Price action means observing and interpreting how prices move: whether highs are rising, how far declines reach or what happens near a particular level. It is not one ready-made strategy. Turning an observation into a testable rule requires saying exactly what you are watching and when you make a decision.
Petr notices that bitcoin has stopped near a similar price several times. That observation can be a starting point for Price action. First, though, he needs to know what he is looking at. CME distinguishes a line of closing prices from a Candlestick chart, which also preserves the open and the period’s extremes. A brief move above a level can show up in a candle but disappear from a closing-price line. A different picture need not mean a different market. [1]
When Petr switches from minutes to days, several small rises become part of a single candle. CME explains that the Timeframe — the interval covered by each chart element — determines how prices are summarized. This is more than zooming in. For comparisons, keep the market, data source, interval length and boundaries consistent. Otherwise, “rising” may refer to something different each time. [1]
Petr marks an area where the price previously turned. Support describes a possible area of support during a decline; Resistance describes a possible obstacle during a rise. CME lists previous highs, lows and trend lines as reference points and emphasizes zones rather than precise barriers. Before a test, record the level and allowable tolerance. Moving it after seeing the outcome changes the question the test was meant to answer. [2]
The price briefly crosses Petr’s line and falls back. Was that a breakout? It depends on the rule. High is the candle’s highest price; Close is its closing price, although while the candle is open it is only the latest value so far. TradingView explains how these live values change. High can exceed a level while the completed Close remains below it. “Breakout” therefore needs a specified price measure and evaluation time. [3]
Completed charts often mark local highs precisely at the peak. Such a point is often called a pivot. But some methods confirm it only after subsequent candles and then draw the marker in the past, as TradingView demonstrates. At the peak itself, that confirmed pivot was not yet known. A test must use the time when it could be identified, not the position of a marker added later. [3]
Petr waits for a candle to close above the level. That establishes his price condition, not a purchase at exactly Close. He still has to send an order and have it executed. TradingView separates these steps even in its broker emulator, which simulates fills. The rule therefore needs timing, order type and execution assumptions. A historical model cannot by itself confirm what a real order would have achieved. [4]
Even if the price moves in the expected direction, costs may consume part of the difference. TradingView simulations can include commission — trading fees — and slippage, the difference between assumed and executed prices. Counting correct direction calls is therefore not the same as measuring results after costs. Petr needs to know not only where the price went, but also what the model assumed about entry and exit. [4]
It is easy to select only charts where a rule worked. TradingView calls this selection bias; overfitting means tailoring a rule too closely to particular historical data. One useful approach separates rule-development data from test data and makes no further adjustments during that test. Even a good historical result cannot guarantee future profit. The aim is to examine an idea honestly, not manufacture certainty from a chart. [4]
The wick crossed the level; the close did not
In a fictional rule, Petr waits for a completed Close above a preselected level of 100000 USD/BTC. The first candle has High 100020 and Close 99990 USD/BTC: the price crossed the level, but the condition was not met. The next candle closes at 100010 USD/BTC and meets the price condition. Its close is above the level by 100010 − 100000 = 10 USD/BTC. This only evaluates the rule; the example promises neither a purchase at that price nor a profit.
For the clearest picture, read this entry together with Technical analysis, Candlestick chart, Backtesting, Timeframe, Risk Management, Slippage. The reverse links also lead from Technical analysis, Timeframe, Candlestick chart, Support and Resistance.
01Is Price action one precisely defined strategy?+
No. It is an approach to studying price movement. A testable strategy also needs specified data, entry, exit, timing, position size and costs.
02Can a test enter earlier using a pivot drawn into the past?+
No, if the pivot was only confirmed later. At each moment, you may use only information available then. The marker’s location on a completed chart is not when you could have known about the pivot.