What Is Candle Range Theory (CRT)?

Candle range theory (CRT) is a price-action model built on one observation: every candle is a range, and price trades around those ranges in a repeatable sequence — sweep one side, reclaim it, run to the other side. That's the whole theory. The rest is definitions and rules precise enough to test.

Most traders meet CRT as a screenshot on social media. A higher-timeframe candle, a wick poking below its low, a rally to its high. The screenshot is easy. The mechanics — which candle counts as the range, what counts as a sweep, when the setup is confirmed, when it's dead — are where most explanations go vague. This guide doesn't.

Every candle is a range

Pick any candle on any timeframe. It has a high and a low. Between them sits everything the market agreed on during that period. Above the high and below the low sit stops, breakout orders, and late entries — resting liquidity.

CRT treats that high-to-low span as a tradable unit of structure. Not support and resistance zones drawn by hand. Not a pattern you squint at. A closed candle's exact high and exact low.

The second piece is nesting. A 4-hour candle contains sixteen 15-minute candles. The H4 candle defines the range; the M15 candles show you how price moves around inside it and through its edges. CRT always works across two timeframes this way — a higher timeframe for the range, a lower timeframe for the entry.

The range candle is called the parent candle. A working definition: the parent is a closed higher-timeframe candle whose high and low haven't yet been touched by any later closed candle. Touches count at the wick, and an exact tag counts — if a later wick hits the parent's high to the pip, that side is swept. A candle that's still forming is never the parent; ranges come from closed candles only.

The three phases: range, sweep, reclaim

A CRT setup plays out in three phases.

Phase 1 — the range exists. A higher-timeframe candle closes. Its high and low are now fixed reference levels.

Phase 2 — one side gets swept. A later candle's wick reaches or exceeds one side of the parent's range. This is a liquidity sweep: price trades through the level where stops cluster, fills resting orders, and — in the cases that matter — fails to hold there.

Phase 3 — price closes back inside. The sweep candle's body closes back within the parent's range. The breakout failed. The level held. Now the opposite side of the range becomes the draw: the side that just rejected price is spent, and the untested side is where the remaining liquidity sits.

Here is the bullish version — a low sweep and reclaim:

CRH ──────────────────────────────────────── ← target
      ┌──┐
      │  │  parent              ┌──┐
      │  │             ┌──┐    ┌┤  │
      │  │             │  │    ││  │
      └──┘             └─┬┘    └┴──┘
                         │    close back inside
CRL ──────────────────── │ ───────────────── ← swept
                         │
                         ▼  wick takes out the low,
                            body reclaims the range

Sweep the low, close back inside, target the high. The bearish version is the mirror image: sweep the high, close back inside, target the low.

Two details that separate a testable rule from a vibe. First, the sweep is evaluated at the wick — the body doesn't need to cross the level, and equal prices count as swept. Second, the close is what decides everything. A wick beyond the range means nothing on its own. The close tells you whether the move was a sweep or a breakout.

Confirmed CRT vs expansion

The sweep candle's close splits every sweep into two outcomes.

Close back inside the range: confirmed CRT. The setup is live. Direction is set by which side was swept — high swept means bearish, low swept means bullish. The target is the opposite side of the parent's range.

Close outside the range: expansion. No confirmed CRT forms. Price swept the level and kept going — that's continuation, not reversal. The old range is finished, and the candle that broke it becomes the new parent. The market rotated to a new range; you wait for the next sweep.

This binary is the discipline of CRT. The same wick through the same level is either a setup or a non-event, and the close decides which. There is no "almost confirmed."

Potential vs confirmed CRT

There's a timing wrinkle worth understanding, because it's where intrabar trading happens.

The sweep occurs the moment a wick reaches the parent's level — and from that exact moment the setup is a potential CRT, even though the higher-timeframe candle that did the sweeping may not close for hours. The sweep has happened; the deciding close hasn't printed yet.

At the higher-timeframe close, one of two things happens. Close inside the range, and the potential CRT becomes a confirmed CRT — same setup, upgraded status. Close outside, and the potential CRT failed to confirm: the structure is dead and the range rotates.

Traders handle this window differently. Entering during the potential phase gets better price and accepts confirmation risk. Waiting for the confirmed close trades later but with the failure case filtered out. Neither is free. What matters is knowing which one you're doing — and an entry taken during the potential phase doesn't retroactively become wrong if the candle later fails to confirm; it was a defined trade with a defined stop the moment it triggered.

What ends a CRT

A confirmed CRT doesn't live forever. Three conditions end it, and the first one to fire wins.

1. Target reached. Price touches the opposite side of the parent's range. The draw on liquidity got hit. This is the setup completing.

