Candle Range Theory (CRT): Reference Candles and Testing Rules
Learn how traders use a prior candle range as a reference in some ICT approaches, how to define a sweep-and-return rule, and why candle ranges do not imply an algorithmic cycle or guaranteed next move.
Framework note: ICT terms are interpretations of price action, not proof of institutional motives or a market algorithm’s intent. No setup is guaranteed. Test clearly defined rules, account for costs and slippage, and use risk limits before risking capital. Read our editorial policy.
Candle Range Theory (CRT) is a term used in some ICT education to compare price action with the high-low range of a reference candle. Some variants interpret later price movement as a sweep and return, but each candle is not proven to contain a separate algorithmic cycle and no three-phase pattern is required to appear.
The CRT Framework — How Each Candle Delivers
In one CRT interpretation, traders look for price to move within a candle’s range in ways that resemble the AMD model. The first phase is the accumulation of the candle — price moves relatively sideways or slowly after the candle's open, establishing the initial range. The second phase is manipulation — price makes a move to one side to collect liquidity (this is the wick of the candle). The third phase is distribution — price moves decisively in the actual direction of the candle, closing near its high (bullish candle) or near its low (bearish candle).
Some ICT interpretations treat a bearish candle upper wick or bullish candle lower wick as a possible sweep. A wick alone does not show that stops were present, identify who traded, or prove that the next move will continue in the opposite direction.
CRT and the Previous Candle's Range
CRT is commonly studied by comparing a candle with the previous candle range. The current candle open can serve as a reference, but it does not prove that an algorithmic delivery cycle has begun or that price must test either side of the prior range.
In a bullish context, the first move after a candle opens will often be a sweep of the previous candle's low — the algorithm collects the sell-side liquidity before the true bullish delivery upward. In a bearish context, the algorithm first sweeps the previous candle's high before delivering lower. Understanding this allows you to anticipate the early-session wick direction and not get stopped out by what appears to be adverse price action.
Trading CRT — Practical Application
- Mark the previous daily candle's high and low before each session. These are the BSL and SSL that the algorithm will likely sweep during the new day.
- In a bullish daily bias, expect an early sweep of the previous day's low (SSL sweep), then look for the bullish reversal entry after the sweep.
- In a bearish daily bias, expect an early sweep of the previous day's high (BSL sweep), then look for the bearish reversal entry.
- On your entry timeframe (5-minute or 15-minute), the same CRT logic applies at the candle level. The previous candle's high and low are local liquidity pools.
- Use CRT to refine your entry: if you are bullish and waiting for a pullback, wait for the sweep of the previous 15-minute low before entering — do not buy before the sweep.
CRT on the Higher Timeframes
CRT can be applied to weekly or monthly reference candles as a hypothesis, but weekly candles do not almost always follow one AMD path. Record every chosen interval and classify outcomes using rules set in advance, including weeks that do not fit.
The classic CRT pattern on the weekly chart: Monday and Tuesday produce the manipulation (wick) to one side. Wednesday marks the turning point. Thursday and Friday deliver the true weekly direction. ICT refers to this as the TGIF (Thank God It's Friday) pattern — the Friday close frequently aligns with the true weekly delivery direction, opposite to the Monday-Tuesday manipulation.
CRT is the bridge between macro ICT analysis and micro execution. Once you internalize that every candle is a miniature market with its own AMD cycle, you stop being surprised by wicks and false moves. They are not random noise — they are the algorithmic manipulation phase of that candle's delivery cycle, and they tell you exactly where the true move is going.
Worked hypothetical example: test a candle-range hypothesis
A learner marks the prior hourly candle's high and low before the next hour starts. Price first trades above the high, then returns inside the range and approaches the midpoint. The learner records the observed sequence under a sweep-and-return rule but does not assume every hourly candle must follow three phases or target the opposite boundary.
Failure case and what to test
Price may break one side and trend, cross both sides, or stay within the range. The outcome depends on candle timeframe, session, volatility and the definition of a valid return. Avoid selecting only days that fit an AMD-style narrative.
Test checklist
- Fix the candle timeframe and when each range becomes active.
- Define a sweep, close-back-inside, midpoint and range invalidation.
- Record single-side breaks, two-sided breaks and no-break cases.
- Compare any trade rule with a naive range-break baseline after costs.
RELATED READING
Continue with guides selected for overlapping concepts and practical context.