Technical analysis for beginners: what works and what doesn't

Fundamentals of technical analysis: three assumptions and Dow theory

Russian Wikipedia defines technical analysis as a set of tools that forecast a probable price change based on how the price moved earlier under similar circumstances. It looks at price charts and the order book. Company reports and news are not taken into account.

The fundamentals of technical analysis rest on three assumptions. The price has already accounted for everything known to the market. The price moves in tendencies, called trends: an uptrend occurs when demand exceeds supply, and there are also a downtrend and a sideways trend. And finally, history repeats itself, so in a similar situation the price behaves similarly.

The method's roots lie in Dow theory. It was derived from 255 editorial articles by Charles Dow (1851-1902) in The Wall Street Journal. Dow himself never used the words "Dow theory" and did not consider his ideas a trading system. The six principles were formulated after his death by Hamilton, Robert Rhea and Schaefer.

The weak spot is visible in the third assumption. If history does not always repeat itself, a rule that made money ten years ago may produce a loss tomorrow.

Technical analysis in trading has nothing to do with the analysis of technical documentation in manufacturing or with the technical analysis of text. The analysis of technological documentation is an occupational function from a professional standard, for example "technologist in the automotive industry". Laboratory research methods and the center for laboratory analysis and technical measurements also belong to a different field; there are no price charts there.

Technical market analysis: trend and levels on the chart

Technical market analysis begins with the chart. The raw material for all technical analysis charts is historical data: for each period, the opening price, high, low and closing price are taken. Candles are built from these, and the trend is visible from the candles.

Levels in technical analysis are prices from which the chart has already reversed. From below, such a level is called support; from above, resistance. On the conditional chart below, the price dropped to 100 twice and moved up. It rose to 110 three times and rolled back each time.

On the 20th candle, the price closed above resistance, at 111.8. A trader who trades a level breakout buys here and places a stop-loss below the level, at 108. The risk of such a trade is 3.8 points. Two candles later, the price dropped to 110.1, touched the level from above and then went to 116.4.

Entry and exit points on such a chart are easy to find in hindsight. Until the candle closes, no one knows whether the breakout will be real or false. That is why the stop is placed before entry, and the trade size is calculated from the amount one is prepared to lose. We discussed how to do this in numbers in the article about stop-loss and take-profit.

Methods of technical analysis: patterns, levels and indicators

Methods of technical analysis are usually divided into three groups. The first is reading the chart: trend, levels, lines. The second is figures (patterns) like the "head and shoulders" or "double bottom". The third is indicators, that is, formulas that recalculate past prices into a line or scale.

A beginner is told: analyze this, analyze that, and a week later there are six indicators hanging on the chart. It is more useful to understand how the two or three best-known ones are structured.

RSI (Relative Strength Index) was invented by J. Welles Wilder. He published it in 1978 in the book New Concepts in Technical Trading Systems and in Commodities magazine. The formula is: RSI = 100 - 100 / (1 + RS), where RS is the average price gain over n periods divided by the average decline over the same periods. The scale runs from 0 to 100, and the most popular period is 14. According to Wilder, after a rapid rise an asset becomes overbought, and after a rapid fall, oversold.

MACD was created by Gerald Appel in the 1970s. In 1986, Thomas Aspray added a histogram to it. The indicator has three parts: the MACD line, the signal line and the histogram. Moving averages are even simpler: it is the average price over the last N periods, which shifts along with the chart.

An indicator is an analyzer of past prices: it recalculates what has already happened. Any such analyzer consists of a formula and one or two settings. By fitting the settings to a past chart, it is easy to get a beautiful picture that says nothing about tomorrow's price.

Technical analysis patterns: what is known about them from the data

Technical analysis patterns are the hardest to verify: every trader sees them differently. Lo, Mamaysky and Wang in 2000 described it this way: geometric patterns on a chart are in the eye of the beholder.

To remove subjectivity, they taught an algorithm to find patterns on its own and ran it over American stocks from 1962 to 1996, that is, over 31 years. Their conclusion is cautious. A few patterns, such as "head and shoulders" and "double bottom", provide additional information about the price and may have practical value. Nothing is said about profit after commissions in the abstract of the paper.

A pattern adds information to the picture. Entering a trade based only on it, without a stop and without testing on history, means guessing.

Technical and fundamental analysis: what is the difference

Fundamental analysis studies the causes of price movement: a company's financial statements, the economy, interest rates. Technical analysis looks at the effect, at the price itself and its chart. In 2020, T-J divided them as follows: technical analysis is needed for trading and short-term investments, fundamental analysis is for long-term investors. For someone who holds a broad set of securities for years, technical analysis, according to the publication, is "absolutely optional".

