ABCD Harmonic Pattern: Rules, Ratios and How to Trade It

The ABCD harmonic pattern is a four-point price structure, labeled A, B, C and D, in which the last leg (C to D) is about as long as the first (A to B), according to specific Fibonacci ratios. Traders use it to identify in advance the point where a price movement may end: they buy at D in a bullish pattern and sell at D in a bearish one. It is the harmonic pattern with the fewest points, and the building block inside the Gartley, the Bat and the other five-point harmonic setups.

This guide gives the exact Fibonacci ratios, a worked example with numbers, how to trade the pattern with entry, stop and objectives, the common mistakes, and an honest answer to the question that matters before any trade: does it work? We give our own test results, on every market, next to the choices the test required.

What is the ABCD trading pattern?

The ABCD pattern is a chart pattern that consists of three legs, or price swings, that join four turning points. The market moves from point A to point B, retraces part of that move to point C, then moves again in the first direction to point D. The idea is symmetry: the pattern assumes that markets move in waves of similar size, so the third swing is expected to match the first one in length, and in some versions in duration too.

H.M. Gartley described this kind of measured move in Profits in the Stock Market (1935). Larry Pesavento later linked it to Fibonacci ratios, and Scott Carney, who coined the term harmonic trading in The Harmonic Trader (1999), set out precise ratio definitions for it. It is also called the AB=CD pattern. This chart pattern does not predict anything by itself: it gives a price zone where a reversal can be watched for, and planned, before price gets there.

The three phases of the ABCD pattern

Each phase has a role. Read them one after the other on the chart.

The A-B leg

The pattern starts with a clear, strong initial movement: a decline from A to B in a bullish pattern, a rise in a bearish pattern. This move sets the size that the last move is expected to match. A weak or choppy first move makes every later measure less meaningful.

The B-C leg

The market then retraces part of the first move: 38.2% to 88.6% of it in Carney's definition. Point C must not go beyond point A: if it does, the setup is broken. The B-C leg is the countertrend phase.

The C-D leg

The last move goes in the direction of the first, beyond point B, and completes at D. CD is about equal to AB, and its projection of BC depends on the depth of C (see the table of ratios below). Point D is where a reversal is expected.

How to identify the ABCD pattern on a chart

To identify one, follow these steps in order. Consider each step a filter: if one fails, move on and follow the next setup. The anatomy chart in this section is the median shape of a random sample of real daily setups from all markets; every setup has the same four points, so its caption's "most common number of swing points" covers all of them.

1. Find a clear first move from A to B when looking at the price chart.
2. Measure the retracement: draw a Fibonacci retracement on the A-B leg and check that C stopped between 38.2% and 88.6% of it, without passing A.
3. Project point D: mark where the last move would equal the first, and where it would reach the multiple of the B-C leg that matches the retracement at C (see the table below).
4. Check time, if your version requires it: some versions want the C-D leg to take about as long as the A-B leg. Our bench only asks for D between half and twice the duration of AB after C, so we have not tested whether a closer match helps.
5. Wait for the market to reach D: until then, it is only a projection, and C can still form elsewhere.

Fibonacci ratios of each leg

The table below provides the usual pairs. The retracement at C and the projection to D are linked: the deeper the retracement, the smaller the multiple of the B-C leg needed for the last move to equal the first.

The two measures for D, the equal-size measure and the multiple of the B-C leg, create a small zone. The tighter it is, the cleaner the setup. An extended version also exists, in which the last move is 1.27 or 1.618 times the first.

A worked example with numbers

Illustrative values, not a live chart. In this example, point A is at 150 and the market falls to point B at 120: AB is 30.

• Point C at a 61.8% retracement: 120 plus 18.5, so 138.5.
• Equal length: a move of 30 from 138.5 puts D at 108.5.
• Projection: 1.618 times BC (18.5) is 29.9, which puts D at 108.6.

The two measures meet at 108.5 to 108.6: a tight zone. The extended version (1.27 times the first move, 38.1) would put D at 100.4, the level beyond which this guide counts the setup as failed (see the exit below).

