Trading patterns: spotting what makes you win or lose

A trading pattern is something that repeats: on a chart, it's a price formation; in your journal, it's a regularity in how you trade that weighs on your results without you being aware of it. Some patterns make you win, others make you lose, and most stay invisible until you look for them. This guide explains what the published numbers say about chart patterns, which patterns to spot in your own trading, and how to use them to improve.

• Your patterns weigh on your results beyond your stated strategy.
• A chart pattern comes with average statistics: yours are measured in your journal.
• They're invisible without analysis: you suffer them until you look for them.
• There are winning patterns to repeat and losing patterns to cut.
• They appear in breakdowns by time, instrument, day and emotion.

When we talk about patterns in trading, we often think of chart patterns. But some of the patterns that weigh on your performance aren't on the chart, they're in your behavior: the moments you win most, the situations where you systematically lose, the emotions that coincide with your best and worst days. These behavioral patterns are a ground for progress that only you can explore.

The problem is that these patterns are invisible to the naked eye. Lost in the flow of your trades, you can't see that you always lose on Friday afternoon, or that your best setup hides in a specific time window. This guide shows you which patterns to look for, how to surface them from your data, and how to turn this knowledge into a concrete advantage, from the initial spotting all the way to the written rule you apply day to day.

Chart patterns: what the published numbers say

On a chart, a pattern is a chart pattern. StockCharts' ChartSchool, on its chart patterns page, sorts them into two broad families: reversal patterns, which indicate a trend change, and continuation patterns, which indicate the trend will continue after a brief consolidation. The inverse head and shoulders belongs to the first, the flag and the VCP pattern to the second.

These patterns have been counted. For each of them, Thomas Bulkowski publishes on his site, ThePatternSite, a failure rate and an average rise measured on stocks, in bull markets. His glossary sets the conventions: a bull market is everything outside the two S&P 500 declines of March 2000 to October 2002 and October 2007 to March 2009, the failure rate counts the patterns that fail to rise at least 5% after the breakout, and the average rise is measured from the open the day after the breakout to the ultimate high. That last point matters: his pages speak of perfect trades, sold at the top, not of a trader's results.

These numbers are read on Bulkowski's pages for the cup with handle, the head-and-shoulders bottom, the Eve and Eve double bottom and the Adam and Adam double bottom. Two things follow. Shape changes the result: between two double bottoms, the failure rate goes from 12% to 16% depending on how the bottoms are drawn. And none of these averages says what the pattern does in your hands, on your market and your timeframe. Even the return to the broken level is not automatic: according to his Throwbacks page, it follows 58% of 10,305 upward breakouts recorded since 2000, which leaves roughly four breakouts in ten without a retest.

Academic research is more cautious still. The study by Lo, Mamaysky and Wang published in 2000 in the Journal of Finance had an algorithm recognize ten patterns, including the head and shoulders and the double bottom, on US stocks from 1962 to 1996. It starts from an observation, that the presence of a geometric shape on a chart is often in the eyes of the beholder, and concludes that several patterns do provide incremental information, especially for Nasdaq stocks, without this implying that they can be used to generate excess profits. So a chart pattern is neither a superstition nor a guarantee: it is a hypothesis, and your journal is where it gets tested for you.

Elliott waves are a case apart: they describe not a pattern but a count, five waves with the trend and three against it, which two readers rarely place the same way on the same chart. The lesson on Elliott wave theory gives its rules, a worked example, and why no published study we found measures its predictive value.

The patterns that really matter

The patterns only you can measure are behavioral, not chart-based. They're the regularities in how you trade: the setups where you excel, the moments you derail, the emotional states that precede your best and worst decisions. Some are documented at scale: across 10,000 accounts at a US discount broker, from 1987 to 1993, Terrance Odean's study of the disposition effect shows stock investors realizing their gains more readily than their losses. Yours are unique to you, they reflect who you are as a trader, and that's precisely why they're so valuable to know.

