CRT: Candle Range Theory
Also known as: CRT, candle range model, range delivery model, three-candle model
What is it?
Candle range theory reads a single higher-timeframe candle as a complete four-phase story - open, accumulate, manipulate, distribute - that the lower timeframe plays out inside it. The claim is that a large candle is not one event but a compressed sequence. The candle opens; price consolidates for a while, building positions; it then runs one side of that consolidation to take stops; and only then does it deliver in the opposite direction to the candle's close.
A 4-hour EUR/USD candle that opens at 1.0862, ranges 1.0850-1.0872 for an hour, spikes to 1.0838, and then closes at 1.0916 shows all four phases - and on the 5-minute chart each is plainly visible even though the 4-hour candle just looks like one long green bar. The practical move is to trade the third phase. Once the manipulation leg has run against the expected direction, the rest of the candle's range is the distribution you are trying to capture.
That requires deciding in advance which way you think the candle closes, which is the hard part - and being wrong there means entering right as the real move goes the other way.
Why it matters: Candle range theory reframes one large candle as four phases, so a spike against your direction can be read as the setup rather than as the trade failing.
It shapes both entry timing and how a move against you is interpreted, which directly affects whether a stop is respected or rationalised away.
Real-world example
A 4-hour EUR/USD candle opened at 1.0862, ranged 1.0850-1.0872, spiked to 1.0838, then closed at 1.0916 - 78 pips of distribution after a 24-pip manipulation leg.
How SignalBots handles it
SignalBots delivers each signal with the level it is built on attached, so you can see whether an entry sits before or after the sweep rather than inferring it from the candle. See /risk-warning.
Pro tip
Drop to a timeframe roughly a twelfth of the candle you are reading - a 4-hour candle needs the 20-minute or 15-minute chart to show its phases clearly.
Common pitfalls
Using the model to justify holding through a stop. A manipulation leg and a losing trade look identical until the candle closes, so the stop still has to be honoured.
Frequently asked questions
How is CRT different from the power of three?
They describe the same accumulation-manipulation-distribution sequence. Power of three is usually applied to a daily or weekly candle; CRT applies the same reading to any candle on any timeframe and puts more weight on the range boundaries.
Which candle should I apply it to?
One large enough that its range matters to your trade - typically the 4-hour, daily or weekly. Applying it to a 1-minute candle leaves no room for four phases to be distinguishable.
How do I know which way the candle will close?
You do not know; you form a bias from higher-timeframe structure and trade it. That bias is the model's weakest link, and the four-phase reading does nothing to validate it, so capital is at risk on the assumption itself.
What if the manipulation leg never comes?
Then there is no CRT entry for that candle. Some candles simply open and run, and forcing an entry because the model predicts a sweep means buying into a move already underway.
Can this be automated?
The phase detection can be coded from range and sweep conditions, but only after the fact - live, a sweep is indistinguishable from a genuine break until price returns. Any automation has to accept that ambiguity explicitly.
Trading involves substantial risk of loss. Historical and backtested results do not guarantee future performance. Read the full risk warning.