Professionals use both. In 2010 Menkhoff surveyed 692 fund managers in the US, Germany, Switzerland, Italy and Thailand. The vast majority of them use technical analysis. Over a forecast horizon of several weeks it matters more to them than fundamental analysis; over longer horizons it gives way.

Which analysis you need depends on the horizon. If you hold stocks for years, you hardly need charts. If you open trades for days, you will have to work with price, levels and stops.

Does technical analysis work: what the research showed

The best-known test was done by Brock, Lakonishok and LeBaron in 1992. They took two simple rules, moving averages and trading-range breakouts, and ran them on the Dow Jones index over 1897-1986. The results supported technical strategies: after buy signals returns were higher than after sell signals.

In 1999 Sullivan, Timmermann and White expanded the set of 26 rules by Brock and co-authors. They ran it on 100 years of daily data for the same index, already adjusted for data snooping.

In 2007 Park and Irwin collected 95 modern studies. Of these, 56 found profits in technical strategies, 20 got a negative result, 19 a mixed one. Early work found profits on currencies and futures; on stocks there were none.

Shares of the 95 studies in Park and Irwin's review (2007), rounded to whole numbers. Verified 06.11.2026.

Most of these studies, according to the review's own authors, have flaws. Rules were fitted to the data, selected with hindsight, and risk and costs were hard to account for. So 56 out of 95 should be read as an upper bound.

The strictest test was by Bajgrowicz and Scaillet in 2012. They took the Dow Jones index over 1897-2011 and a new method for selecting rules. Conclusion: an investor could not have chosen the future best rules in advance, and the in-sample gains are fully eaten up even by low transaction costs. The authors seriously question the economic value of rules from early periods. The main limitations of technical analysis are visible from these studies: fitting to history and transaction costs.

Scientists' research methods are the same backtest a trader does, only adjusted for the number of variants tried. Without such an adjustment, almost any search over settings finds a "working" rule.

Technical analysis of stocks: US and Russia

Technical analysis of stocks in the US has been tested on long data series: the work of Brock, Lo and Bajgrowicz is built on the American market. There is little work on Russia.

Volodin and Kulagina of HSE in 2013 took the most liquid Russian stocks and six indicators: simple moving average, moving average crossover, MACD, RSI, stochastic and negative volume index. In most cases the indicators gave a negative result. On daily data the authors advise against using them, and the same came out in Volodin and Baulin's 2012 paper.

An exception was found in financial sector stocks. There, moving average crossover and negative volume index could produce a plus. The authors see the reason for the failure in the indicators' overly simple formulas.

The HSE paper is 13 years old; we did not find a more recent Russian study. The authors themselves wrote in 2013 that there are almost no serious academic papers on Russia. Those that existed disagreed: one showed weak efficiency, another found successful application on the MICEX spot market in 2007-2009.

Technical analysis in trading: what happens to beginners' money

Technical analysis in trading is needed primarily by those who hold a trade for a short time: from minutes to several days. The US Securities and Exchange Commission (SEC) calls day trading extremely risky: it can lead to substantial losses in a very short time.

In Brazil, economists Chague, De-Losso and Giovannetti studied everyone who started day trading stock-index futures in 2013-2015 and lasted more than 300 days. 97% of them lost money. More than a bank teller's daily wage, 54 dollars, was earned by 0.4%. The best made 310 dollars a day with a standard deviation of 2,560 dollars. The authors found no signs that people learn as they trade.

This sample included only those who did not quit trading within 300 days. The weakest were not included.

In Taiwan, Barber, Lee, Liu, and Odean analyzed day traders over 1992-2006. Predictably and consistently, fewer than 1% of them earned money after commissions. The top 500 of one year earned 37.9 basis points per day after commissions the following year, while the bottom ones lost 28.9 points.

In Russia, clients of forex dealers, according to their reporting, had a loss of 546.5 million rubles in Q3 2024, Kommersant wrote. This was the worst result of that year and the third-largest loss since 2017.

Trading can end in the loss of all invested money. Technical analysis does not remove this risk. It only helps to set a stop and calculate the position size.

How to test an approach before risking money

It is convenient to build technical analysis training around one habit: testing any rule on data. First, write down the rule so that another person could repeat it: where the entry is, where the stop is, where the exit is, what the position size is. The rule "I buy when the chart looks strong" cannot be tested.

Next comes history. Choose settings on the first part of the data and test them on the second part, which you did not touch. In MetaTrader 5, this is done by the forward test option. The period is divided into two parts: half, a third, a quarter, or up to a specific date. Optimization runs on the first part, then the best 10% of runs (with full enumeration) or 25% (with the genetic algorithm) are tested on the second.

