News outlets occasionally highlight terms like "golden cross" and "death cross" in the cryptocurrency space, signaling potential market reversals—upward or downward, respectively. Technical analysis enthusiasts view these crossovers as validation of their market outlook and a cue to enter long positions.
The last bearish pattern occurred in late autumn of the previous year. Over the following three months, Bitcoin lost around 25% of its value. Since then, part of the crypto community has been eagerly anticipating a reverse signal, indicating a trend shift and the end of a prolonged recession.
This article explores the effectiveness of these cross patterns, their reliability for trading strategies, and what market cyclicality suggests about the current bear market.
Nearly 300 Days Under the Cross
Six weeks after Bitcoin's all-time high (ATH) above $126,000, the market received a signal interpreted by many as confirmation of a downward trend. In mid-November 2025, the 50-day moving average (MA) fell below the 200-day MA, while the price remained slightly above the psychological threshold of $100,000.
"Death cross" on the BTC/USDT chart from Binance. Source: TradingView.Since then, 277 days have passed without an occurrence of the opposite crossover, known as the "golden cross."
The current market phase has become one of the longest in cryptocurrency history. Only three previous instances saw prices remain "under the cross" for longer: after signals in March 2018 (389 days), January 2022 (388 days), and September 2014 (314 days). The historical median duration of bearish trends following such crossovers is 91 days.
Duration of Bitcoin's bearish moving average crosses, 2011–2026. Source: Coin MetricsThree of these "crosses" were negated by a subsequent crossover in less than two months:
- The signal from April 2025 lasted 45 days;
- September 2023 signal—49 days;
- March 2020 signal—56 days.
Meanwhile, three others extended beyond ten months, coinciding with significant bear markets, while the remaining five ranged from 79 to 141 days.
This raises the ongoing debate about the veracity of the cross on the chart. The most popular technical analysis pattern often appears the same at the outset, with the distinction between a prolonged sideways or bearish market only becoming clear in hindsight.
By the end of August 2026, the faster 50-day indicator line is projected to be at $65,000, while the slower 200-day MA will be around $69,000, narrowing the gap to approximately 7%. These lines began converging in spring, but until recently, this occurred with minimal price involvement.
As summer concluded, Bitcoin saw a gain of about 25%, surpassing the $75,000 mark for the first time since early June. CryptoQuant identified this as the strongest short squeeze since November 2024. Analysts noted that the rise was driven by the closure of short positions and derivatives rather than spot purchases, raising questions about the sustainability of the momentum.
Short and long pressure ratio on Binance. Source: CryptoQuant.SMA vs. EMA
Both types of crosses are constructed from two lines that are calculated slightly differently. The simple moving average (SMA) sums the closing prices and divides by the number of days in the period. For the 200-day MA, this means averaging seven months of price history where a price from six months ago carries the same weight as yesterday's price.
Conversely, the exponential moving average (EMA) weighs more recent data more heavily, allowing it to respond more quickly to market changes.
The difference between these approaches is significant. In January 2026, some reports claimed a "golden cross" had formed, referring to the EMA crossover of the 50 and 200 periods, while the simple averages remained in a bearish configuration, as they still do. Readers were presented with a clickbait headline amidst a less-than-optimistic technical analysis landscape.
However, the practical difference between the methods is minimal. Trading based on either signal from historical data since 2015 has underperformed compared to passive asset ownership. Yet, the drawdown depths for "cross" strategies are lower: 72% and 76% versus 84% when simply following a Buy & Hold approach.
Comparison of simple and exponential moving averages as trading signals, 2015–2026. Source: Coin Metrics.Over 16 years of Bitcoin's price history, simple averages have produced 25 crossovers, while exponential ones yielded 23. The methods are nearly identical in quantity, but their reliability differs.
With the EMA, every third signal reverses, while with the SMA, it happens every fourth time. For example, on April 18, 2018, the EMA lines generated a "golden cross" that was canceled by May 24, while the simple averages did not cross.
A less reliable "golden cross" from April-May 2018. Source: TradingView.Half the Signals Mean Nothing
Analyzing all thirteen "death crosses" since 2011 yields results that do not align with the dramatic name of the pattern.
After three months following the signal, Bitcoin was lower in only six of the thirteen instances. The median price change over the same period was +25.7%.
Bitcoin's performance 90 days after bearish moving average crossovers, 2011–2026. Source: Coin Metrics.Thirty days post-signal, the scenario is no better: declines occurred in seven of the thirteen cases. The same pattern holds after six months. The so-called "death cross" historically predicts direction with about the same accuracy as a coin flip.
Some instances explain why certain analysts view the signal as a buying opportunity. The crossover on September 12, 2023, at a price of $25,135 preceded a 74% increase over three months and a 172.5% rise over six months. The August 2024 signal yielded a +25.7% return over the quarter, while April 2025 resulted in +37.5%.
All three signals coincided with local minima: those who closed positions at the crossover sold at the lowest correction point, while buyers capitalized on double-digit growth. After such episodes, the pattern gained a reputation as an indicator best interpreted in reverse.
