Golden Cross and Death Cross: The Signal Wall Street Watches

After the S&P 500 flashed a golden cross in late 2023, the index went on to gain ~32% over the next twelve months — while Tesla (TSLA)'s April 2025 death cross preceded losing roughly half its value. Same two patterns, different outcomes — and most investors misread both.

Walk into any trading desk and you will hear the phrases "golden cross" and "death cross" as if they were verdicts. They are not. They are simple, lagging indicators that confirm a trend already in motion — useful for sizing decisions, dangerous as standalone signals. Here is the version of the story most retail investors do not get.

What Is a Golden Cross?

A golden cross is the moment when a stock's 50-day simple moving average (SMA) closes above its 200-day SMA. Both averages are calculated as the running mean of the prior 50 or 200 trading days. When the shorter-window line moves above the longer-window line, technicians treat it as evidence that recent momentum is strong enough to pull the long-run trend higher.

The mechanical reason: the ~50-day average reflects the past two-and-a-half months of price action; the ~200-day average reflects the past nine-plus months. When the recent past is consistently higher than the longer past, the longer trend has — by definition — already turned up. The cross is the visible scoreboard of that change.

A golden cross is not a forecast. It is a confirmation that the trend has already shifted. Treating it as a forecast is the most common mistake retail investors make.

What Is a Death Cross?

The death cross is the inverse: the 50-day SMA falls below the 200-day SMA. Recent price action has been weak enough for long enough that the longer trend is now bending down. Like its cousin, it confirms a regime change rather than predicting one.

Famous prior death crosses include the broad market index heading into 2008, the late-2018 cross, the COVID-era March 2020 cross, and the 2022 cross. Each was followed by additional downside, but each also ended with a violent reversal that reversed the cross within months. The pattern's reputation as a sell signal overstates its precision.

How Are They Calculated in Practice?

Both averages use the closing price of the prior N trading days. There is no smoothing, no weighting, no exotic math — the 50-day SMA is simply the average of the last fifty closing prices, recalculated daily. The 200-day works the same way over a longer window.

For example, if Apple (AAPL) has the last 50 closes summing to roughly $9,250, its 50-day SMA is around $185. If the last 200 closes sum to roughly $35,000, its 200-day SMA is around $175. Because the 50-day sits above the 200-day, AAPL is currently in a "golden cross" state. The signal "fires" the day this relationship first flips from one side to the other.

Most charting software draws both lines automatically. The crosses are visible to the naked eye on any year-or-longer chart of Microsoft (MSFT), Nvidia (NVDA), or Amazon (AMZN).

Do Golden and Death Crosses Actually Work?

Sometimes — and the conditions matter more than the cross itself. Backtests on the S&P 500 since 1950 show that golden crosses precede positive 12-month returns roughly 70-75% of the time, with average returns near low-double-digits. Death crosses precede negative 12-month returns at a lower hit rate near 55-60%, with wider variance.

The honest read: golden crosses are decent confirmation signals, death crosses are noisier. The asymmetry exists because equities have a long-run upward drift — any signal that confirms an uptrend has the underlying market math working in its favor.

For individual stocks, the pattern works far better when the 200-day is sloping up. A golden cross on a stock with a flat or declining ~200-day SMA is much more likely to be a whipsaw than a true regime change. Tesla (TSLA) had several "fake" golden crosses in 2022-23 during sideways chop. Walmart (WMT)'s 2023 golden cross fired at the start of a multi-quarter uptrend.

The signal is only as good as the regime it fires in. Same cross, different context, different outcome.

Why Do Most Investors Misuse These Signals?

Three errors come up over and over.

First, treating the cross as a buy/sell trigger. Golden crosses do not say "buy now" — they say "the trend has confirmed." If you are already long, that is reassuring. If you are not long, you are getting confirmation of a move you missed at least part of. The signal is most useful for sizing decisions, not entry decisions.

Second, ignoring the slope. A 200-day SMA that has been going sideways for ~6 months is a different market than one that is rising. Golden crosses inside flat regimes whipsaw frequently. The slope is the filter; the cross is the trigger.

Third, applying the signal to the wrong instrument. Golden and death crosses work best on broad indices and large-cap, fundamentally-driven stocks like Johnson & Johnson (JNJ) or Costco (COST). They work poorly on illiquid small-caps, biotech binary names, and any stock whose price is dominated by a single news event.

For a more complete framework, see our technical analysis library and the fundamental analysis guide for how to combine technicals with valuation work.

Pro Tips: How Practitioners Actually Use Them

Three habits separate practitioners who use these signals well from those who get whipsawed.

First, they require slope confirmation. A golden cross on a flat 200-day SMA is treated as a "watch" signal, not a "go" signal. Practitioners want the 200-day to be rising for ~30+ days before treating a golden cross as actionable.

Second, they use the cross as one of several confirmations. A golden cross on Procter & Gamble (PG) that lines up with rising free cash flow yield, expanding gross margins, and improving relative strength is a much higher-conviction signal than a golden cross alone.

Third, they accept that a meaningful share of crosses fail. The discipline is to size accordingly. A failed cross on a position sized at ~2% of portfolio is a learning experience. The same failure at ~10% is a portfolio-level event.

For long-term thinking, see our investment strategies guide to layer technical signals over a value or quality framework.

When Should You NOT Use This Signal?

Three situations where the cross is essentially noise.

First, in chop. If a stock has been ranging within ~10% of a midpoint for ~6 months, the moving averages will be tangled and crosses will fire and fail repeatedly. Avoid acting on either pattern in clearly sideways markets.

Second, on event-driven stocks. Earnings, drug approvals, regulatory rulings, and acquisition rumors can drive prices in ways that have nothing to do with momentum continuation. Names where a single binary outcome dominates the next 12 months will produce crosses that mean very little.

Third, on illiquid stocks. Low-volume names can produce crosses purely from one or two outsized days of trading. The signal needs ~$50M+ daily dollar volume to be statistically meaningful as a momentum read.

Ready to analyze these stocks yourself? Search any ticker on MainRatios to see valuations from 6 legendary investors - free.