CRT Handbook · October 31, 2026 · 7 min read

Multi-Timeframe CRT: How to Align Your Timeframes

A CRT Handbook lesson on multi-timeframe confluence: which candle sets the range, where the turtle soup prints and which chart you enter on.

The question I get more than any other about Candle Range Theory isn't "what is CRT?" It's "which timeframe do I trade it on?" And the honest answer is: not one. CRT is a multi-timeframe model by design. A higher-timeframe candle sets the range, a middle timeframe proves the sweep failed, and a lower timeframe gives you the entry. Get those three out of alignment and the setup stops making sense, no matter how clean it looks on one chart.

That's what Chapter 6 of the CRT Handbook, Multi-Timeframe Confluence, is about: how higher-timeframe ranges define lower-timeframe opportunities. This post walks through the core of that lesson the way I apply it every session.

The three jobs in a CRT setup

Every CRT trade I take has three separate jobs, and each one belongs to a different timeframe:

1. The range. A higher-timeframe Candle 1 marks the high and low (CRH and CRL). This is the map. It tells you where the liquidity sits and where the target is.
2. The proof. Candle 2 runs one side of that range. Then a lower timeframe has to show a turtle soup: a candle that runs the previous candle's high or low and closes back across it. That's the evidence the run failed.
3. The trigger. Only after the turtle soup do I look for an entry on the entry timeframe: an order block at the purge, an inverse FVG, a balanced price range, or a CISD (a close back through the open of the candles that ran into the purge).

If you try to do all three jobs on one chart, something gets lost. On the H4 you can see the range but you can't see a clean entry. On the M1 you can see entries everywhere but you've lost the range that tells you which ones matter. The skill is letting each timeframe do its own job.

Real alignments I use

Talk is cheap, so here are the actual pairings the presets in my CRT Radar are built on. These come straight from how I trade the models:

Notice the pattern. The turtle soup timeframe sits well below the range candle, and the entry timeframe sits at or below the turtle soup. You never confirm on a timeframe higher than the one you're trying to confirm, and you never hunt entries on a chart that's too slow to show them.

Notice also that the fast entries — M5 and M3 — only show up in London and Asia. That's not an accident.

The lesson I learned the hard way: faster isn't better

When I first built these models, I used M5 entries almost everywhere. Smaller timeframe, tighter stop, bigger R. On paper it was beautiful.

Live, it was noise. Signals fired, flipped and invalidated before I could act on them. I was trading the lower timeframe's mood swings instead of the higher timeframe's intent. Moving Model 1, Model 2 and the Classic model to M15 entries made everything calmer — fewer alerts, cleaner setups, and decisions I could actually make with a clear head. I wrote about that change in how I built the CRT Radar.

The rule I took from it: the entry timeframe should be the slowest one that still gives you a precise invalidation. London and Asia move fast enough that M5 and M3 earn their place. A New York H6 setup usually doesn't need anything faster than M15.

Confluence means agreement, not decoration

A lot of traders hear "multi-timeframe confluence" and start stacking every timeframe on one chart until it looks like a Christmas tree. That isn't confluence. Confluence is when the timeframes agree on the same story.

Here's what agreement looks like in practice:

• Candle 2 purges into a past key level from the higher timeframe. For Models 1 and 2 that means an H4-or-higher level: an order block, rejection block, FVG, volume imbalance, or a previous day/week high or low. The range alone isn't enough; the sweep has to hit something the higher timeframe cares about.
• The turtle soup prints inside Candle 2, at that level. Not before, not somewhere else on the chart.
• The entry requirement prints after the turtle soup. An order block or IFVG that formed before the proof doesn't count. That one sequencing rule removes most of the low-quality setups I used to take.

My 1:188 MNQ trade at the Sunday Asia open is a good example of all three agreeing: the previous day's high and a daily FVG above, an M30 turtle soup right at that FVG, then M15 entry requirements printing on the way down. The target, an H4 FVG below, also came from the higher timeframe. It's one trade, and most of mine are much smaller or lose. But it shows the structure clearly.

The target belongs to the higher timeframe too

This is where most traders break their own alignment. They find a great M15 entry and then take profit at some random M15 swing because it "looks like resistance."

In CRT, the target is the opposite side of Candle 1 — the range you started with. The higher timeframe defined the map; the lower timeframe just got you in cheaply. If you manage the trade off the entry chart, you'll cut most of your good trades short and hold your bad ones too long. A CRT-timeframe candle closing through the target is what clears the setup, not a wiggle on the M5.

That's also why the risk-to-reward on these trades can look so extreme. The stop comes from the lowest timeframe (the turtle soup wick), and the target comes from the highest (Candle 1). The gap between those two timeframes is the R. I break down that geometry in how a 1:175 trade happens.

Time is a timeframe

One more layer people forget: the clock. The Classic model only works because the H4 candles open at specific times — C1 at 2 AM, C2 at 6 AM, C3 at 10 AM, or 1-5-9 on the other alignment. London uses the 2, 3 and 4 AM hourly candles. Asia uses the 3 PM candle.

If your chart timezone is wrong, your H4 candles are wrong, and your whole stack is built on the wrong range. Before you study anything else, check your chart's timezone and session settings. Chapter 5 of the Handbook covers killzone timing, and the Time Theory hub has more free reading on why the clock matters so much in CRT.

A simple routine to practise alignment

If you want to train this skill without any indicator at all, here's how I'd do it:

1. Pick one model and stick to it for a few weeks. For most people, the Classic 2-6-10 on H4 is the easiest to start with.
2. Mark Candle 1 on the H4 before the session. Write down its high, low and the key levels above and below it.
3. Drop to M15 and wait. Do nothing until Candle 2 runs one side into a key level and an M15 turtle soup closes back inside.
4. Only then look for the entry requirement. No turtle soup, no trade.
5. Go back to the H4 for the target. The opposite side of Candle 1. Write it down before you enter.
6. Journal every one, including the losers. You'll quickly see which misalignments cost you money.

It's slow. It's supposed to be. Rule the candle, rule the trade.

Go deeper

The full lesson — including how each chapter connects, from market structure and entry mechanics to stop and target placement — is in the CRT Handbook. The Handbook is the method; it doesn't include indicator access. If you'd rather have the chart enforce the alignment for you, the CRT Radar presets do exactly that, and the free CRT Radar Lite runs the Classic H4 → M15 model with no settings.

New to all of this? Start with what CRT is, then come back here. And if you want to see how the alignment plays out across real trades, wins and losses, every trade I take goes on the results page.