Written lesson
Lesson transcript
Does technical analysis work? For decades, academics and traders have fought over this. One side says charts predict the future. The other says it's no better than astrology. Here is a single reframe that helps you approach this in a systematic way. Technical analysis does not predict the future. It helps you read the present. This single mindset will help you understand what charts actually reveal and what they cannot. We will approach this through behavioral science, academic research, and probability mathematics. My name is Ashim Nandi. I'm a system architect with ten years of experience between running an IT company and full time trading. Charts are like system logs. They don't predict what will happen. They reveal what is happening now. Here are the chapters that will help you understand this concept better. First, the behavioral foundation. You will understand why price patterns form in the first place. Second, the academic evidence. You will know what the research actually shows about technical analysis effectiveness. Third, the systematic framework. You will have a method for using charts as decision support tools rather than crystal balls. Once you go through all of these concepts, you will approach technical analysis systematically… rather than mystically. And with that understanding, you will also see how charts integrate with everything we have built so far. Risk management, position sizing, expected value, volatility, liquidity, market structures, and market regimes. Think in odds. Act with discipline. Let's begin. Chapter one, the behavioral foundation…Why patterns exist? Price patterns are not mystical formations. They are visual representations of aggregate human psychology. And we humans have decades of behavioral finance research including two Nobel prizes that explains exactly why. The first mechanism, anchoring bias. Kenman and Tversky demonstrated that humans fixate on reference points when making decisions. In trading, these reference points become support and resistance levels. Previous highs, previous lows. Round numbers, purchase prices. Other researchers, Donaldson and Kim confirmed that multiples of hundred act as psychological barriers in the Dow Jones. Not because of magic, because traders collectively anchor to these prices and act accordingly. When enough traders believe a level matters, it matters. The concentrated buying or selling pressure makes it real. The second mechanism… prospect theory. Kenman and Tversky proved that losses feel twice as painful compared to gains. This two to one pain ratio isn't theory. It's a measurable neural response. In practice, this is what happens. After gains, traders become risk averse. They sell too early. They lock in profits. After losses, they become risk seeking. Hoping to break even, they hold too long. This creates predictable selling pressure around breakeven points. And it explains why trend persist. Loss aversion prevents timely selling. The third mechanism, the disposition effect. Sheffrin and Staitman documented that traders realize gains at a fifty percent higher rate than losses. This explains…why price hesitate at previous resistance where many bought, and why breakdowns from support accelerates forced selling when pains from losses exceed tolerance. It is not mysticism. It is measurable human behavior…repeating across millions of trading decisions. The fourth mechanism, herding. Information cascades cause early investor actions to be observed and followed. This generates self reinforcing momentum. Research shows herding intensifies during market stress. The two thousand and eight crisis, COVID nineteen, both triggered pronounced herding behavior. This is why trends once established tend to persist beyond what fundamental changes would justify. George Soros calls this reflexivity. Perceptions influence actions. We change reality. We change perception. A feedback loop. Technical analysis captures these reflexive dynamics. The patterns are not predicting the future. They're revealing the psychology happening now. But understanding why patterns form does not tell you whether they will work in the future. That requires looking at the evidence, which is what chapter two gives us. Chapter two, What the research shows. The most comprehensive analysis comes from Perk and Irwin's two thousand and seven meta study. They reviewed ninety five modern academic studies on technical trading strategies. The results? Fifty nine percent showed positive outcomes, twenty one percent negative, twenty percent mixed. That sounds promising, but there's a critical detail. Profitability declined over time in developed markets. Early research found technical analysis profitable in forex and futures. Modern studies showed strategies working at least until the early nineteen nineties. Markets became more efficient as strategies became known. The edge got arbitraged away. Andrew Lowe at MIT provides the intellectual framework that explains this. His adaptive markets hypothesis connects both sides of the debate… Markets aren't either efficient or inefficient. Efficiency varies over time based environmental factors, number of competitors, magnitude of profit opportunity…participant adaptability. In Lowe's own words, investment strategies will perform well in certain environments and poorly in others. This explains why technical analysis profitability declined in developed markets while persisting in emerging ones. Low's research also validated something important. Using nonparametric statistical methods, his team tested chart patterns rigorously. Head and shoulders, double tops, rectangle bottoms, these patterns showed statistical significance. His key insight, the distinction between technical analysis language, support and resistance levels, and academic language autocorrelation patterns is primarily linguistic, not substantive. Both describe identical statistical phenomena. The strongest academic support exists for momentum strategies. Jagdish and Titman's nineteen ninety three research established that buying recent winners and selling losers produces approximately one percent monthly excess returns. This effect has persisted across decades, international markets, and in multiple asset classes. Their two thousand and twenty three follow-up concluded, the widespread nature of momentum returns provides perhaps the strongest evidence against the efficient market hypothesis. Momentum is essentially systemized trend following. It is technical analysis validated by academic discipline. Now the critics deserve acknowledgement. Nassim Taleb called technical analysis no better than astrology. Burton Malkiel argued a blindfolded monkey throwing dirts could match expert stock pickers. Eugene's efficient market hypothesis directly attacks the premise. Here's the