2. The 50% retrace. Every CRT range has a midpoint — halfway between the parent's high and low. Once the reclaim leg trades back to that midpoint — a wick touching it is enough — the setup is retired for new entries. The logic: after a low sweep you want to be entering near the lows, in the cheap half of the range. If price has already recovered to the middle before your entry triggered, the location edge is gone. The structure may still resolve toward the target, but it's no longer offering an entry.

3. A close outside the range. If any later higher-timeframe candle closes beyond the parent's high (for a bearish CRT) or below its low (for a bullish CRT), the setup is invalidated. The range broke. Note the symmetry with confirmation: wicks beyond the range don't kill a CRT, closes do.

One more rule that matters in practice: a CRT ending does not close a trade that's already open. The end conditions stop new entries on that structure. A position that triggered earlier runs to its own target or stop on its own terms. Setup lifecycle and trade lifecycle are separate things.

When a CRT ends — any reason — the candle that ended it becomes the next parent, and the process restarts. Ranges rotate; the model never runs out of structure.

Worked example: H4 parent, M15 entry

Say gold prints an H4 candle with a high of 2350.00 and a low of 2340.00, and that candle closes. Parent range: 2340.00–2350.00. Midpoint: 2345.00.

Three H4 candles later, price drops. On the M15 chart you watch a candle wick down to 2339.20 — below the parent's low. Swept. The next few M15 candles close back above 2340.00. The reclaim is happening, and the last lower-timeframe candles of that move leave behind an order block — the footprint of the rejection, and the zone a CRT trader wants price to retest for entry.

Now the branches:

• Price retraces into the order block at 2339.80, then the H4 closes at 2343.80 — inside the range. Confirmed bullish CRT. The entry triggered on the retest, the stop sits at the sweep low, and the target is the parent's high at 2350.00.
• Price rallies straight to 2345.00 — the midpoint — before ever retesting the entry zone. Setup retired. No entry. The move may still reach 2350.00, but chasing the middle of the range was never the trade.
• The H4 closes at 2338.40 — below the range. No CRT. That was distribution, not a sweep. The breaking candle becomes the new parent and you start over.

Same sweep in all three branches. Three different outcomes, each decided by a rule, not a feeling. That's the point of CRT as a theory rather than a chart pattern: every branch is defined in advance, which means every branch can be backtested.

Where CRT sits next to ICT

If you've studied ICT (Inner Circle Trader) concepts, CRT will feel familiar — liquidity sweeps, order blocks, and draw on liquidity are shared vocabulary. The difference is scope. ICT is a sprawling framework with dozens of interlocking concepts; CRT compresses the core sequence into a single mechanical model built on one unit of structure, the candle range. The full comparison is in CRT vs ICT.

CRT's popularization is credited to the trader known as RomeoTPT, who built the modern formulation around the parent-candle model. Background on that history is in who created CRT.

Trading it without guessing

Everything above is mechanical: wick-level sweeps, close-based confirmation, three defined end conditions. That precision is what makes CRT testable — and testing is the part most traders skip.

CRT Terminal runs this exact model as a live detection engine on real market data: parent selection, sweeps, potential and confirmed CRTs, order blocks, and signals, with frame-by-frame replay and backtesting so you can verify the behavior on history instead of trusting screenshots. The entry logic itself is covered in the CRT trading strategy guide, and pricing is flat and public.

Start the 7-day trial — no card required — and watch the engine mark CRTs on live data.

FAQ

What is candle range theory in simple terms?

Candle range theory treats each higher-timeframe candle's high-to-low span as a tradable range. When price sweeps one side of that range at the wick and closes back inside, the model expects price to move toward the opposite side.

What is a parent candle in CRT?

The parent candle is the closed higher-timeframe candle whose range defines the current setup — its high and low haven't been touched by any later closed candle. Sweeps against it are measured at the wick, and an exact touch counts as swept.

What is the difference between a potential and a confirmed CRT?

A potential CRT exists from the moment a wick sweeps the parent's level, while the sweeping candle is still forming. It becomes confirmed only if that candle closes back inside the parent's range; a close outside ends the setup without ever confirming it.

What invalidates a CRT setup?

Three things end a CRT: price reaches the target (the opposite side of the range), price retraces to the 50% midpoint of the range (a touch counts), or a later higher-timeframe candle closes outside the range. An already-open trade is not closed by the setup ending.

Does CRT work on forex, gold, and indices?

The mechanics are instrument-agnostic — any market with candles has ranges, sweeps, and closes. Whether it carries an edge on a specific instrument and timeframe pair is an empirical question, which is why backtesting it beats believing it.

Educational content, not financial advice.