Updated May 2026: This review has been refreshed with newer research, a clearer summary of the Park and Irwin study, and a more practical trader’s view on when technical analysis helps or misleads.
Introduction
The question “does technical analysis work?” is harder to answer than most traders want it to be.
Some traders treat chart patterns and indicators as useless astrology. Others treat them as a shortcut to easy money. The truth sits somewhere messier: technical analysis can help in some markets and under some conditions, but it can also mislead traders who use it without context, testing or risk control.
This article reviews Park and Irwin’s 2004 paper, The Profitability of Technical Analysis: A Review, and adds newer research plus my own view after more than two decades trading futures professionally.
The aim is not to prove that every indicator works, or that every chart pattern is nonsense. It is to answer the more useful question: when does technical analysis help, when does it fail, and what should traders be careful about?
Does technical analysis work? The short answer
Technical analysis can work, but the evidence is mixed.
Academic reviews have found more support for technical trading rules in futures and foreign exchange markets than in stock markets. Some strategies have shown profitability in certain periods, especially trend-following rules such as moving averages and breakout systems. But the results often weaken once transaction costs, data snooping, changing market conditions and out-of-sample performance are considered.
That means technical analysis is better understood as a tool, not a guaranteed edge. It can help traders structure decisions, manage risk and understand market behaviour, but it does not remove the need for discipline, testing, position sizing and awareness of fundamentals.
My answer as a futures trader
After more than two decades trading futures professionally, my answer is: yes, technical analysis can work, but not in the lazy way it is often sold online.
I have used charts, MACD, DMI, market profile and order flow, but I have never treated technical analysis as a standalone truth machine. For short-term futures trading, I have found it most useful as a way to frame risk, identify where other traders may be positioned, and decide when not to trade.
The danger is when beginners treat an indicator reading as an instruction. “Overbought” does not mean a market must fall. “Oversold” does not mean it must rise. A strong trend can punish anyone who blindly fades it.
So the better question is not simply “does technical analysis work?” It is: what market are you trading, what timeframe are you using, what costs are involved, and are you using technical analysis as part of a broader process or as a magic signal generator?
Why technical analysis fails for many traders
Technical analysis often fails not because every chart signal is useless, but because traders use it badly.
The most common mistake is treating an indicator as a command. RSI says overbought, so they sell. Price hits a moving average, so they buy. A support line breaks, so they chase. None of those actions is automatically wrong, but none is automatically right either.
Markets trend, reverse, trap, squeeze and drift. A signal that works in a clean trend can be chopped to pieces in a sideways market. A setup that looks good before a major central bank announcement can become irrelevant once the news hits.
The other problem is hindsight. Charts look obvious after the event. In real time, the trader has to deal with incomplete candles, spreads, slippage, position size, emotions and the possibility that the next tick invalidates the whole idea.
That is why technical analysis is best used as a decision framework. It can help answer questions such as:
- Where has price rejected before?
- Where might stops be clustered?
- Is the market trending or ranging?
- Is volatility expanding or contracting?
- Where is my trade idea clearly wrong?
Those are useful questions. They are very different from pretending that a chart pattern guarantees the next move.
When technical analysis is most useful
Technical analysis is most useful when it helps a trader define risk and behaviour.
For short-term traders, charts can show where momentum is building, where a market has stalled, and where other traders may be forced to react. For swing traders, technical analysis can help identify trends, breakouts, failed breakouts and areas where the risk-reward is becoming more attractive.
It is less useful when it becomes detached from context. A chart signal ahead of Non-Farm Payrolls, CPI, a central bank decision or an earnings release is not the same as a chart signal on a quiet afternoon. The event can overwhelm the setup.
This is why I prefer to think of technical analysis as one layer. Order flow, fundamentals, positioning, volatility, market structure and upcoming catalysts all matter. Technical analysis can help organise the trade, but it should not be asked to carry the entire decision.
Background of the Study
The Origins of Technical Analysis of the Financial Markets
Technical analysis has a long history, with its roots in the West dating back to the early 20th century yet it was being used as far back as the 1600s in Japan. Over the years, it has evolved from a simple method of analyzing price charts to a sophisticated discipline that incorporates a wide range of techniques and indicators.
Technical analysis long predates the efficient markets hypothesis. The EMH became important later because it gave academics a framework for arguing that past prices should not provide easy, repeatable profits. Much of the research on technical analysis can be read as a test of that claim: if price-based rules make money after costs and proper statistical checks, markets may not be fully efficient in the weak-form sense.