Bullish vs. bearish ABCD patterns

Identifying a bullish setup

A bullish pattern forms in a downward market: price falls from A to B, followed by a rebound to C below A, then falls again to D, below B. On a price chart, the shape is a zigzag of falling highs and falling lows. Traders looking for a trading opportunity buy at D, expecting an upward reversal. Understand that reversals at D are not guaranteed.

Identifying a bearish setup

A bearish ABCD pattern is the mirror image in a rising, trending market: price rises from A to B, followed by a fall back to C above A, then rises again to D, above B, a new high. Traders sell short at D with a sell limit order, or close open positions, expecting the market to turn down.

Step-by-step guide to trading the ABCD pattern

As with other harmonic setups, the appeal of this one for trading is that the entry, the exit if wrong and the objectives are known before the trade.

Entry strategy and confirmation

Keep in mind what our test bench found on those signals alone: on 12,005 closed trades on daily hammers, the 2:1 trade reached its target 35.4% of the time against 36.1% for the same trade on random dates, a difference rated No clear edge: its average result per trade (+0.06 R against +0.09 R) is within the margin of error of random dates. A reversal signal at D adds information, not certainty.

Setting stop loss and take profit levels

With the numbers above, buying at 108.5 with an exit at 100.4 risks 8.1. The move from A to D is 41.5, so the 38.2% objective is at 124.4 (a gain of 15.9, about 2.0 times the risk) and the 61.8% objective at 134.1 (a gain of 25.6, about 3.2 times). These figures describe the plan, not the odds.

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 this pattern's own stop: no limit order, but an entry at the close of the first bar that reaches the zone, a stop where CD is 1.27 times AB and a target at twice the risk. The 61.8% retracement of AD is measured separately.

Position size and trading risk

Size each position on the distance between your order and your exit, so that being wrong costs a small, fixed share of your trading account. For example, risking 1% or less of the account per position keeps a run of losses small, whatever the account size. Consider that a few losing trades in a row are normal, and size for them.

ABCD vs. XABCD harmonic patterns

Add a first move before A, labeled X, and the setup becomes a five-point structure: the XABCD pattern. The Gartley, the Bat, the Butterfly and the Crab all contain one in their last three swings. Point X provides an extra measure for D and, for the Gartley and the Bat, a natural exit. The four-point version has no X, so its exit is less obvious.

The three drives pattern extends the same idea of symmetry to three pushes in a row.

Is the ABCD pattern reliable?

A fair test needs a fixed definition of each turning point and of the tolerance around each ratio, and we set ours from chart reviews, never from results. Turning points are swings of 2, 4 or 8 times the average true range, and A must be a real turning point: over a stretch as long as AB before it, price never went more than 10% of AB beyond A. C must retrace 0.382 to 0.886 of AB. The reversal zone is where CD measures 0.95 to 1.05 times AB, AB=CD within 5%; the BC projection is not checked separately, since equal legs already give the pairs of the table above. Equal timing of the legs is not required, but D must come between half and twice the duration of AB after C. The signal is the first bar that trades into the zone without reaching the point where CD is 1.27 times AB. Each of these choices changes the outcome, and trying many versions until one looks good would produce a figure that means nothing.

It is by far the most frequent harmonic setup our bench detected, even though the strict rule misses about one bullish setup in five in our recall test on synthetic charts. On 20,366 closed trades on bullish setups on daily charts, all markets together, the 2:1 trade reached its target 36.6% of the time, against 35.9% for the same trade on random dates. On 30,798 trades on bearish setups, it was 28.9% against 28.0%. On bullish setups the hit rate is a little higher than on random dates, but the average result per trade (+0.13 R against +0.12 R) does not pass our statistical test. On bearish setups, the average result per trade (−0.11 R against −0.14 R) is also within the margin of error, which is why the bearish 2:1 trade reads No clear edge: it loses a little less than the same trade on random dates, but still loses money. The 61.8% retracement of AD was reached before the stop, within 100 bars, in 35.0% of bullish daily setups and 26.4% of bearish ones. Those rates are not in the tables: they describe the shape, not an edge. Be wary of win rates quoted elsewhere without the count of trades, the markets, and a comparison with chance.

