From a chart to a decision: what technical analysis actually does

Technical analysis examines recorded market behavior: where price traded, how movement developed, and how much trading activity accompanied it. Its professional purpose is not prophecy. It is to turn an observation into a testable hypothesis, with conditions that support the idea, conditions that invalidate it, and a defined cost of being wrong. A chart can help organize timing and risk; it cannot independently establish fundamental value, project quality, or the probability of a future announcement.

Separate observation from interpretation. “Price closed above the previous high” is an observation. “Buyers therefore have permanent control” is an unsupported interpretation. A testable alternative is: if price remains above the broken area and demand appears during a retest, we will examine continuation; if it moves back below the area and stays there, the hypothesis weakens or fails. This distinction makes the analysis easier to audit after the outcome is known.

Reliable data comes before lines and indicators

A candle records open, high, low, and close for a defined interval. Its body represents the difference between open and close; its wicks show the traded range. A long wick alone does not prove manipulation, institutional buying, or a reversal. Its location, liquidity conditions, and subsequent price behavior matter.

Record the symbol, spot or derivative market, data provider, and chart timezone before analyzing. Equity prices may need adjustment for splits and distributions. Continuous futures charts depend on their contract-roll methodology, which can create artificial discontinuities. Cryptocurrency volume on one venue is not necessarily representative of the whole market. Comparing dollar turnover with units traded, or daily candles with different timezone boundaries, without harmonizing the inputs can produce misleading conclusions.

A logarithmic chart may be more useful when the price range is very large and the question concerns percentage changes. Equal vertical distances then represent equal ratios rather than equal absolute differences. Keep the scale consistent: a trendline drawn on a linear chart may not carry the same meaning on a logarithmic one. Document the choice instead of changing it to make an existing interpretation look better.

Market structure: trend, range, and transition

An uptrend is commonly described by higher swing highs and higher swing lows; a downtrend by lower highs and lower lows. In the first diagram, HH means higher high and HL means higher low. This description is useful only when swing selection is consistent. Selecting a tiny wick in one section and an entire major wave in another makes the classification arbitrary.

Schematic, not market data: HH denotes a higher high and HL a higher low; a short pullback can occur within a larger trend.
Schematic, not market data: HH denotes a higher high and HL a higher low; a short pullback can occur within a larger trend.

A range repeatedly trades within an area, and unsuccessful breakouts are possible. A transition has lost the earlier structure without yet establishing a clear new one. Forcing an immediate bullish or bearish label during a transition can be premature. “The structure is currently unclear” is a legitimate analytical result, not a failure to produce a signal.

A lower low on a short timeframe does not automatically end a trend on a larger timeframe. A daily chart can advance while an hourly chart corrects. Choose a context timeframe, a decision timeframe, and, if needed, an execution timeframe. Switching repeatedly until a favorable view appears is confirmation seeking rather than a coherent multi-timeframe process.

Treat support and resistance as areas

Support and resistance describe areas where price previously reacted, not guaranteed barriers. Nearby highs, a significant low, or an area from which a sharp move began may deserve attention on the next visit. The width of the zone should reflect volatility and data precision. Adding decimal places does not create analytical precision.

Distinguish rejection, acceptance on the other side, and a brief penetration followed by a return. An intrabar move through a level is not equivalent to a close beyond it, sustained trading there, and a successful retest. Define “breakout” before seeing the outcome; an educational example is a decision-timeframe close beyond the area followed by a retest that holds. That definition still needs testing for the chosen market and horizon.

A former resistance area becoming support is also a hypothesis. A break followed by a hold above the area provides evidence for a role change; a return below it does not. The label should follow the behavior, not override it. The number of previous touches alone proves neither increasing strength nor inevitable exhaustion.

Volume and volatility: assess the quality of a move

Volume measures trading activity within the selected interval. It does not identify participants or their intentions. Every transaction has both a buyer and a seller, so high volume alone is not proof of capital inflow or a guaranteed advance. Expanding activity during a move out of a range can indicate broader participation; follow-through and closing location still need examination.

Compare volume with the same instrument's history and similar trading hours. The opening session and midday are not interchangeable. Cryptocurrency spot turnover and perpetual-contract turnover should not simply be combined without understanding their definitions. Tick volume in some markets measures recorded activity or changes rather than a centralized count of all traded units. State the limitations of the feed in the analysis.

