Elliott waves in crypto: testing the count on Bitcoin's rise to $69,000

Elliott waves describe price moves as a fixed sequence: five waves run in the direction of the larger trend, and three waves correct it. The theory thus provides a grid into which you can sort any chart, including that of Bitcoin. Whether this grid predicts prices is still disputed today, and that is exactly where this guide comes in.

We traced the theory on real Bitcoin prices from March 2020 to November 2022. The result: one count meets all three basic rules in hindsight. A second count meets them as well, but reaches the opposite conclusion in April 2021. Anyone using Elliott waves in crypto trading therefore needs above all a tool that documents counts cleanly. Our comparison shows which crypto analytics platforms offer wave counting, Fibonacci tools and price alerts.

Elliott wave theory: five impulse waves and three corrective waves

Elliott wave theory distinguishes two building blocks. An impulse wave is a move in the direction of the trend made up of five sub-waves, numbered wave 1 to wave 5. Waves 1, 3 and 5 drive the price forward, while waves 2 and 4 are countermoves within the trend. A corrective wave is the countermove after a completed impulse; it consists of three waves labelled A, B and C.

Together, five waves and three waves form the eight-wave cycle, the basic pattern of Elliott wave analysis. Each wave in turn consists of smaller waves of the same kind. A wave 3 on the weekly chart consists of five smaller waves on the daily chart, and each of those again consists of five or three sub-waves on the hourly chart. Elliott called these levels wave degrees. Today this is described as a fractal structure: the pattern looks the same in every time frame; only the scale changes.

Behind this lies an assumption about how market participants behave. Sentiment swings between confidence and fear, and because people react similarly in groups, price patterns repeat. For the crypto market the idea is an obvious one: Bitcoin trades around the clock, a large share of volume comes from retail investors, and shifts in sentiment feed through to the price faster than with a blue-chip stock.

Ralph Nelson Elliott and the wave principle of 1938

Ralph Nelson Elliott was an American accountant who had to retire after an illness and in the 1930s studied decades of price data from the US stock market. His book The Wave Principle appeared in 1938, followed by Nature’s Law in 1946. Elliott built on Charles Dow's Dow theory, which already described trends as a sequence of moves and countermoves, and arranged these moves into fixed counting patterns.

After Elliott's death in 1948, only a few market analysts kept the wave principle alive. It only became widely known again in 1978 with the book Elliott Wave Principle by A. J. Frost and Robert Prechter, which is still regarded as the standard work. The three rules and the guidelines analysts count with today largely go back to this version.

The method was applied to crypto as soon as Bitcoin had enough price history. On analytics platforms, drawing tools for wave counts are now standard equipment. That says something about how widespread the method is, but nothing yet about its accuracy.

Three fixed rules of Elliott wave counting

Elliott analysts distinguish between rules and guidelines. A rule may never be broken; if it is, the count is wrong and has to be redone. For an impulse wave there are exactly three rules:

1. Wave 2 never falls below the start of wave 1. Wave 2 may retrace at most 100 percent of wave 1.
2. Wave 3 is never the shortest of the three driving waves 1, 3 and 5. It is usually the longest.
3. Wave 4 does not enter the price territory of wave 1. In an uptrend, the low of wave 4 stays above the high of wave 1. Diagonals, a rare special form of impulse, are the exception.

Everything else is a guideline: how deep wave 2 typically corrects, that waves 2 and 4 often alternate in form, how long a wave 5 should be relative to wave 1. Guidelines describe what is usual, not what is mandatory. This very distinction makes the method flexible, and it is also its biggest problem.

For corrective waves, the theory knows several forms. The zigzag is a sharp correction in a 5-3-5 pattern, meaning a five-part wave A, a three-part wave B and a five-part wave C. The flat runs sideways in a 3-3-5 pattern, and the triangle contracts in five sub-waves. On top of that come combinations of two or three of these forms.

Fibonacci ratios as guide values for wave lengths

The Fibonacci sequence starts with 1, 1, 2, 3, 5, 8, 13, 21, and each number is the sum of the two before it. Dividing a number by its predecessor gives a result that approaches 1.618, the golden ratio. Elliott analysts use ratios derived from this as guide values: wave 2 often corrects 50 or 61.8 percent of wave 1, wave 3 often reaches 1.618 times wave 1, and wave 4 tends to correct 38.2 percent of wave 3.

