XABCD Pattern: How to Draw, Read and Trade It

The XABCD pattern is the five-point frame behind the Gartley, the Bat, the Butterfly, the Crab and the Cypher. Five turning points, X, A, B, C and D, join four moves (XA, AB, BC and CD), and the Fibonacci ratios between those moves decide which pattern you are looking at. It is used to calculate in advance where a pullback or a reversal may end, at point D.

This page explains how to draw it, how the rules of each version work, a worked case with numbers, how to start trading it with an entry, a stop and targets, and what we can and cannot say about their reliability. We tested these patterns on our bench, and we give the results where they matter.

What is an XABCD pattern?

An XABCD pattern is a price structure with five pivots that alternate between highs and lows. It is a tool of technical analysis: the trader draws it on a chart with lines between the pivots, then uses Fibonacci retracements to find the zone where the last price movement should end.

The five points of the structure

Each letter of the name is a turn on the chart. X is the origin, A the end of the first move, B and C the two intermediate turns, and D the completion point, where the trade is taken. In a bullish setup, X, B and D are swing lows; in a bearish one, the opposite.

Defining the four legs: XA, AB, BC and CD

• XA: the initial impulse leg. Every other measure is based on it.
• AB: the first retracement of XA.
• BC: a move back in the direction of XA, which stays inside AB in the classic patterns.
• CD: the final move, which completes the structure at D.

In simple terms, the frame alone predicts nothing. It becomes a valid setup when each move respects its rule: where B sits on XA, where C sits on AB, and where D lands on XA and on BC.

Where the XABCD frame comes from

H.M. Gartley described a five-point retracement shape in Profits in the Stock Market (1935). Larry Pesavento later tied the moves to these ratios, and Scott Carney set precise rules for each pattern in The Harmonic Trader (1999), where he coined the term that now names the method. Scott Carney credits Bryce Gilmore with the structure of the Butterfly; the Bat and the Crab were defined by Scott Carney.

Understanding the core Fibonacci ratios

Every version is built from a short list of numbers: 0.382, 0.50, 0.618, 0.786 and 0.886 for retracements, 1.272, 1.618, 2.24 and 2.618 for extensions. A retracement measures how much of the previous move is given back; an extension measures how far a move travels beyond its starting point.

Fibonacci levels and their role in pattern validation

The ratios do two main jobs. At B, they orient the reading: a B at 0.618 of XA suggests a Gartley, a B at 0.786 a Butterfly. The B ranges overlap, though (0.618 also falls inside the Crab range), so only B and D together decide the pattern. At D, they define the potential reversal zone (PRZ): the area where the XA measure, the BC extension and the AB=CD measure meet. The tighter these three prices, the cleaner the setup. If a single measure falls outside its range, the pattern is not valid, whatever it looks like.

Major types of XABCD pattern

The same five pivots produce different patterns depending on two numbers: the depth of B on XA, and where D lands on XA.

The Gartley pattern: rules and identification

The first of the four to be described (Gartley, 1935). B sits at 0.618 of XA and D at 0.786, inside the initial impulse. In the classic description, CD extends 1.272 to 1.618 of BC and is about as long as AB.

The Bat pattern: critical retracement levels

A shallower B (0.382 to 0.50) and a deeper D, at 0.886 of XA, close to X. A stop just beyond X is therefore near the entry.

The Butterfly pattern: the extended CD move

A deep B at 0.786 of XA, and a D that goes beyond X, at the 1.272 extension of XA. It marks a new extreme that fails, which is why it is called an extension pattern.

The Crab pattern: extreme reversal zones

The most extended of the four: D at the 1.618 extension of XA, with a long CD move. It aims at the extreme of a sharp move.

The Cypher and Shark patterns: modern variations

The Cypher breaks one classic rule: C goes beyond A, to the 1.272 to 1.414 extension of XA, after a B at 0.382 to 0.618 of XA. D is then taken on the XC move, at 0.786 of it. The Shark also lets C go beyond A. Other structures on the same frame include the alternate Bat and the 5-0; the three drives is a related structure built on three pushes.