When the Signal Worked
There are only two confirmed instances of the indicator working correctly in the sample.
The crossover on September 5, 2014, at a price of $489 preceded a 23.2% decline over the quarter and a 42.3% drop over six months. This signal occurred during a prolonged bear market that ultimately drove Bitcoin to a low of $176 in January 2015.
The second instance is even more telling. On January 16, 2022, the indicator lines converged at a price of $43,207, followed by the Terra crash, the bankruptcy of Three Arrows Capital, and the collapse of FTX. Six months later, Bitcoin was valued at 52.4% less, and a year later, it was down 51.7%. In this case, the indicator was effective, although the "warning" merely confirmed an ongoing decline: from the November 2021 peak ($67,542), the price had already dropped by over a third by the time of the crossover.
The signal from last year added to a modest track record: -6.8% after one month, -25.9% after a quarter, and -16.1% after six months. The indicator showed effectiveness for the first time in several years—just as the market had started to disregard it.
Not Just Cryptocurrencies
The signal shows ambiguous results in the stock market as well. According to Reuters' analysis using LSEG data, the S&P 500 index formed 24 "death crosses" over the last fifty years, with the signal appearing after the market had already experienced a significant intraday decline in 54% of cases. Adam Turnquist, the chief technical strategist at LPL Financial, noted that historically, buying in response to such patterns has been more profitable than selling.
On the other hand, Schaeffer's Research reported that following 34 signals since 1950, the index showed returns below average after six months and a year.
Bitcoin's price moves more vigorously than many stocks, resulting in a higher cost of error: a misread signal in the stock market might cost an investor a few percentage points, while in the cryptocurrency market, it could mean tens of percentage points.
In this context, the bullish version of the signal appears much more convincing. The challenge lies in its delayed appearance.
Signal with a Two-Month Lag
On July 13, 2014, the second "golden cross" in Bitcoin's history formed on the daily chart. The price was $637—73.6% above the April low. This bullish signal coincided with a local peak: the following year, the asset traded only below this level, dropping to $176.
Such a mistake was not repeated by the indicator afterward. However, it has consistently shown a lag—just the extent varies.
Calculating all twelve bullish crossovers since the first one in February 2012 reveals a consistent pattern. The signal typically appears around 60 days after the actual minimum phase, when the price is already 54% above the bottom. By this time, the market has usually completed about one-fifth of the journey from the minimum to the annual maximum.
Proportion of the journey from minimum to annual maximum completed by the time of a "golden cross," 2012–2025. Source: Coin Metrics.For conservative market participants, missing part of the movement isn't a death sentence. The median growth from the crossover point to the peak over the next 12 months is +122%. While the indicator misses the initial impulse, it captures most of the ensuing rise.
Conversely, there is a downside. In the first 90 days following a "golden cross," the price has, on average, decreased by 11.2%. In some cases, the decline has been much steeper. The February 2020 signal at a price of $10,195 resulted in a 51.4% drop amid the COVID-19 pandemic, but a year later, Bitcoin's price was five times higher.
Variability Matters More Than the Average
Behind the median of 60 days lie extremes: in some cycles, the indicator lagged the bottom by just over a month, while in others, it was half a year. The increase over the market's minimum varied from 29.8% to 177.4%.
An almost ideal crossover occurred on October 29, 2015: 37 days after the bottom, with the price just 34.2% above the lowest extreme, followed by a 152% increase. The October 2024 signal at $69,863 operated similarly—52 days after the bottom, with the price above the low by 29.8%, leading to a further rise of 78.7% by the October 2025 peak.
In contrast, the indicator performed poorly in spring 2025. The lines converged on May 23 at a price of $111,477, 44 days after the minimum of $76,351. The pattern appeared early, but by then, the market had already traversed 72.5% of the way to its peak. Only 12% remained until the maximum, and a year later, Bitcoin was trading roughly a third lower than the level where the signal emerged.
Confirmation, Not Prediction
The "golden cross" indicates a trend reversal that has already occurred rather than predicting the future. In three out of four cases, the fixation proves accurate, making it more reliable than its bearish counterpart. However, confirmation usually arrives after many market participants have already recognized the trend change without needing moving averages.
Thus, this signal is not particularly effective for entering a position. However, it serves as a good verification tool: if a reversal is already visible from other data, the crossover affirmatively confirms it.
The question remains: when can we expect the next "bullish" confirmation?
Arithmetic Instead of Forecasting
By August 2026, the 200-day MA will exclude significant February and March prices above $80,000. Each such replacement pulls the line down—on its own, without market involvement.
The faster line, assuming prices remain unchanged, stays in place, and the averages converge mechanically.
If we assume prices will hold at the August level without moving until the year's end, a "golden cross" will still form in December—not due to a trend change, but because the "expensive" winter prices will exit the calculation window.