The Purpose of the Study
The Park and Irwin paper was written to review the existing evidence on whether technical trading rules had been profitable across stock markets, futures markets and foreign exchange markets.
Rather than treating technical analysis as one simple yes-or-no question, the review looked at different types of evidence: survey studies, theoretical models and empirical tests of trading rules. It also paid close attention to the weaknesses in many studies, including data snooping, ex-post rule selection, transaction costs and out-of-sample performance.
The purpose of the review was therefore not simply to find profitable chart patterns. It was to ask whether the published evidence for technical analysis was reliable, whether it varied by market, and whether the results challenged the weak-form efficient markets hypothesis.
Why This Study Still Matters in 2026
Although Park and Irwin’s review was published in 2004, it still matters because many of the same questions remain unresolved today. Can price-based trading rules produce returns after costs? Do those returns survive out-of-sample testing? Are profitable rules genuinely predictive, or are they the result of testing too many variations until something appears to work?
Those questions have become even more important in 2026. Retail traders now have free charting tools, automated scanners, broker APIs, social-media trading ideas, and AI-assisted coding tools. It is easier than ever to build and test a technical strategy, but it is also easier than ever to overfit one.
The key lesson from the original review is not that every technical indicator works. It is that technical analysis has shown stronger evidence in some markets and periods than others, especially in foreign exchange and futures, while many stock-market results become weaker once transaction costs, data snooping and changing market conditions are considered.
Recent studies still reach mixed conclusions. Some find useful technical signals in specific markets, while others show how quickly apparent profits disappear after data-snooping checks, out-of-sample testing and transaction costs.
That is why this older review remains useful. It gives traders a framework for thinking clearly about technical analysis: not as magic, not as nonsense, but as a set of tools whose usefulness depends on market, timeframe, costs, testing method and trader behaviour.
Evolution of Technical Analysis
Technical analysis has evolved significantly over the years, with advancements in technology and the development of new theories contributing to its growth. The study categorizes modern technical analysis into several distinct studies:
| Category | Description |
|---|---|
| Standard Studies | Typically conduct parameter optimization and out-of-sample tests. |
| Model-based Bootstrap Studies | Focus on resampling techniques to assess the profitability of trading rules. |
| Genetic Programming Studies | Use genetic programming techniques to optimize trading rules. |
| Reality Check Studies | Evaluate the performance of multiple trading rules to identify the best-performing ones. |
| Chart Pattern Studies | Concentrate on recognizable patterns in price charts. |
| Nonlinear Studies | Investigate complex, non-linear relationships in market data. |
The evolution of technical analysis has also been influenced by the development of various theoretical models already touched on. These models provide a theoretical foundation for the profitability of technical trading rules.
Benefits of Technical Analysis
The best case for technical analysis is not that it predicts the future with certainty. It is that it can help traders organise market information.
Used properly, technical analysis can help define trend, range, momentum, volatility and risk levels. It can also help traders decide where a trade idea is wrong, which is often more useful than trying to predict exactly where price will go next.
That is different from saying every technical signal is profitable. A moving average, RSI reading or chart pattern still needs context. Market regime, liquidity, news events, transaction costs and position sizing all matter.
Methodology
The Approach of the Study
The study’s approach comprehensively reviews survey, theoretical, and empirical studies regarding technical analysis and discusses the consistency and reliability of technical trading profits across markets and over time. The report pays special attention to testing procedures used in empirical studies and identifies their salient features and weaknesses. It aims to improve the general understanding of the profitability of technical trading strategies and suggest directions for future research.
The empirical studies surveyed include those that tested technical trading systems, trading rules formulated by genetic algorithms or some statistical models (e.g., ARIMA), and chart patterns that can be represented algebraically (Page 5).
The Types of Trading Rules Examined
The study examines various types of trading rules. Early models in this area provided the first theoretical foundation for the possibility of profitable technical trading rules by taking account of the speed and efficiency with which a speculative market responds to new information. These models hypothesized two types of traders, “insiders” and “outsiders,” and the dynamics between them (Page 13).
The study also examines the profitability of technical trading rules in various different types of markets. The results varied greatly between them. For example, in the early studies, very limited evidence of the profitability of technical trading rules was found in stock markets, yet they often realized sizable net profits in futures markets and foreign exchange markets (Page 27).