XABCD vs. ABCD

The ABCD pattern is the same structure without X: three moves, in which CD is roughly equal to AB. Adding the first move XA changes a lot: D can now be placed on XA, which gives a tighter zone and, for retracement patterns, a natural invalidation point at X.

How to identify an XABCD pattern on your chart

Draw the point labels in order, then check each rule. If one measure fails, the structure is either another design or no pattern at all.

Visualizing M and W formations

A bullish XABCD pattern starts with a rise from X to A. B is a low, C a high, and D a low where traders buy: the line joining the five pivots looks like an M. The bearish version is the mirror image, after a fall from X to A, with D at a peak where traders sell: it looks like a W. The rules are the same in both directions.

Manual drawing techniques using Fibonacci tools

1. Start with a clear XA: a strong, clean, large move that stands out. Draw a line from X to A. A weak or choppy first move gives measures that mean little.
2. Measure B: draw a Fibonacci retracement from X to A and read where B stopped. This tells you which type you may be looking at.
3. Check C: a second retracement on AB. In the four classic patterns, C sits at 0.382 to 0.886 of AB and never beyond A.
4. Project D: compute the XA ratio for that pattern, the BC extension, and the AB=CD level. Mark the zone they form.
5. Wait for price to reach D, the last point of the pattern: before that, the pattern is only a projection, and C can still form elsewhere.

Using automated scanners and indicators

Finding the pivots and monitoring many markets by hand takes a lot of work. Automated indicators and charting platforms find the five pivots and check the rules for you, and some platforms can alert you when a pattern completes, which is especially handy on lower timeframes. They help, but the pivots found by a scanner depend on hidden choices: how big a swing must be to count as a point, and how much tolerance is allowed around each measure. Two indicators can show different results on the same chart.

An XABCD pattern example with numbers

Illustrative values, not a real chart. Reading the table below row by row shows why point B matters. Take a bullish setup where X is at 100 and A at 150: XA is 50. Here is where D would sit for each of the four classic patterns.

The same X and A can give a buy zone anywhere from 110.7 down to 69.1. The difference comes from B and D: if B stopped at 119.1 (0.618 of XA), you may have a Gartley and are watching 110.7, or a Crab if price runs on to 69.1; if it stopped at 110.7 (0.786), you may have a Butterfly and are watching 86.4, below X. Our Gartley page works through more examples of the calculation.

Step-by-step guide to trading the XABCD pattern

With an XABCD pattern, the entry, the stop and the targets are all known before the trade.

Determining the reversal zone at D

Mark the PRZ from the three measures of D. A narrow zone gives a precise entry point; a wide one means the measures disagree, and the trading plan is less clear.

Setting entry points with confirmation

Two approaches. The aggressive one places a limit order inside the zone, buying a falling price with no sign of a turn yet. The conservative one waits for a sign at D: a reversal candle, a break of a short trend line drawn along the CD move, or a divergence on an oscillator such as the RSI. Our bench tested one such candle alone: on 15,363 daily hammers (12,005 closed trades), the 2:1 trade reached its target 35.4% of the time against 36.1% for the same trade on random dates; its average result per trade (+0.06 R against +0.09 R) did not differ measurably from random dates. Confirmation adds information, not certainty.

Our bench tests one standardized version of this trade, with the same entry and target rules for every harmonic pattern so that results can be compared, and each pattern's own stop: no limit order, but an entry at the close of the first bar that reaches the zone, the stop each pattern's page teaches (X for the Gartley and the Bat, the 1.618 extension of XA for the Butterfly, the far edge of the zone for the Crab) and a target at twice the risk. We also measured how often price reached the 0.618 retracement of AD before the stop, within 100 bars: 36.9% of bullish daily setups. That rate has no random baseline, so it describes the shape rather than an edge.

Managing risk: stops and take profit targets

For retracement patterns (Gartley, Bat), the natural stop sits just beyond X, and a horizontal line at X marks it on the chart: if price passes X, the structure is invalid. For extension patterns (Butterfly, Crab), D is already beyond X, so the stop goes beyond the next extension of XA (for a Butterfly, 1.618) or just beyond the zone. The classic targets are the 0.382 and 0.618 retracements of the AD move, then C, then A. One way to manage the trade is to take part of the position off at the first target and move the stop to breakeven.