A Fibonacci retracement is the percentage pullback of a move, measured against these ratios. A Fibonacci extension is the projection of a move beyond its starting point, for example to 1.618 or 2.618 times its length. You will find both tools in practically every charting software.

Importantly, the Fibonacci ratios are not rules of the theory. They help to narrow down possible turning zones. Whether the price actually turns there is decided by the market, not by the number. How often the guide values missed with Bitcoin is shown by the count in the next section.

Bitcoin from $3,782 to $69,000: an Elliott count that meets every rule

For our reconstruction, we analysed Bitcoin's daily highs and lows against the US dollar (BTC/USDT on the Binance exchange) from March 2020 to November 2022. Every high or low after which the price ran at least 20 percent in the opposite direction counts as a turning point. This produces a series of turning points into which an Elliott count can be placed without us setting points by feel. The following count is one of several possible ones. We chose it because it interprets the two all-time highs of 2021 as waves 3 and 5.

Wave	End point	Price in dollars	Length or retracement
Start	March 13, 2020	3,782	Covid low
Wave 1	August 17, 2020	12,468	+$8,686
Wave 2	September 5, 2020	9,825	30.4 percent of wave 1
Wave 3	April 14, 2021	64,854	+$55,029
Wave 4	June 22, 2021	28,805	65.5 percent of wave 3
Wave 5	November 10, 2021	69,000	+$40,195

The rule check is clean. Wave 2 ended at $9,825, well above the start at $3,782. At $55,029, wave 3 was the longest of the three driving waves, ahead of wave 5 at $40,195 and wave 1 at $8,686. Wave 4's low at $28,805 stayed far above the wave 1 high at $12,468. Overall, the price rose more than eighteenfold.

The Fibonacci guide values, by contrast, almost never hit. Wave 2 retraced only 30.4 percent instead of the usual 50 to 61.8 percent. Wave 3 was 6.3 times as long as wave 1; 1.618 times would have been at around $23,900, and 2.618 times at around $32,600. Wave 4 corrected 65.5 percent instead of 38.2 percent. For Bitcoin, this means the rules can be satisfied in hindsight, but the target zones derived from the ratios would almost always have been too early or too late in this cycle.

ABC correction in 2022: Bitcoin's fall to $15,476 in three waves

The high of November 10, 2021 was followed by the decline that this count interprets as corrective waves A, B and C. Wave A ran to $32,917 by January 24, 2022, a loss of $36,083. Wave B recovered to $48,190 by March 28, 2022, retracing 42.3 percent of wave A. Wave C ended on November 21, 2022 at $15,476, shortly after the collapse of the FTX exchange.

At $32,714, wave C was about 0.91 times as long as wave A. A common guide value says wave C is roughly as long as wave A, and here it fits reasonably well. From high to low, Bitcoin lost 77.6 percent. In hindsight, the count thus looks like a textbook example. In hindsight, mind you: which wave was running was an open question at every one of these points.

For an example of how differently chart patterns are read in such a phase, see our guide to bull flags and bear flags. The same applies there: a pattern is only confirmed by the breakout, never before.

A second wave count with a top in April 2021: where the Elliott method fails

The same turning points allow a second count. This variant places wave 3 at the interim high of January 8, 2021 at $41,950, wave 4 at the low of January 22, 2021 at $28,850 and wave 5 at the high of April 14, 2021 at $64,854. All three rules are met this way too: at $32,125, wave 3 was longer than wave 1, wave 5 at $36,004 was longer still, and the wave 4 low sat above the wave 1 high. Wave 4 even corrected almost exactly the guide value, namely 40.8 percent of wave 3.

Anyone using this count in April 2021 saw a completed five-wave impulse and therefore the end of the uptrend. The 55.6 percent crash to June seemed to confirm it. Then Bitcoin rose to $69,000 in November, 6.4 percent above the supposed end of the impulse. For this count, that is a contradiction the theory can absorb with auxiliary constructions, such as an irregular correction whose wave B runs beyond the start of wave A.