It also looks into genetic trading rules, a sophisticated approach in financial markets that leverages genetic programming. Genetic programming is a form of artificial intelligence that simulates the process of natural evolution. This method is used to autonomously develop rules for trading – essentially, strategies for when to buy or sell financial assets like stocks or currencies.
One of the key advantages of using genetic programming in this context is its potential to sidestep the pitfalls of data snooping. Data snooping occurs when analysts inadvertently tailor their strategies based on the idiosyncrasies of the specific data set they are studying, rather than finding genuinely predictive patterns. This often happens when too many strategies are tested on the same set of data, leading to seemingly significant findings that are, in reality, the result of random fluctuations. By using genetic programming, the development of trading strategies is less influenced by human biases or preconceived notions, as the program evolves these strategies based on their actual performance data.
However, this innovative approach is not without its own conundrum. The study points out a potential methodological issue with relying on genetic programming. The concern lies in claiming predictability in financial markets using a method that did not exist during the period being analyzed. For instance, if a genetic algorithm uncovers a pattern that appears to have been predictable in the 1980s, it’s contentious to claim this predictability when the algorithm itself, along with the computational resources required, only became available much later(Page 39). This could lead to a misleading conclusion, as it assumes the availability of advanced tools and analysis methods in a time when they were not actually present.
In essence, while genetic programming presents a novel and less biased way of developing trading rules, thereby potentially avoiding the trap of data snooping, there remains a critical consideration. It is crucial to contextualize the findings within the time frame and technological capabilities of the period being studied, to avoid anachronistic conclusions about market predictability.
Methodology Limitations
The study acknowledges the challenges and limitations in its process.
Some of these include:
- Data Snooping: This refers to the misuse of data analysis to find patterns in data that can be presented as statistically significant, thus leading to unreliable study outcomes.
- Ex Post Selection of Trading Rules or Search Technologies: This refers to the selection of trading rules after the fact, based on what has already happened. This can lead to overfitting, where a model is tailored to fit the historical data perfectly but performs poorly on new data.
- Difficulties in Estimation of Risk and Transaction Costs: Accurately estimating risk and transaction costs is a challenging task. Misestimation can significantly impact the results of a study.
The document suggests that future research must address these deficiencies in testing to provide conclusive evidence on the profitability of technical trading strategies (Page 2).
The Impact of Time and Market Changes on the Methodology
As we review this study more than twenty years after its publication, it’s crucial to consider how time and market changes have impacted the methodology and findings of the study. The study itself acknowledges the evolving nature of financial markets and the impact this has on the effectiveness of technical trading rules.
The review was published during a period when financial markets were already changing quickly. Electronic trading was growing, financial instruments were becoming more complex, and market structure was evolving. Even so, there was still far less HFT and algorithmic activity than traders face today, and much more trading was still visibly human-driven.
Key Findings
The Main Discoveries of the Study
“The Profitability of Technical Analysis: A Review,” presents several significant findings that shed light on the effectiveness of technical analysis in trading. Here’s a breakdown of the results across different markets:
| Market | Technical Method | Result | Reference Page |
|---|---|---|---|
| Foreign Exchange Markets | Single moving average rules | Demonstrated significant profitability, especially when considering long positions in certain currency pairs. | Page 104, Page 86 |
| Futures Markets | Dual moving average crossover rules | Showed promising results, indicating potential trend-following opportunities. | Page 104 |
| Stock Markets | Various technical trading rules | Found mixed results, with some rules being more effective than others. | Page 27, Page 39 |
These findings suggest that the profitability of technical analysis may vary based on market conditions, the asset class, and the specific technical method employed.
The Profitability of Technical Analysis
As we see above, certain technical analysis strategies, such as single moving average rules, were particularly profitable (Page 104). This suggests that traders who are able to effectively use these strategies may be able to generate significant profits.
However, the study also notes that the profitability of technical analysis can vary depending on a variety of factors, including the specific market conditions and the type of trading position that is taken e.g. as we note in the above section where long only positions were considered for certain FX pairs.
Technical analysis may be particularly profitable in markets where information is rapidly changing and traders need to quickly adjust their strategies in response to new data.
The Question: “Can Technical Analysis Make You Rich?”
The study does provide insights that can help us infer an answer. The authors found that certain technical analysis strategies, such as single moving average rules, were particularly profitable (Page 104). This suggests that traders who are able to effectively use these strategies may be able to generate significant profits. However, the profitability of these strategies can vary depending on a variety of factors, including the specific market conditions and the type of trading position that is taken. For example, the study found that there was a marginal improvement to five and four currencies for moving average rules and channel rules, respectively, when only long positions were considered (Page 86). This indicates that while technical analysis can potentially lead to wealth accumulation, it is not a guaranteed path to riches and requires skill, knowledge, and the right market conditions.
