OPCIONARIO Options Encyclopedia
EN ES opcionsigma.com

Technical Analysis

ES: Análisis Técnico PT: Análise Técnica

The study of price and volume to make decisions: what it assumes, what it can and cannot do, and how to use it without slipping into magical thinking.

The three assumptions it rests on

Technical analysis starts from three explicit premises. First: price discounts everything — any relevant information, whether fundamental, macro or sentiment, eventually shows up in price, so studying price is an indirect way of studying everything else. Second: price moves in trends that tend to persist longer than pure randomness would predict. Third: history repeats, because patterns of collective behaviour — fear, greed, capitulation — are stable over time even as the participants change. It is worth being honest about their status: these are working assumptions, not proven laws, and their validity is partial and depends on the market and the horizon.

Cuatro familias, cuatro preguntas distintas Tendencia ¿hay direccióny qué fuerza tiene? medias · MACD · ADX Osciladores ¿está el movimientoextendido? RSI · estocástico Volatilidad ¿qué distanciaes normal aquí? Bollinger · ATR Volumen ¿hay participaciónreal detrás? VWAP · perfil Tres osciladores coincidiendo no son tres confirmaciones: son el mismo dato repetido IV Rank elige la estructura · el técnico elige los strikes

The tools and what each family is for

The toolkit falls into four families answering different questions. Trend indicators — moving averages, MACD, ADX — answer "is there direction and how strong is it?". Oscillators — RSI, stochastic — answer "is the move extended relative to its recent range?". Volatility indicators — Bollinger Bands, ATR — answer "how much is it moving and what distance is normal?". And volume indicators — VWAP, volume profile, accumulation — answer "is there real participation behind this move?". On top of these sit the visual structures: support and resistance, chart patterns and candlestick patterns. The most common error is stacking indicators from the same family, which say the same thing with a different face and produce a false sense of confirmation.

What it can actually do

With magical expectations set aside, technical analysis contributes four concrete and valuable things. Identifying the regime: distinguishing whether an asset is trending or ranging completely changes which strategy makes sense, and that diagnosis is probably its most useful contribution. Defining operational levels: where to place the stop, where to take profits, which strike to choose; even if the prediction fails, having objective levels prevents deciding in the emotion of the moment. Managing risk: ATR allows sizing a position by the asset’s actual volatility rather than an arbitrary figure. And imposing time discipline: rules dictating when to enter and exit reduce impulsive trading, which is the largest source of retail capital destruction.

What it cannot do

It does not predict the future, and any presentation suggesting otherwise is selling something. Patterns have hit rates that rarely exceed 60–70% under the best conditions, and many rigorous studies place them considerably lower. Nor does it work equally across horizons: evidence that signals carry information is weaker the shorter the timeframe, which is precisely where they are most used. And it carries two serious methodological problems: overfitting — tuning parameters until the system works perfectly on the past and fails the moment it goes live — and partial self-fulfilment, which makes some levels work only because many people are watching them, until they stop working when flows change.

How it combines with options

In options trading, technical analysis plays a different and considerably more defensible role than in pure directional trading: it defines the range, not the direction. Selling an iron condor requires a view on where price will not be, and there support and resistance levels, ATR width and trend regime are direct inputs into strike selection. The most robust combination in practice is two-layered: IV Rank decides which type of structure makes sense — buying or selling premium — and technical analysis decides where to place the strikes within that structure. Using it the other way round — letting a moving average crossover decide whether to buy expensive volatility — is the ordering that costs the most money.

Frequently Asked Questions

Does technical analysis actually work?
The academic evidence is mixed and depends heavily on the question asked. There is reasonable support for momentum as a return factor over multi-month horizons, and for the usefulness of historical volatility in position sizing. There is much less support for specific chart patterns and for very short-term signals. The defensible position is to use it as a management and level-definition framework, not as an oracle, and to distrust any claim of high hit rates without a large sample and costs included.
Which timeframe should I use?
The one matching your trade’s horizon, plus the next one up for context. If you sell 45-day options, the daily chart is the natural frame and the weekly gives you the regime; the five-minute chart is irrelevant noise for that decision. The common error is trading one horizon and analysing another: entering on a fifteen-minute signal for a position you will hold six weeks guarantees the signal expires long before the position does.
Technical or fundamental analysis?
They answer different questions and do not compete. Fundamental answers "what is this worth?" and helps decide what to trade and in which structural direction. Technical answers "what is price doing now?" and helps decide when and at what levels. Many options traders use fundamentals to select the universe, IV Rank to choose the structure, and technicals to place the strikes.
Are support and resistance real or a self-fulfilling prophecy?
Both, which is why they work intermittently. They have a real basis — zones where significant volume traded leave participants with positions to defend — and a self-fulfilling component: if enough people place orders at the same level, the level acts. The problem with the self-fulfilling part is that it exhausts itself: when a level becomes crowded, flow front-runs it and it stops behaving as expected. Treat them as zones with width, not exact lines.
How do I avoid overfitting a system?
Four tested measures. Use few parameters: each one added multiplies the combinations you can fit to the past. Hold back out-of-sample data you do not touch during development. Check that performance is stable under small parameter changes: if it works at 14 but not at 13 or 15, you have fitted noise. And always include commissions and slippage, because many systems profitable on paper stop being so once real costs are subtracted.