As of 09/17/2025   Indus: 46,018 +260.42 +0.6%     Trans: 15,502 -145.76 -0.9%     Utils: 1,086 +2.00 +0.2%     Nasdaq: 22,261 -72.63 -0.3%     S&P 500: 6,600 -6.41 -0.1%	YTD  +8.2%   -2.5%    +10.5%    +15.3%    +12.2%	Targets    Overview: 09/15/2025     46,700 or 44,700 by 10/01/2025   16,700 or 15,200 by 10/01/2025   1,150 or 1,050 by 10/01/2025   22,700 or 21,400 by 10/01/2025   6,750 or 6,400 by 10/01/2025
Chart Pattern Indicator: on 9/12/25

As of 09/17/2025   Indus: 46,018 +260.42 +0.6%     Trans: 15,502 -145.76 -0.9%     Utils: 1,086 +2.00 +0.2%     Nasdaq: 22,261 -72.63 -0.3%     S&P 500: 6,600 -6.41 -0.1%	YTD  +8.2%   -2.5%    +10.5%    +15.3%    +12.2%
Targets    Overview: 09/15/2025     46,700 or 44,700 by 10/01/2025   16,700 or 15,200 by 10/01/2025   1,150 or 1,050 by 10/01/2025   22,700 or 21,400 by 10/01/2025   6,750 or 6,400 by 10/01/2025
Chart Pattern Indicator: on 9/12/25

Bulkowski on the Death Cross

Initial release: 9/5/2025.

The death cross occurs when the 50-day simple moving average (SMA) crosses below the 200-day SMA in a major index. The death cross is a sell signal, that the markets are going to tumble.

Does it work and how big of a drop does the market make? This article answers these and other questions.

• Summary
• Introduction
• Serious Decline
• Lag
• Drop
• Hold Time
• Recovery
• See Also

Death Cross: Summary

Don't depend on this indicator. Here's why.

• Because of indicator lag, between 57% and 72% of the decline will have occurred before a signal triggers. It suggests you're selling closer to the bottom than the top.
• Only 29% to 30% of the time will a decline measure more than 10%.
• Eighty-five percent of the time, a death cross means price will decline less than 20% in the indices. In other words, you'd be making a mistake selling a portfolio to avoid a big decline 85% of the time.
• If you hold onto your market-based ETF, you can expect to recover your losses in a median of 3 to 4 months. Selling a long-term holding for such a short term decline puts you at risk of higher capital gains taxes.
• If you use the pair, death cross and golden cross, you will be selling low and buying high between 75% and 79% of the time. That's the opposite of what you want to do.

Death Cross: Introduction

The idea behind the death cross is to have an indicator that warns of a substantial decline. Let's take a look at an example, shown in the figure.

This is a chart of the Dow industrials on the daily scale. At D, Trump entered office as our new president. At C, the industrials crested. However, Trump's threats of imposing tariffs sent the markets plummeting, reaching a low in April 2025, E.

The blue line (which looks black) is the 50-day simple moving average (SMA) of closing prices in the Dow. The red line is the 200-day SMA.

At A, we see a death cross, where the 50d SMA drops below the 200d SMA. That was a sell signal. However, notice that the market peaked at C and it had already reached its low (E) and started to rebound. In other words, the death cross indicator flagged a sell near the bottom (E) and not the top (C).

At B, we see a golden cross, where the 50d SMA rises above the 200d SMA. That was a buy signal.

Compare to the two cross signals (death and golden). Notice that the sell price below A and the buy price above B suggests to sell low and buy high. That's the exact opposite of what you should be doing.

Death Cross: Terms

I use several terms in this article, mostly as column headers in the tables.

Hold Time Drop. This is the hold time (assuming you held onto the security or sold it short) or percentage drop from the death cross signal (A) to the lowest close, at F (F is the lowest close between A and B).

Median. It's the middle value in a sorted list of numbers. For example, the median of 3, 1, 5, 4, 2 is 3. Sort the list (1, 2, 3, 4, 5) and pick the middle value, which is 3. Half the values are higher than 3 and half will be below.

Prior High to Cross. This is the hold time or the drop from point C in the chart (the highest close within 130 days preceding a death cross signal) to A (the death cross signal).

Recovery. This is the time or percentage gain from the lowest close at F to the crossover at B. It's the time it would take before a buy (golden cross) signal occurs (the 50-day moving average crossing above the 200-day moving average).

Death Cross: Serious Decline

Table 1 is what I consider the most important table in this article. It shows the results of how often a serious decline occurs after a death cross. What is meant by serious? A 10% drop is commonly called a correction. It's worrisome and it's a warning signal to pay attention to the market. However, a 20% drop (usually measured from high to low) is serious. It signals a bear market when it occurs in the indices.