The Long-Term Sustainability of Profits from Technical Analysis
While technical analysis can potentially generate profits in the short term, the sustainability of these profits over the long term is uncertain and likely depends on a variety of factors, including changes in market conditions and the ability of traders to adapt their strategies in response to these changes.
Implications for Traders
The Practical Application of the Findings
The study’s findings have practical implications for traders. The research suggests that technical analysis can be profitable, especially when applied to foreign exchange and futures markets. The study found that single moving average rules generated the best results, followed by dual moving average crossover rules and relative strength index rules (page 104). These findings suggest that traders can potentially use these specific technical analysis methods to guide their trading decisions.
The Potential Risks and Rewards of Using Technical Analysis
While the study provides evidence of the profitability of technical analysis, it’s important to note that there are potential risks involved. The study categorizes modern technical analysis studies and highlights the risks associated with each category (page 82). For instance, some studies lack trading rule optimization and out-of-sample tests, and do not address data-snooping problems. These issues can lead to inaccurate predictions and potential losses for traders.
However, the rewards can be significant. The study found that technical analysis rules generated positive returns including foreign exchange and futures markets. For instance, when only long positions were considered, there was a marginal improvement to five and four currencies for moving average rules and channel rules, respectively (page 86).
The Question: “Does Technical Analysis Actually Work?”
The study provides evidence that technical analysis can work, but it’s not a guaranteed strategy for success. The research suggests that the effectiveness of technical analysis can vary depending on the market and the specific technical analysis method used. For instance, the study found that technical trading rules formulated by genetic programming appeared to be unprofitable in stock markets, particularly in recent periods. In contrast, these rules performed well in foreign exchange markets, with their performance decreasing over time (page 39).
The study also suggests that the success of technical analysis may depend on the trader’s ability to interpret and apply the analysis correctly. The study notes that ill-qualified traders who have little opportunity to acquire valuable information early and little ability to interpret the information may choose to “go with the market” (page 13). This suggests that education and experience can play a significant role in the success of technical analysis. Keeping abreast of fundamental releases will assist traders with interpreting when to react to what technical analysis signals suggest and when to hold off executing solely on their basis.
Critiques and Limitations
The Criticisms of the Study
We touched on this in the Methodology section earlier. While the study provides valuable insights into the profitability of technical analysis, it acknowledges several limitations in its testing procedures, which could potentially impact the validity of its findings. Let’s address these three areas again:
One of the main criticisms of the study is related to data snooping. Data snooping refers to the misuse of data analysis to find patterns that can be presented as statistically significant, leading to unreliable study outcomes (page 2). This issue is particularly relevant in the context of technical analysis, where the effectiveness of trading rules is often evaluated based on their past performance. The risk of data snooping could potentially lead to overestimation of the profitability of technical analysis.
Another criticism is related to the ex post selection of trading rules or search technologies. The study acknowledges that this could potentially lead to overfitting, where a model is tailored to fit the historical data perfectly but performs poorly on new data (page 2). This could potentially overstate the effectiveness of technical analysis and lead to inaccurate predictions about future market trends.
The study also acknowledges the difficulties in estimating risk and transaction costs accurately. Misestimation can significantly impact the results of a study and can potentially overstate or understate the accuracy of technical analysis (page 2). This is a significant limitation, given that the profitability of technical analysis is often evaluated based on the assumption that trading rules can be implemented without any transaction costs.
In conclusion, while the study provides valuable insights into the profitability of technical analysis, these criticisms highlight the need for caution when interpreting its findings. Traders and investors should be aware of these limitations when using the study’s findings to guide their trading strategies.
What newer research adds
The Park and Irwin review is still useful because it explains the main academic debate: technical analysis appears to work better in some markets than others, but many studies are vulnerable to data snooping, transaction cost assumptions and weak out-of-sample testing.
More recent research has not produced a simple yes-or-no answer. It has mostly reinforced the more cautious conclusion: some technical rules can work in some markets, but the edge is highly dependent on the market, period, rule design and trading costs.
For example, a 2023 study of 14 currency pairs tested filter rules, trading range breakouts, moving averages and Bollinger Bands. It found mixed results, but reported that the 1% filter rule, 150-day moving average and Bollinger Bands performed better than the other techniques tested, with trading range breakout rules adding value in strongly trending markets.