Unlike chart patterns, available to everyone, your behavioral patterns are your private information. No one but you can discover them, because they're in your own data. Identifying them gives you an advantage no one can copy: the intimate knowledge of what makes you win and lose, which lets you adjust your behavior where it truly counts, rather than following generic advice that ignores who you actually are.

The winning patterns to repeat

The first family of patterns to look for is your winning patterns: the situations where you win most often and biggest. Maybe a specific setup shows an excellent win rate, a certain time window concentrates your best trades, or a particular instrument suits you far better than others. These positive regularities are gold mines, because they tell you where to concentrate your energy.

Fixing a weakness means learning something new. Repeating a strength only requires having spotted it.

Identifying your winning patterns enables an often-neglected optimization: instead of correcting weaknesses, you amplify strengths. Discovering that a single setup produces a large share of your gains pushes you to play it more, prepare it better, not miss it, and give it a sheet in your trading playbook. Doing more of what already works is one of the simplest and most effective ways to improve your performance, and it starts with recognizing your winning patterns.

The losing patterns to cut

The second family is your losing patterns: the situations that systematically cost you money. A weekday where you always lose, an hour when your trades go wrong, a type of market that doesn't suit you, or an emotional state that coincides with your worst days. These negative patterns are leaks you can often plug simply by stopping the activity that produces them.

Optimization by removal, stopping what makes you lose, is one of the easiest gains in trading. If your data shows a certain situation has a clearly negative expectancy, the decision is obvious: avoid that situation. Removing your worst patterns requires changing nothing else, provided the gap holds on a sufficient sample. But to cut them, you first have to see them, and that's where analysis becomes indispensable. It's also psychologically easier than fixing a technical weakness: it's not about learning to trade a situation better, just about no longer exposing yourself to it at all.

Surfacing patterns through breakdowns

Patterns don't appear in a raw list of trades, they emerge when you break down performance along different dimensions. By grouping your trades by time of day, day of week, instrument, setup or emotional state, you surface performance differences that were drowned in the mass.

Each of these breakdowns is a lens that reveals a different type of pattern, and none of them is optional if you want the full picture rather than a partial, potentially misleading one. It's by crossing these dimensions that you get the finest information: not just that you lose on Friday, but that you lose on Friday afternoon on a specific instrument when you're in a degraded emotional state. The more precise the breakdown, the more actionable the pattern, and the more targeted and effective the correction.

Beware of false patterns

Searching for patterns is worth doing seriously, not as a one-off exercise but as a recurring habit, because your patterns shift as you gain experience, as market conditions change, and as your life circumstances evolve. A review that made sense six months ago can be stale today. An important trap awaits the pattern-seeker: the false pattern, an apparent regularity that's actually just chance. On a small sample, you can always find a motif, like three consecutive losses on a Tuesday, that seems significant but isn't at all. Acting on a false pattern means mistaking noise for signal, and complicating your life for nothing.

To distinguish a real pattern from a false one, you need a sufficient sample size and a marked, consistently repeated difference, not a one-off streak that could just as easily have gone the other way. A pattern based on five trades is worthless; the same based on dozens of trades, with a clear performance gap, deserves attention. Statistical caution is essential: better to act on a few solid, well-established patterns than on a multitude of fragile motifs that are just disguised chance. The order of magnitude can be computed: with the confidence interval as set out in the NIST statistics handbook, from the US standards institute, a 60% win rate measured over 30 trades sits, at 95% confidence, between roughly 42% and 75%, and over 100 trades between 50% and 69%. A time window showing 60% over thirty trades is therefore not yet distinct from a window at 50%.

Time-based patterns: hour, day, session

Time is a dimension to break down first, because the hour and the day of every trade are already in your data. A trader can have hours when their focus, market reading and discipline are better, and hours when they degrade, without realizing it. They might excel at the London session open, for instance, and perform noticeably worse late in the day, when fatigue builds up and patience runs out.

