -5.9% | +50.7% |
| Mar 14, 2022 | 4,173.11 | -13.0% | -14.3% | -6.7% |
| Apr 14, 2025 | 5,405.97 | -12.0% | -4.6% | +30.3% |
Two of the 12 caught genuine bear markets early. The October 2000 cross came before a further 30.9% slide within a year, and the dot-com bear didn't bottom until October 9, 2002, at 776.76. The December 2007 cross came about 5% off the October 2007 peak, ahead of a further 49.3% slide within the next 252 sessions.
The 2022 cross sits in between. The S&P fell 25.4% from its January 3, 2022 close to its October 12, 2022 low. But the index was already 13.0% down when the cross printed on March 14, 2022, so the signal caught only about half of the decline.
The other nine were followed by gains a year out, several of them large. In five of those nine, the worst close over the next 12 months was less than 5% below the cross-day close.
Where the S&P 500 stands at the October 30, 2026 close
The current regime is golden. The S&P's 50-day SMA rose above its 200-day on July 1, 2025, and as of the October 30, 2026 close it had stayed there for 336 sessions.
| Measure (October 30, 2026 close) | Level |
|---|---|
| S&P 500 close | 7,580.07 |
| 50-day SMA | 7,697.08 |
| 200-day SMA | 7,309.76 |
| 50-day vs 200-day | +5.30% |
| Close vs 200-day | +3.70% |
| Close vs 50-day | -1.52% |
The index closed below its 50-day but well above its 200-day. The gap between the two averages narrowed every session from October 19, 2026 (5.90%) to October 30, 2026 (5.30%). From the July 1, 2025 golden-cross close of 6,198.01 to the October 30, 2026 close, the S&P gained 22.3% on price. None of this says when or whether the next cross comes.
Traders who track volatility alongside trend often pair the cross with the VIX. Our breakdown of what a VIX above 20 has meant historically and our explainer on VIX term structure cover that side of the dashboard.
What the cross can and can't tell you
On an index, the cross is a regime label. It sorts the tape into "trend up" and "trend down" after the fact, and it keeps you from arguing with a market that is clearly going one way. It is not a timing tool. It flashed within a week of the closing low in both March 2020 and April 2025, and in 2022 it arrived with the index already 13.0% off its high.
On single stocks the same arithmetic applies, but the signals come more often and with more noise. The setup is the same whether it's a golden cross on a large-cap like $CSCO or a stock building a tight base around its 200-day. The cross starts the research. Volume, earnings and breadth decide whether it holds. More desk-level takes are on the Wall Street War Room homepage.
Desk Notes
What the numbers say: Since 1975, golden crosses on the S&P 500 were followed by 12-month gains 84% of the time vs a 78% baseline. Death crosses came in at 76%, and two of the 12 since 2000 came before real bear markets.
What they don't: 25 signals a side, overlapping windows and price-only returns make this a loose tendency. The cross can't tell you whether the next death cross is a 2007 or a 2025.
FAQ
What is a golden cross in stocks?
It's when a stock's or index's 50-day simple moving average of closing prices rises above its 200-day average. The S&P 500's most recent golden cross was on July 1, 2025, and it was still in place at the October 30, 2026 close.
Does a death cross mean the market will crash?
Not historically. Of the 25 S&P 500 death crosses since 1975, 19 were followed by a higher index 12 months later. Since 2000, the October 2000 and December 2007 crosses preceded deep bear markets. Most of the others did not.
How often does the S&P 500 get a golden or death cross?
There were 50 crosses from October 1975 through October 30, 2026, about one a year on average. The median gap between them was 203 sessions, but four reversed within 30 sessions.
Why do different charts show different cross dates?
Platforms differ in whether they use simple or exponential averages, closing or intraday prices, and the price index or a total-return series. This analysis uses simple averages of S&P 500 daily price closes from Cboe.