A 2024 study of Chinese stock market indices tested 38,456 trading rules and specifically controlled for data-snooping bias. Its conclusion was much more cautious: only a few complex rules outperformed the Shanghai Composite in some sub-periods, no rules outperformed the Growth Enterprise Market, and many apparently lucrative rules became negative out of sample after transaction costs.
That contrast is useful. It suggests technical analysis should not be discussed as one thing. A moving-average trend rule in foreign exchange is not the same as a high-turnover stock-market rule after costs. Market, timeframe and implementation matter.
A Professional Trader’s View
Personally, after 21 years trading futures professionally, and having started before the original study reviewed here was compiled, I would say technical analysis has made me a lot of money. But it’s not necessarily a requirement for traders to use it to be profitable, nor even look at a simple chart either for that matter. I certainly haven’t purely relied on it either, I’ve always used it in combination with analysing fundamental events and order flow.
As a short-term futures trader with seconds to minutes hold times, I’ve tended to use a combination of order flow on a DOM (depth of market), market profile and 15 minute and 60 minute candles with MACD and DMI but that’s just me. I know an extremely successful gasoil prop trader who I worked on one of the same trading floors as I did, who only ever looked at order flow via his WebICE platform. He found that charts overcomplicated his thinking when trading front month gasoil spreads, so he never used them. He is still hugely profitable and continues trading the same way today.
Many traders starting out look for some sort of holy grail indicator, some secret sauce, whereas I think it is probably more important to focus on the dynamics of intermarket relationships and fundamentals such as upcoming economic figure releases. Too many times people cry ‘overbought’ or ‘oversold’ due to indicator readouts and helplessly keep selling or buying repeatedly against a stubborn trend. Never outsource the blame of a losing trade to technical analysis being wrong. It’s a tool in a patchwork of considerations, you always own the position. There are a thousand ways to skin a cat and more than a thousand ways to pull money from the markets but the more you learn what’s available to assist you, the more you can settle into finding your own strategy that works for you.
Understanding who else is in the market and what they are looking at will help you exploit them and avoid them exploiting you, be they man or machine. Quite often chart patterns can be a self-fulfilling prophecy and maybe you can find ways to flush others out of positions to your benefit when you figure out where their stops would be. Certainly HFT algos visibly undertake these tactics every day if you watch closely.
Key Takeaways:
- Technical analysis can work, but the evidence is mixed and depends heavily on market, timeframe, costs and testing method.
- Park and Irwin’s 2004 review found stronger evidence for technical trading rules in foreign exchange and futures markets than in stock markets.
- The study remains relevant in 2026 because traders now have more tools than ever, but also more ways to overfit strategies and fool themselves with backtests.
- The biggest research problems are data snooping, ex-post rule selection, poor out-of-sample testing, and underestimating transaction costs.
- Newer research still does not give a simple yes-or-no answer. Some technical rules perform well in certain markets, while others fail once costs and more rigorous testing are applied.
- From a practical trading perspective, technical analysis is best used as a risk and decision framework, not as a magic signal generator.
- The trader still owns the position. A chart can help define the trade, but it cannot remove the need for judgement, risk control and market context.
Resources
You can download the full 106 page study as a free pdf file here:
Park, Cheol-Ho and Irwin, Scott, The Profitability of Technical Analysis: A Review (October 2004). AgMAS Project Research Report No. 2004-04, Available at SSRN: https://ssrn.com/abstract=603481 or http://dx.doi.org/10.2139/ssrn.603481
“Technical Analysis of the Financial Markets: A Comprehensive Guide to Trading Methods and Applications” by John J Murphy
“Technical Analysis from A to Z” by Stephen Achelis (Second Edition)
Technical-trading-rule profitability, data snooping and reality check: Evidence from the foreign exchange market
ScienceDirect: https://www.sciencedirect.com/science/article/abs/pii/S1042444X23000270
This is the 2023 FX study testing filter rules, trading range breakouts, moving averages and Bollinger Bands.
Profitability of technical trading rules in the Chinese stock market
ScienceDirect: https://www.sciencedirect.com/science/article/abs/pii/S0927538X24000295
This is the 2024 Chinese stock market study testing 38,456 trading rules and controlling for data-snooping bias. It is also listed on RePEc here: https://ideas.repec.org/a/eee/pacfin/v84y2024ics0927538x24000295.html