What the bench measured on the ABCD

Every occurrence our rule finds on full price history, in crypto, forex, US stocks, ETFs, indices and commodities, compared with the same trade placed on random dates on the same assets.

Read each row in pairs: the pattern against the same order, exit and objective placed on random dates. On daily charts, all markets together, the ABCD showed no confirmed edge in either direction: the 10-bar move of bearish setups passed a simple test (promising) but not our correction for multiple testing, the bullish one read no clear edge, and the bearish 2:1 trade did a little better than random dates in average result per trade, within the margin of error. The verdict reads No clear edge, and the trade still loses money. A few results stand out. On daily US stocks, price was lower 10 bars after bearish setups in 47.6% of cases, against 45.4% after random bars, an excess return of +0.20% that survives our correction, holds in both halves of the history and stays positive after costs (proven edge); the 2:1 trade there is labeled "No clear edge" and still loses money. Bullish setups on daily forex look better at first sight: the 2:1 target was reached 41.9% of the time against 34.2% on random dates, for +0.29 R against +0.09 R, but on only 267 trades that gap is within the margin of error: no clear edge. After bearish setups on hourly charts, mostly crypto, price was lower 10 bars later in 53.9% of cases, against 50.4% after random bars, a gap that passes a simple test but not our correction for multiple testing (promising). With thousands of tests across markets, timeframes and patterns, a single confirmed result among many null ones is a lead, not a strategy. The method is published in our methodology.

Common mistakes to avoid when trading ABCD patterns

• Forcing the symmetry: a last move 40% longer than the first does not have equal legs. Stretching the definitions finds more setups, but they are no longer the pattern you tested.
• Acting before D: the setup only exists once the market reaches the zone.
• Ignoring the larger trend: a buy setup in a market making new lows on the higher timeframe fights that trend, and the last move can run further.
• Moving the exit: once price passes it, the setup has failed. Moving it can turn a small loss into a big drawdown.
• Remembering only the winners: on past charts, the setups that worked are easy to see. Keep a record of every one you would have traded.

Technical confluence

Some harmonic traders want the zone at D to line up with other levels at the same price: a prior support or resistance, a prior high or low, a Fibonacci retracement of a larger move, a round number, a moving average, or the end of a larger movement on a higher timeframe. A momentum indicator can also show whether the C-D leg is losing strength. We have not tested these filters on this pattern.

Tools and indicators for harmonic trading

TradingView offers a drawing tool for it that shows the ratios as you place the points. The TradingView community library also has free indicators that detect the ABCD pattern automatically, and versions exist for the MetaTrader platform. Choose a tool whose rules are published, and check which ratios and tolerances each one uses: two indicators can show different setups on the same chart. Practice on a demo trading account before risking money, and create alerts so you do not have to watch every chart.

Frequently asked questions

Summary of the ABCD harmonic pattern

• Four points, three moves: BC gives back 38.2% to 88.6% of AB, and CD is about as long as AB, with a multiple of BC that depends on the depth of C.
• Traders act at D, place the exit beyond it or beyond the 1.27 extended level, and aim for the 38.2% and 61.8% retracements of the move from A to D.
• It is the core of the Gartley, the Bat, the Butterfly and the Crab harmonic patterns.
• Tested on our bench with a fixed rule: no confirmed edge on daily charts with all markets pooled; the one confirmed result is narrow, the 10-bar direction of bearish setups on daily US stocks, and no 2:1 trade beats chance.

This guide is general information about technical analysis, not financial or investment advice. Trading carries a risk of financial 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 ABCD rule (swings of 2, 4 or 8 times the average true range, AB=CD within 5%) was set from chart reviews, never tuned on results. Statistics on this page were generated on November 5, 2026. The full methodology explains every step.