Diamond Chart Pattern: What the Statistics Actually Say
Search the diamond chart pattern and three of the guides that come back will tell you how often it works. One says 81 percent. One says 38 percent for the top variant and 41 percent for the bottom. One says diamond bottoms break upward 73 to 74 percent of the time.
Those numbers cannot all describe the same thing, and none of the three pages acknowledges that the others exist.
The figures are traceable, which makes this unusual. Most pattern statistics circulating online have no origin at all, but these lead back to one published database, and reading it settles which of the three is closest, what the numbers actually measure, and where one of them has been read wrong.
Key takeaways
- Three widely read guides give three incompatible success rates for the same pattern: 81 percent, 38 to 41 percent, and 73 to 74 percent.
- Bulkowski publishes the underlying figures himself, and the 73 percent is the share of diamond bottoms that reached their price target after an upward breakout, not the share that broke upward.
- Every published figure is labelled a bull market result and drawn from what the author calls perfect trades: 733 of them for tops, 477 for bottoms.
- The same source ranks diamond tops with upward breakouts 39th of 39, and diamond bottoms with downward breakouts 1st of 36. The pattern has no single reliability number.
- Break-even failure rates run 21 and 15 percent for tops and 13 and 15 percent for bottoms, which is the closest thing published to a failure rate.
- A diamond is a broadening formation followed by a symmetrical triangle, so whether one exists at all depends on where you place four trendlines.
Table of contents
Three Guides, Three Incompatible Success Rates
Set the three claims side by side and the problem is immediate.
| What the page claims | What it offers as backing | Date on the page |
|---|---|---|
| 81 percent success rate | Nothing at all | 26 November 2024 |
| 38 percent accuracy for tops, 41 percent for bottoms | Its own counts of 904 and 1,052 formations, with no method, market or date range | 9 February 2024, never updated |
| Bottoms break upward 73 to 74 percent of the time | Named as Bulkowski research, with no link to it | 12 March 2026, no modified date |
An 81 percent success rate and a 38 percent accuracy rate differ by more than a factor of two. Either they are measuring different things without saying so, or at least one is wrong.
The word doing the damage is the label. Success, accuracy and breaks upward are three different measurements, and a reader comparing three pages has no way to know that, because not one of the three defines its term.
This is a recurring shape in pattern writing rather than a fault unique to the diamond. Another count with strict voiding conditions attracts the same treatment: a percentage travels from page to page and the definition behind it does not travel with it.
What the Primary Source Actually Reports
Two of the three point, directly or by name, at the same place. Thomas Bulkowski publishes per-pattern statistics on his own site, and the diamond pages there were updated on 31 July 2025.
He reports each variant separately, and each breakout direction separately again. Both sets are labelled Important Bull Market Results.
| Measure, up breakout then down breakout | Diamond tops | Diamond bottoms |
|---|---|---|
| Overall performance rank, 1 is best | 39 of 39, then 3 of 36 | 27 of 39, then 1 of 36 |
| Break-even failure rate | 21 percent, 15 percent | 13 percent, 15 percent |
| Average rise or decline | 29 percent, 17 percent | 39 percent, 19 percent |
| Throwback or pullback rate | 57 percent, 58 percent | 52 percent, 67 percent |
| Percentage meeting price target | 65 percent, 63 percent | 73 percent, 55 percent |
| Sample | 733 perfect trades | 477 perfect trades |
Three of those rows need their terms before they mean anything. The break-even failure rate counts patterns whose move after the breakout was too small to clear a nominal cost threshold, so it is a measure of moves that went nowhere rather than of moves that reversed.
The throwback and pullback rate counts how often price returned to the breakout level shortly after leaving it, which is a comment on entries taken at the break rather than on the eventual direction. The percentage meeting price target counts how often the move reached the level a measuring rule projects from the pattern.
Read the first row before any other. Diamond tops breaking upward rank last of 39 patterns. Diamond bottoms breaking downward rank first of 36.
The same shape sits at both extremes of the same ranking depending on which variant formed and which way it broke. Any single number offered as the reliability of the diamond pattern has collapsed that spread into one figure, and the spread is the finding.
None of this is a claim about how the figures were produced. That is a separate question, and how a pattern statistic is actually produced determines how much weight any of them carries.
Target Attainment Is Not Breakout Direction
Now the specific error.
The claim that diamond bottoms break upward 73 to 74 percent of the time attributes itself to Bulkowski. On his diamond bottom page, 73 percent is the percentage meeting price target for upward breakouts.
Those measure different events. Percentage meeting price target asks a conditional question: given that the pattern broke upward, how often did price go on to reach the target the measure rule projects. It says nothing whatever about how often the upward break happened in the first place.
A reader who takes 73 percent as a directional probability concludes that roughly three diamond bottoms in four resolve upward. The source supports no such statement. What it supports is that among the ones that did break upward, roughly three in four reached their projected target.
The neighbouring figures in the same claim show the same drift. An average rise of 35 to 39 percent brackets the 39 percent that the source gives for diamond bottoms with upward breakouts. An average decline of about 17 percent matches the 17 percent the source gives for diamond tops with downward breakouts, which is the other variant entirely. Two variants have been merged into one set of numbers.