On the flip side, a price increase accelerates the convergence of the averages. If the price stabilizes around $77,000, the MA crossover will occur on September 11; at $75,000, it will happen on September 15; if it drops to $66,000, the crossover won't occur until December 1. If there is a correction and stagnation below $60,000, no signal will appear at all, as the faster line simply won't catch up with the declining longer one.
Projected timing for the formation of the "golden cross" under different price scenarios. Source: Coin Metrics.How to Strengthen the Signal
The crossing of the averages is merely a derivative of price history. Metrics based on holder behavior and coin cost can provide a more compelling picture of the current market state.
VanEck tracks twelve such indicators, of which eight were in the capitulation zone as of August 18, 2026. For example:
- The MVRV Z-Score indicator fell to 0.37—indicating a "green zone" of deep overselling;
- The Puell multiple, comparing current miner revenues to the average over the past 365 days, shows 0.67;
- The aSOPR indicator remains at 0.989: a value below one means that market participants are, on average, closing positions at a loss.
A cluster of "bullish" signals supports the notion that the bulk of the sell-off is behind us. However, it does not promise a quick recovery—VanEck's estimates suggest Bitcoin's returns in the first six months following such periods remain below average.
The logic behind the aforementioned metrics mirrors that of the crosses, only based on different data. Indicators effectively show where the market currently stands but poorly predict future movements.
Additionally, over the past 30 days, long-term holders have reduced their holdings by approximately 356,000 BTC (to 11.84 million BTC), and their share in circulation has dipped below 60% for the first time in several months. This profit-taking after a nearly 50% drop from the completed accumulation phase seems uncharacteristic.
What History Teaches Us
The market has another benchmark that has little to do with price dynamics and on-chain metrics: the comparable duration of cycles.
For instance:
- The decline following the December 2013 peak lasted 406 days;
- The 2017–2018 cycle lasted 364 days;
- The crash in 2021–2022 lasted 366 days.
Given that bearish phases historically last about a year, counting from the last ATH places the minimum of the current cycle between October 5 and November 16, 2026. At the time of writing, the decline has persisted for 319 days.
A significant caveat: Bitcoin tested a minimum of $58,525 on July 1, marking the 268th day of the phase. If this level is indeed the cycle's bottom, the reversal occurred earlier than historical timelines would suggest. However, this hypothesis can only be confirmed with time, similar to the MA/EMA situation.
Duration and depth of Bitcoin's bearish phases across cycles. Source: Coin Metrics.VanEck estimates the average duration of the last three bear markets at 12.7 months and predicts the start of an accumulation phase between September and November.
Galaxy Research anticipates a bottom in the fourth quarter at a price range of $40,000–46,000. Analysts describe the October peak as the most "modest" in the asset's history, noting that the rise to it occurred without the widespread excitement characteristic of previous bullish phases. Therefore, the market bottom may not be as deep compared to past cycles.
Early signs of phase change are already visible. For instance, CryptoQuant notes a restoration of spot demand to February levels. The indicator nearly transitioned from negative to positive territory. In previous cases, such a transition yielded a median increase of 18.1% over 60 days with a 78% probability. However, analysts regard the improvement in market conditions as a tailwind that raises the likelihood of growth but does not guarantee a trend reversal.
"The Cycle is Dead"
The current price drop from the ATH is not deep enough to confidently assert that the cycle's bottom has been reached.
At the July minimum, the decline relative to the cyclical peak was approximately 53%. In past bear markets, the depth of decline ranged from 77% to 85%—a significant difference.
Analysts from VanEck attribute the "shallowing" of phases to the emergence of U.S. spot ETFs, increased institutional participation, and the absence of systemic collapses like Celsius, Three Arrows Capital, and FTX.
Bitwise's Chief Investment Officer, Matt Hougan, stated that cycles tied to halving events are no longer relevant. The reasoning is straightforward: over 90% of all coins have already been mined, so the next reward reduction for miners hardly alters the supply-demand balance. Previously, halving events significantly restricted the influx of new coins into the market, but that influx is now already minimal.
However, the institutional demand that supports this framework works both ways: miners continue to sell off reserves to fund their transition to AI infrastructure.
***
Technical analysis indicators cannot predict the future like a magic crystal ball. The bearish crossover guesses the direction in half of the cases, while the bullish one often appears a couple of months after the price trend reversal.
Moving averages do not initiate a new market phase. They merely confirm what is already visible from holder behavior, inflows into funds, and the "calendar" of price cycles. For hypothesis verification, a selection of indicators may suffice, but for precise forecasting, it falls short.
The diminishing drawdown from the ATH indicates a fundamental shift in the structure of the maturing market: volatility is decreasing, behavior is becoming more predictable, and institutional participation is rising. A decline of 53% instead of the typical 80–90% is likely not a sign of an incomplete phase but a new norm for an asset with a market capitalization of around $1.5 trillion.
Dashboards with on-chain metrics can assist in decision-making, but relying solely on them is unwise. The two moving averages with their attention-grabbing names are no exception.