This is where the method's weakness lies. Two rule-compliant counts led to opposite conclusions at the same moment: one expected a further rise in wave 5, the other a bear market. Which one was right only became clear seven months later. A forecast that can be fitted to any outcome after the fact is hard to refute, and a statement that cannot be refuted does not deliver a reliable prediction either.

Studies on the predictive power of Elliott wave analysis

Robust studies on Elliott wave analysis are rare, and their results contradict each other. A review by CXO Advisory took 521 medium-term market outlooks by Elliott analyst Tony Caldaro from 2006 to 2008 and compared them with the subsequent performance of the S&P 500. The correlation with returns over the following seven trading days was 0.00, and with returns over the preceding seven days 0.31. The authors' conclusion: the outlooks reacted to past prices rather than predicting future ones.

On the other side is a 2017 paper by Eugenio D’Angelo and Giulio Grimaldi in the journal International Business Research. The authors report that the US dollar to euro exchange rate between 2009 and 2015 could be predicted with high accuracy using Elliott waves. We are not aware of any independent replication of this result, and for cryptocurrencies there is no comparable study with a large sample.

For the crypto market, the assessment that emerges from our own count therefore stands: Elliott wave theory organises price moves; it does not prove a direction. As a language for talking about trend phases, it is useful. As a signal for buying or selling, it is not proven.

Elliott waves on analytics platforms: tools and limits

If you want to count Elliott waves yourself, you do not need expensive specialist software. The common charting platforms offer drawing tools for impulse and corrective waves that set numbers and letters automatically, plus Fibonacci retracements and extensions. Some providers offer automatic wave detection; it follows the same rules, however, and faces the same problem that several counts can be valid at the same time.

A sensible choice is a platform that can store several counts side by side and link them to price alerts. For each count, set the level at which it becomes invalid, for example the wave 1 low for a wave 4 in progress. If the price breaks this level, the count is finished. That is the most honest way to use the method: not to defend the count, but to know the point at which it is wrong.

An indicator that measures independently of the wave count also helps, for example the momentum of a trend. How the Relative Strength Index works and what it recently showed for Bitcoin is explained in our guide to the Bitcoin RSI. If you prefer to have your rules executed automatically, crypto trading bots include providers that carry out price alerts and stops without manual intervention.

Leverage, stops and position size: why a wave count is no substitute for risk management

Elliott analysis becomes dangerous where it meets leverage. A count that expects a wave 3 tempts traders into large positions, because wave 3 is considered the strongest move of the cycle. If the count is wrong, as the second variant was in April 2021, leveraged positions risk forced closure before the picture clears. A Bitcoin pullback of 55.6 percent, as in spring 2021, wipes out a long with twofold leverage completely.

Three principles help to use the method without excessive risk. Position size is based on the distance to the count's invalidation level, not on the hoped-for price target. A stop sits at exactly that level. And every count is recorded in writing with a date, so that you can later check how often it was right. Anyone who wants to trade crypto assets with leverage or via derivatives should compare the terms of crypto brokers in advance, because financing costs and liquidation rules differ markedly. With leveraged products, a total loss of the capital invested is possible.

Elliott waves: only the invalidation level makes a count testable

Elliott wave theory gives you a vocabulary for trend phases and three clear rules. Our reconstruction on Bitcoin shows, however, that several rule-compliant counts are possible at the same time and that the Fibonacci guide values mostly missed in the 2020 to 2022 cycle. If you still want to use the method, proceed as follows:

1. Document your counts: Choose a platform that combines wave counts, Fibonacci tools and price alerts. Our comparison of crypto analytics platforms gives an overview.
2. Automate your rules: Set an invalidation level for every count and let an alert or stop trigger automatically. Suitable providers are listed among the trading bots compared.
3. Start without leverage: If you buy Bitcoin on the basis of an analysis, a regulated exchange with low fees is the best choice. The terms are in our comparison of crypto exchanges.

(As of November 14, 2026. This article is not investment advice. Prices and fee structures change; check the terms with the provider before you buy.)