The day of the week adds another layer: some traders perform much better mid-week, when their rhythm is stable, and noticeably worse on Fridays, caught between wanting to close the week on a win and accumulated fatigue. Crossing hour and day refines the pattern further: a trader might discover they almost systematically lose on Friday after 4pm, a very specific window they can simply decide to avoid, without changing anything else about their strategy.

Patterns tied to instrument and setup

Not all instruments are equal for a given trader, even within the same asset class. A trader might have a real edge on one index but be structurally unprofitable on another, simply because their reading of context, execution speed or tolerance for volatility fits one better than the other. This pattern is easy to miss if you trade several instruments without ever comparing them separately.

Setup adds an even finer layer of granularity. Two setups that look similar in theory, a range breakout and a trend breakout for example, can have radically different performance profiles for the same trader. One might show a solid edge, the other a slightly negative expectancy, even though they belong to the same family of chart patterns. Without a per-setup breakdown, these two opposite realities stay drowned in a single, misleading average statistic.

Emotional and physiological patterns

Some patterns are neither time-based nor tied to an instrument, but purely internal. Fatigue, lack of sleep, stress accumulated outside trading, or even hunger, can weigh on decision quality without you being aware of it in the moment. For sleep, the effect has been measured in the lab: in the study by Venkatraman and coauthors published in 2011 in the Journal of Neuroscience, healthy volunteers deprived of one night of sleep and faced with monetary gambles moved from defending against losses to seeking increased gains. Those are lab gambles, not trades, but it is a reason to check it in your own data: a trader might, for instance, find that trades taken after a night with less than six hours of sleep have a noticeably lower win rate than those taken after a full night's rest.

These physiological and emotional patterns are worth looking for early, because they touch causes rather than surface symptoms, and because one cause can sit behind several of your other patterns. Spotting that a degraded emotional state systematically precedes a losing streak doesn't just say what to avoid, it also reveals an early warning signal: the next time that state appears, the trader knows to cut size, or even stop, before the losses pile up.

From discovery to action

Discovering a pattern is useless if you don't turn it into a concrete rule. Once a pattern is confirmed on a sufficient sample, good practice is to write it down as a rule in your trading plan: 'I don't trade after 4pm on Fridays,' or 'I cut my size in half if I slept less than six hours.' A written rule is infinitely easier to respect than a vague intuition about what works or doesn't.

This transformation also requires ongoing tracking: a pattern identified today can evolve with your experience, your life circumstances or market conditions. Regularly reviewing your patterns, rather than treating them as fixed once and for all, ensures your trading plan stays aligned with who you actually are as a trader, not who you were six months ago.

A worked example: uncovering a hidden pattern

Let's take a concrete example. A trader who takes an average of 200 trades over a quarter has an overall win rate of 48% and a profit factor of 1.1, decent but nothing special. Breaking these same trades down by hour, they discover their 40 trades taken between 2pm and 4pm show a 61% win rate and a 1.9 profit factor, well above their overall average. Conversely, their 35 trades taken after 7pm show a 34% win rate and a 0.6 profit factor, clearly losing.

This pattern, invisible in the overall statistic, changes everything once isolated. By choosing to concentrate their activity on the 2pm to 4pm window and stopping trading entirely after 7pm, this same trader, without changing a single rule of their strategy, could see their overall profit factor climb significantly, simply by playing their best pattern more often and cutting their worst one. It's the perfect example of the value of breakdowns: the same strategy, the same trader, but a radically more efficient allocation of time.

How Tradoshi reveals your patterns

Tradoshi automatically does the breakdowns that reveal your patterns, on your real trades. You see where and when you win or lose, which setups suit you, and which emotional state coincides with your worst days.

• Breakdowns by hour, day and instrument to spot your winning and losing windows.
• Per-setup statistics to identify your most profitable configurations to repeat.
• Emotion crossed with performance to reveal the state that coincides with your losses.
• P&L calendar to see at a glance your patterns by day and week.
• Pre-session check-in to connect today's state to the emotional patterns you've identified.

Frequently asked questions