Why such complications are needed was explained by Bailey, Borwein, Lopez de Prado, and Zhu in 2014. A high result on history is easy to obtain by trying just a few variants of settings. The more variants tried, the more likely overfitting is. Strategy authors rarely write how many variants they tried, so overfitting cannot be assessed from the outside. If prices have a memory effect, an overfitted strategy on average loses money on new data.

After history, move on to manual practice. TradingView has a Bar Replay mode for this: it plays back past candles one at a time, with no real financial risk. You can pause it and then return to the same place with the same settings. That way you trade the past without knowing what comes next, and you count how many trades you got right.

Then a demo account and a trading journal. The journal and the backtest are your diagnostic methods: they show where the approach breaks down and in which markets it does not work. We compared how a demo differs from a cent account and a prop challenge in the article about demo trading.

Technical analysis software: TradingView or MetaTrader 5

Two programs are enough for practice, each with its own purpose. TradingView is convenient for manual visual checking in Bar Replay mode. MetaTrader 5 is needed once the rule is already written down and needs to be run on history with a forward test.

If the description of a paid technical analysis program contains the word "analyzer", before paying, figure out what an analyzer means for that seller: an indicator, a market scanner, or a backtester. For a beginner, the built-in analyzers of TradingView and MetaTrader 5 are usually enough.

The platform also matters for a future challenge. Among the seven prop firms in the PMC catalog, TradingView as a platform is available only at FTMO. The others use MetaTrader 5, cTrader, TradeLocker and web terminals. FTMO is closed to traders from Russia, Belarus, Kazakhstan, Uzbekistan, Kyrgyzstan and Tajikistan, while it accepts traders from Armenia, Georgia, Azerbaijan, Moldova and EU countries. We covered what to replace it with in the CIS in the article about FTMO alternatives.

Technical analysis: which book a beginner should read

If you are looking for a complete course, it is more convenient to learn technical analysis from one thick textbook than from a dozen videos. T-J names John Murphy's book "Technical Analysis of the Futures Markets. Theory and Practice" as the main textbook. Murphy also has Technical Analysis of the Financial Markets, in Russian "Технический анализ финансовых рынков": in the Open Library catalog it is listed with the year 1999.

Jack Schwager in the book "Technical Analysis" separately examines trading systems and the analysis of their performance. This is closest to the topic of testing an approach. T-J's list also includes "Technical Analysis from A to Z" by Steven Achelis and "Entries and Exits" by Alexander Elder. One of the older books in the genre is Technical Analysis of Stock Trends by Robert Edwards and John Magee: in the Open Library, its AMACOM edition is listed with the year 1964.

We don't provide download links for pdf and fb2: those are pirated copies. The books are available in stores and libraries. Check what you've read the same way as any rule: on history and on demo.

From chart to prop account: when the approach has already been tested

A prop firm lets someone trade with its account if they pass a paid evaluation, a challenge. The trader trades on a simulated account, and the prop firm pays the payout from its own funds according to the terms of the contract. We described the scheme in detail in the article what is prop trading.

The challenge does not evaluate your method. It checks whether you reached the profit target and did not breach the daily and overall loss limit. That is why the approach first goes through history, a forward test and demo, and the challenge comes last, when you already know your drawdown under these rules.

In the PMC catalog, the cheapest two-step challenge for 10,000 is Starter way from The Funded Way at 92 dollars. There the targets are 6% and 6%, the daily limit is 3%, and the overall limit is 6%. At Era Trade and Hash Hedge, two steps for 10,000 cost 99 dollars each, at FTMO 89 euros, at SpiceProp 99 euros. The table lists first the prop firms with the highest overall rating on PropMarketCap.

Hash Hedge, Era Trade and SpiceProp accept traders from Russia and Belarus. The Funded Way sells its challenge to Russians, but payouts go through Rise, and Rise does not work with Russia and Ukraine. Because of its payouts, The Funded Way suits traders from Kazakhstan, Uzbekistan and Kyrgyzstan.

Before buying, read your program's rules in full: the free Academy lesson on challenge rules explains where to look for the daily limit, prohibitions and payout conditions. Check yourself against four questions.

Summary

Technical analysis helps describe a chart, set a stop, and calculate position size. It does not promise ready-made profit: in the Park and Irwin review, 56 of 95 studies found profit, and a 2012 check showed that the best rules cannot be selected in advance. So test every approach on history, on a holdout portion of the data, and on demo. The basics of prop trading and the challenge rules are covered in free PMC Academy lessons: 4 of 28 lessons are open without payment. Once the approach passes the test, pick a challenge in the calculator to match your budget and limits.