Volatility measures the size of changes, not their direction. Average True Range, or ATR, summarizes true ranges over a chosen lookback and can help place distances in context. A fixed ATR multiple does not guarantee that a stop will survive. If volatility expands, an appropriate invalidation distance may widen and the corresponding position size may shrink. Increasing quantity is not a substitute for managing that risk.

Indicators: new information or repeated price information?

A moving average smooths prices, RSI measures the relative magnitude of recent gains and losses, and MACD tracks differences between moving averages. None observes the future directly. Lag, sensitivity to settings, and unsuccessful signals are normal properties of these tools.

Agreement among a moving average, RSI, and MACD does not necessarily provide three independent pieces of evidence: all are largely derived from the same prices. Assign each tool a role. Structure can describe context, volume participation, volatility typical distances, and a momentum measure the pace of movement. Adding tools without a specific question mainly adds visual complexity.

During a strong trend, RSI can remain in a region commonly called overbought or oversold for an extended period. Divergence may flag changing momentum, but it does not determine the timing of a reversal. A trade still requires an independent trigger, invalidation rule, and risk budget. The indicator should support the process rather than replace it.

A complete scenario using hypothetical numbers

The following example uses a fictional asset and an unleveraged spot purchase. It is not live market data or a trade recommendation. Price has repeatedly rejected an area between 104 and 106. It then closes above 106 and remains above the area during a retest. The educational hypothesis is continuation, not certainty that the breakout will succeed.

  • Context: the higher-timeframe structure is rising, and the 104–106 area was recorded in advance.
  • Example entry condition: a valid retest and a hypothetical entry at 107. If price moves away without the retest, this scenario has no entry.
  • Example invalidation: loss of the retest structure, with an illustrative exit level of 102.
  • Potential destination: 117, an area for evaluating an exit rather than a promise that price will reach it.
  • Price risk: 5 units; hypothetical reward: 10 units; reward-to-risk before costs: 2:1.

With hypothetical equity of 10,000 and an illustrative risk budget of 0.5%, planned loss is 50 currency units. If estimated round-trip costs and slippage are 0.25 per asset unit, quantity is 50 divided by 5.25, or approximately 9.52 units. Round down to the permitted order increment. This estimate is not a guaranteed loss ceiling: a price gap or execution failure can make the actual loss larger.

Hypothetical example: 104–106 area, conditional entry 107, exit 102 and destination 117; pre-cost reward-to-risk is 2:1.
Hypothetical example: 104–106 area, conditional entry 107, exit 102 and destination 117; pre-cost reward-to-risk is 2:1.

Write three branches in advance. If all conditions occur, assess a risk-sized position. If price returns inside the range, question the breakout hypothesis. If the planned entry never occurs and price moves away, do not chase it under this plan. Professional analysis provides a response to each branch instead of requiring every prediction to be right.

Testing and journaling: move beyond attractive screenshots

One successful historical chart does not establish a method's performance. Fix the market, period, timeframe, swing definition, and entry and exit rules before viewing results. Record unsuccessful examples and quiet periods too; excluding them makes the evidence look better than it is.

Using a future high or low to make a past decision introduces look-ahead bias. Repeatedly adjusting settings on the same sample can cause overfitting. Reserve data for an out-of-sample test, then examine the process in simulation or carefully limited implementation. Include fees, spreads, slippage, delays, and incomplete fills in the assumptions. A result without these inputs is only a simplified model.

Keep a pre-entry screenshot, the decision rationale, invalidation, position size, and net result in a journal. Evaluate decision quality separately from profit or loss. A profitable trade can violate the plan and expose the account to excessive risk; a losing trade can be a correct execution of a valid but unsuccessful hypothesis.

Practice exercise and final checklist

Choose a historical chart with future candles hidden. Mark only three major areas, then write a continuation scenario, a failure scenario, and a no-trade condition. Reveal the next candles and check which conditions actually occurred. Do not replace the original rules with a story that fits the completed chart.

  • Are the market type, data provider, timezone, and scale documented?
  • Are trend and swings defined consistently?
  • Do breakout and retest have operational definitions?
  • Do the tools answer different questions, or repeat the same price information?
  • Are invalidation, execution costs, and position size known before entry?
  • Is doing nothing an explicitly permitted outcome?

The Nexito principle: technical analysis becomes useful when a clear observation leads to a testable hypothesis and then to a decision with defined risk. The number of lines on the chart is not the measure of professionalism. Clear conditions and a willingness to accept invalidation are more important.