Benefits of using the XABCD pattern

High risk-to-reward potential

In retracement patterns, the stop sits near the entry. In the Gartley example above (X at 100, A at 150, B at 119.1), suppose C turned at 135, about half of AB and inside the 0.382 to 0.886 range. Buying at 110.7 with a stop below 100 risks about 10.7; a target at C offers about 24.3, about 2.3 times the risk. That figure describes the plan, not the odds of reaching the target.

Versatility across different timeframes

The rules apply the same way in stock, forex and crypto trading, and on any timeframe, from intraday to weekly.

Improving trade consistency with structural rules

Explicit ratios make the patterns easier to check than a shape drawn by eye, and they force a written plan before each trade.

Challenges and limitations of XABCD trading

The difficulty of subjective pattern recognition

The choice of X and A is subjective: two traders may label the same chart differently, and each will find a valid pattern. The temptation is to stretch the ratios to find more. Human emotions play their part: after a few losses, it is easy to see the pattern you want to see.

Avoiding the trap of over-trading

With five designs, several timeframes and dozens of assets, there is always a setup somewhere. Each new design you add means more trading decisions, not better ones. Fewer, cleaner patterns are easier to manage than many loose ones.

Why market structure matters more than the pattern

Is the XABCD pattern reliable?

A fair test needs a fixed detection rule: which swings count as pivots, and how much tolerance to allow around each ratio. We set ours from chart reviews, never from results: swings of 2, 4 or 8 times the average true range, and single ratios accepted within 5% of their value. On our bench, the XABCD is not a rule of its own: it is any bullish (or bearish) Gartley, Bat, Butterfly or Crab, the four patterns described above, as defined on their own pages. In all four, X must be a true turning point: over as many bars as XA took, before X, price never went more than 10% of XA beyond it. When one X-A-B-C structure reaches more than one reversal zone, only the first one counts. The alternate bat is left out. These counts are low estimates: on simulated prices with mid-level noise and planted patterns, the narrow rule found about 70% of bearish patterns and 63% of bullish ones.

On 6,634 bullish setups on daily charts (6,598 closed trades), all markets together, the 2:1 trade reached its target 34.8% of the time, against 34.3% for the same trade on random dates. On 8,022 bearish setups (7,976 closed trades), it was 26.4% against 26.3%. In neither direction does the average result per trade (+0.15 R against +0.13 R on bullish setups) beat random dates by a margin that passes our statistical test. Each of these detection choices (swing size, tolerance) changes the result, and trying many versions until one looks good would produce a number that means nothing. Be wary of win rates quoted elsewhere without the number of cases, the markets, and a comparison with chance.

What our tests on the XABCD and its four patterns show

The rows below are measured on every occurrence our rule found on full price history, in crypto, forex, US stocks, ETFs, indices and commodities, against the same trade on random dates. The first two are the combined XABCD figures; the others are the four patterns it is made of. They overlap: never add them up, and do not read the combined result as a separate finding.

Read each row in pairs: the pattern against the same entry, stop and target placed on random dates on the same assets. On daily charts, all markets together, neither the combined figures nor any of the four patterns showed a measurable difference from chance in average result per trade. On other timeframes, one result survives our correction for multiple testing, on the direction 10 bars later rather than the 2:1 trade: the bearish butterfly on 4-hour charts. The combined bearish XABCD on 4-hour charts leans the same way but is rated no clear edge on direction (54.6% of 3,511 setups lower 10 bars later, against 51.5% for random bars). Hourly bearish setups read no clear edge too, while the bullish bat and the combined bearish XABCD on weekly charts pass a simple test but not our correction (promising). The same 4-hour bearish setups did not beat random dates on the 2:1 trade (−0.05 R against −0.04 R). That does not prove these setups fail, but it is a good reason to test them on your own market before trusting them. Our methodology explains the method.