Let's discuss the results in Table 1 for the indices.

The Dow industrials, Nasdaq, and S&P 500 index see about the same size drop after a death cross. Slightly less than a third of the time (29% to 30%), the three will see prices drop 10%. About half that, 15% to 16% of the time, the indices will see price drop at least 20%.

To put it another way, about one in six death crosses will see price drop at least 20% after the sell signal. You'd be smart to ignore the signal 85% of the time.

Individual stocks have the highest percentage (45% and 27%) seeing price drop from 10% and 20%, respectively. When applied to individual stocks, the indicator flashes a bearish move about one in four trades (27% of the time). To me, that's a dreadful success rate.

Death Cross: Lag

Table 2 shows the result of indicator lag. Lag is the delay it takes an indicator to signal. For moving averages, the longer the moving average (like 200 days), the longer the lag.

With the death cross, we have one moving average crossing the other, but there is still lag. I measured that by finding where the sell signal occurred (at a crossover) and looked back 130 days (about half a year's worth of trading days) to find the highest high between those two points.

The median time from the high to the death cross was about 3 months. The signal to sell came 3 months too late (well after the high). This compares to a hold time (between a sell signal and the resulting low) of about a month.

By the time a death cross signal occurs, the market is near the bottom of the plunge.

The price decline from the peak to the death cross sell signal varied from 8% to 18%, as the table shows, depending on the market. The drop from the sell signal to the lowest low (before a buy signal occurs) varied from 5% to 8%. Again, this means most of the drop is over by the time a death cross signal happens.

Death Cross: Drop

Table 3 shows the recent price drop for data since year 2000 versus all data.

The Hold Time Drop column is the amount you would lose if you held your securities from the sell signal to the day price bottomed (measured by highest to lowest closing prices, not highest high to lowest low prices).

For example, the Dow lost a median of 6% with data going back to 1929. More recent data shows the median decline is just 2%. For the other markets, the median decline hasn't varied much over time.

Death Cross: Sell Low, Buy High

I logged the closing price at the death cross and again when the 50-day SMA moved above the 200-day SMA (golden cross) and compared the two prices. I found that you'd be selling low and buying back at a higher price between 75% and 79% of the time. That's the opposite of the type of signals you want from an indicator.

Death Cross: Hold Time

Table 4 shows the hold time (assuming you shorted at the death cross sell signal or decided to hold during the decline), in months.

For example, the first Nasdaq signal I found was in late 1971. The hold time from the sell signal to the day price reached the lowest close was a median of 0.8 months. Since year 2000, the median hold time was about the same, 0.9 months. Except for the Dow, the remainder of the table showed similar hold times.

The data suggests the signal occurs near the end of a decline and not the beginning. Table 2 (not this table) shows a comparison of the hold time from the peak to the death cross compared to the time from the death cross to the lowest close before a buy signal. It's about 3 months versus 1 month, respectively.

Death Cross: Recovery

Table 5 shows how long it takes price to recover after reaching the lowest close.

I measured the time after reaching the lowest close (after a death cross sell signal) to the next buy signal (a golden cross signal).

For example, the S&P 500 using all data since 1953 showed a recovery time of 3.9 months. Since year 2000, the recovery time is a median of 3.7 months. Except for the Dow row, the column to column results are similar to each other.

If you compare the hold time (Table 4) to the recovery time (this table), recovery takes three to four times as long as the drop. Note that this is not the amount of time to break even after a death cross sell signal.

-- Thomas Bulkowski

See Also

• Golden cross buy signal or how to buy low and sell lower.
• Buy low or buy high? Buy near yearly low for larger gains, less risk.
• Moving average study & stock trends. Buy in a downtrend, sell on the uptrend.
• Percent-down indicator. A new indicator suggests weakness ahead.
• Weinstein stages. Does Stan Weinstein's four stages work? Yes!

 

Support this site! Clicking any of the books (below) takes you to Amazon.com If you buy ANYTHING while there, they pay for the referral.
Legal notice for paid links: "As an Amazon Associate I earn from qualifying purchases."

My Stock Market Books Invest for Two-Comma Wealth Chart Patterns: After the Buy Encyclopedia of Chart Pattern, 3rd Edition Fundamental Analysis and Position Trading Getting Started in Chart Patterns, 2nd Edition Swing and Day Trading Trading Basics

My Novels Bumper's Story Orb Foresight  Head's Law Remember Me

Copyright © 2005-2025 by Thomas N. Bulkowski. All rights reserved.

Disclaimer: You alone are responsible for your investment decisions. See Privacy/Disclaimer for more information.

Some pattern names are registered trademarks of their respective owners.

Home Advertise Contact Privacy/Disclaimer ☰