key, these critics are attacking predictive claims and on prediction, they have substantial ground. Chars cannot reliably forecast future prices. Data mining creates illusory edges. Transaction costs erode returns. Survivorship bias inflates expectations, but sophisticated practitioners have largely abandoned prediction for something different. Probability assessment, risk calibration regime identification. Critics attacking fortune telling and practitioners focused on present momentum awareness are often talking past each other. Both can be correct, but knowing the evidence still does not tell you how to use technical analysis systematically. That requires the right framework… and which is what exactly chapter three delivers. Chapter three, the systematic framework, reading, not predicting. Senior strategist James Stanley articulates the thesis directly. Technical analysis does not work because it does not have a job. It is simply a form of analysis, a way of looking at something. But the trader has a job, and the job is to manage risk. And for that, technical analysis can be helpful. This is the reframe. Charts do not predict the future. They help you read the present. Application one, regime identification. This may be the technical analysis highest value use. Markets spend approximately seventy percent or more of the time in ranging conditions. Only fifteen to twenty percent of the time do they trend. A trend following strategy in a ranging market loses money. A mean reversion strategy in a trending market gets crushed. Technical tools identify which regime is active now, not which regime will come next. ADX measures current trend strength. Bollinger band shows current volatility compression. Higher highs and higher lows reveal current bullish structure. The institutional approach uses hidden mark off models and statistical methods, but the underlying principle is the same. Match strategy to current conditions. That is situational awareness, not forecasting. Application two, probability calibration. Thomas Wilkoski built a database of over thirty thousand chart patterns. His research reveals critical context. A trader expecting to make forty percent or more on downward breakouts can expect success only about twenty six percent of the time… only in four trades. Failure rates have increased over decades. Failures in nineteen ninety one were a third of what they were in two thousand and seven. Long patterns outperform short ones. Heavy breakout with volume improves performance. These aren't predictions. They are base rates. The systematic trader uses pattern statistics to establish…prior probabilities. Current price action provides likelihood evidence. Probability is continuously updated. This is Bayesian thinking applied to charts. Not certainty, probability calibration. Application three, risk calibration. Technical tools measure current conditions for position sizing. ATR, average true range, measures current volatility, not future volatility, current. In our position sizing video, we used ATR to adjust position size. Higher ATR means smaller position. Lower ATR means larger position. Support and resistance levels define stop losses. Not because price will definitely reverse there, but because these levels represent current psychological concentration points. If support breaks, the thesis changes. Exit makes sense. If support holds, the thesis remains valid. Stay in the trade. Application four, integration with expected value. Remember the expectancy formula from our earlier video? Expectancy equals win rate times average win minus loss rate times average loss. Technical analysis provides inputs for this calculation. Pattern statistics inform win rate estimates. Support and resistance distance defines risk reward ratios. Volatility measures calibrate realistic profit targets. A forty percent win rate with one is to three risk and reward produces sixty dollars profit per trade on an average. A sixty percent win rate with one is to three risk and reward produces only twenty dollars. Win rate does not equal profitability. Expectancy does. Technical analysis contributes to expectancy calculation… not by preceding outcomes, but by providing the probability context. Let me show you how this integrates. You identify a potential trade. Market structure shows an uptrend. Higher highs, higher lows. That's current regime. Trending. A pullback to previous resistance, now support, presents entry opportunity. ATR is elevated compared to the twenty day average. That's volatility context. Reduced position size. Pattern statistics suggest similar setups work fifty five percent of the time with two is to one reward and risk. Now calculates expectancy. Fifty five percent times two r minus forty five percent times one r. That's one point ten r minus zero point five r plus zero point sixty five r per trade. Positive expectancy. The trade is worth taking. But notice what technical analysis provided, current regime identification, entry context, stop loss placement, probability estimate, risk and reward calculation. Not a prediction. A probability framework. Reading the present, not forecasting the future. This is how quantitative traders use technical analysis as one input among many. Filters, not signals. Regime identification. Execution timing. The CFA Institute notes that the technical analysis community split in the last twenty years with a great big piece cleaving off and calling themselves quants. They took a large body of knowledge from the technical approach, made it disciplined, and backed it with higher math and statistics. That's the systematic path. Technical analysis is neither astrology nor alchemy. It's a toolkit for reading current conditions, establishing probability contexts, and managing risk, not predicting the future. The behavioral science explains why patterns form. Anchoring, prospect theory, disposition effect, herding, human psychology made visible on a chart. The academic evidence shows what works and what doesn't. Momentum has the strongest validation. Probability has declined as markets became efficient. Predictive claims deserve skepticism. The systematic framework provides the application, regime identification, probability calibration, risk management, integration with expected value. In our first principles of trading series, we are building a complete system. Risk management, position sizing, expected value, volatility, liquidity, market structure, market regimes… and now technical analysis. Now that you have this framework, the next question becomes technical analysis reads price. But what about the value? When price separates from fundamentals, where does the advantage come from? That is what fundamental analysis gives us. The other lens. Reading value, not just price. Think in odds. Act with discipline. See you in the next one