Execution tips

Incorporating price and time analysis

Time measurements can be added to the price rules: a CD move that takes far longer than AB is a less balanced structure. Time measurements based on the same numbers add more choices to an already flexible method. We have not tested them.

Using confluence with technical indicators

A PRZ that coincides with a former support or resistance line, a round number or a moving average gives more than one reason to act at the same price. A pattern combined with such a level is easier to defend than one in empty space. An oversold or overbought oscillator is another possible filter.

The importance of backtesting your specific setup

Before trading real money, open a free demo account, mark the last completed patterns on your market with fixed rules, write the plan for each, and record every outcome, winners and losers, in a trading journal. Compare with the same trade on random dates, under the same conditions: it is the test we run at scale. Past performance is no guarantee of future results.

Best practices for risk management

Position sizing for XABCD setups

Size each trade on the distance between entry and stop, and risk only a small, fixed share of the account. Extension patterns have wider zones, so the same loss per trade means a smaller position.

How to handle failed patterns

When price breaks through D and the stop, the trend may continue in the direction of CD. Major economic news can also push straight through D. Take the loss and continue with the next setup: do not move the stop, and do not redraw the pattern with new pivots to keep the trade alive.

Common mistakes to avoid

• Forcing the ratios: calling a structure a Gartley when B is at 0.786. That is a Butterfly, with another D and another stop.
• Entering before D: until price reaches the zone, the pattern is incomplete.
• Ignoring scheduled news: a central bank decision or an earnings release can move the market far beyond any Fibonacci level.
• Counting only the winners: looking back, the patterns that worked are easy to spot. Record every one you would have traded.

Tools to draw and scan the XABCD pattern

Drawing tools

• TradingView XABCD pattern tool: a free drawing tool. You click X, A, B, C and D, and dashed lines between X and B, A and C, B and D, and X and D show each ratio as you draw. The line style, colors and labels can be changed in the settings.
• Fibonacci tools: a retracement on XA and AB and an extension on BC are enough to check any setup by hand, on any trading platform. Draw each line from swing to swing.

Scanners and other products

• Scripts and scanners: free community scripts and paid products (some with a free trial) detect XABCD setups automatically. Additionally, some products add alerts and practical applications such as backtests. Check the rules and tolerances each one uses, and test it on past data, for at least a month on a demo account, before relying on it or paying for one.

Further learning resources

Each pattern built on this frame has its own page with its rules, a worked case and trading plan: the Gartley, the Bat, the Butterfly, the Crab and the Cypher. These resources, and the ABCD page, are the next step if you want to learn each design in detail.

Frequently asked questions about the XABCD pattern

Summary

• The XABCD pattern is a five-point, four-move frame; fixed proportions define the Gartley, Bat, Butterfly, Crab and Cypher on it.
• B orients the reading and B and D together decide the pattern; D, the reversal zone, is calculated in advance from XA, BC and AB=CD.
• Retracement patterns end inside X; extension patterns end beyond X and need a different stop.
• Tested on our bench as the union of the Gartley, Bat, Butterfly and Crab: no measurable difference from chance on daily charts, all markets together. Only direction stood out, not the 2:1 trade: the bearish butterfly on 4-hour charts survives our correction, while the bullish bat and the combined bearish setups on weekly charts pass a simple test only.

This guide is general information about technical analysis, not financial advice. Trading carries a risk of loss: never trade money you cannot afford to lose.

About the author and the data. Written by Amaury, founder of PatternsWizard, who is passionate about finance, quantitative trading and data analysis. He built the PatternsWizard test bench, which detects each setup with fixed, published rules on full market history: crypto, forex, US stocks, ETFs, indices and commodities. Trades and win rates are compared with random bars on the same assets, excess returns with the rest of the market, with error margins and a correction for multiple testing, and the whole pipeline is checked on random-walk prices, where nothing can be predicted. The XABCD figures combine the Gartley, Bat, Butterfly and Crab rules (swings of 2, 4 or 8 times the average true range, ratios within 5%), never tuned on results. Statistics on this page were generated on November 5, 2026. The full methodology explains every step.