The remaining figure in that claim, 54 percent for tops breaking downward, does not appear anywhere among the bull market results published for diamond tops.
What Perfect Trades and Bull Market Exclude
Two qualifiers govern every figure in the table above, and neither one survives into any page that restates them.
The first is bull market. Both result sets are published under that heading. Whatever they describe, they describe it under one market condition, and the pages that repeat the numbers apply them to any market a reader happens to be in.
The second is the sample description. The figures rest on 733 perfect trades for tops and 477 for bottoms. Perfect is the author’s own qualifier for the subset he counted, not a synonym for all instances found.
The practical consequence is that these are not base rates for what a pattern on your screen will do. They are summary statistics for a filtered set of historical examples in one market condition, published as such, and stripped of both qualifiers by the time they reach a forex blog.
The direction of that filter is not stated either. A subset described as perfect could have been selected for clarity of shape, for completeness of the move, or for both, and each choice pushes the resulting averages a different way. Without the selection rule, the numbers cannot be adjusted for it.
There is a further gap that nobody restating these figures mentions: the database they come from is a chart-pattern database built on stock charts, and no comparable that cites it states which market the numbers came from.
How the Shape Is Defined, and Where That Bites
The definition is where the reliability question actually starts, because it decides whether you have a diamond at all.
A diamond is two formations in sequence. Price widens out into a broadening formation, with higher highs and lower lows, and then narrows into a symmetrical triangle. Drawn on a chart the outline resembles a diamond, usually tilted rather than symmetrical. Trending up into the shape makes it a top; trending down into it makes it a bottom.
That construction requires four trendlines, two for the expansion and two for the contraction, and every one is placed by eye. Move any of the four to a different swing and the shape either appears or disappears. This is why the pattern is described as rare: a stricter hand finds fewer.
That subjectivity also explains why the pattern is reported as rare while appearing frequently in tutorials. A shape assembled from four hand-placed lines can be found in a great deal of price action if the placement is generous, and almost nowhere if it is strict. Rarity is a property of the rule being applied, not only of the market.
It also means the statistics and your chart may not be discussing the same object. A figure computed from patterns identified by one set of rules describes those patterns, not whatever a looser or tighter reading produces on a different instrument and timeframe.
The entry question inherits the same problem. The trigger is a break through one side of the contracting half, which puts the whole apparatus of what separates a real breakout from a false one in front of the pattern statistics rather than behind them. Reading volume alongside the shape is the confirmation most commonly recommended, though no source consulted here quantifies how much it changes the outcome.
What a Forex Reader Can Honestly Take From This
Three things survive the check, and they are worth more than a success rate.
The first is that the pattern has no single reliability figure, and any page offering one has discarded the distinction between the two variants and the two breakout directions. Ask which variant and which direction before accepting any number.
The second is that the closest published equivalent to a failure rate is the break-even failure rate, not the target-attainment rate that gets quoted. Those run 21 and 15 percent for tops and 13 and 15 percent for bottoms, and they answer a different question from the 65 and 73 percent figures that appear in headlines.
The third is a habit rather than a fact. When a percentage appears with a name attached and no link, the name is doing the work of the evidence. Here the source was two clicks away and the number had been changed in transit.
This page will not serve a reader who wants a probability to size a position against. No figure here is a base rate, the published set is bull-market and filtered, it was not computed on currency pairs, and nothing in it forecasts what a particular chart will do next. A reader wanting entry rules, stop placement and targets for the pattern will also find those elsewhere and not here, because the sources that state them do not source them.
Frequently Asked Questions
Is a diamond pattern bullish or bearish?
Neither on its own. A diamond forming after an uptrend is a diamond top and is usually read as a bearish reversal, while one forming after a downtrend is a diamond bottom and is read as bullish. The direction of the break out of the contracting half is what settles it, and both variants can break either way.
How do you read a diamond pattern on a chart?
Look for price first widening into a broadening formation with higher highs and lower lows, then narrowing into a symmetrical triangle. Four trendlines outline the shape, two for the expansion and two for the contraction. Because all four are placed by eye, two traders can disagree about whether one is present.
How reliable is the diamond chart pattern?
There is no single answer, and that is the honest one. The published statistics rank diamond tops with upward breakouts last of 39 patterns and diamond bottoms with downward breakouts first of 36. Reliability depends entirely on which variant formed and which way it broke.
What separates a diamond from a head and shoulders?
A head and shoulders is defined by three peaks with a neckline drawn under them. A diamond is defined by a change in volatility, expanding and then contracting, and is outlined by four converging trendlines rather than by a neckline. Some diamonds contain a shape that could also be read as a head and shoulders, which is one reason the two get confused.
Risk warning: this page is educational. It compares published statistics for a chart pattern against the source they are drawn from, and describes how the pattern is defined. It is not advice to trade this or any pattern, no figure here is a prediction or a base rate for any instrument, and past statistical results do not indicate future outcomes. Leveraged trading carries a high risk of loss.