If you've opened the sleep section of your Oura, Apple Watch, Garmin, or WHOOP app, you've seen a sleep cycle graph. It usually looks like a jagged staircase: your brain moving down into deep sleep, rising back up through lighter stages, sometimes reaching near-waking, then descending again.

Most people look at this graph and wonder what they're supposed to take from it. Is it normal to have this many stages? Why does deep sleep happen early and REM later? What does a bad night actually look like versus a normal one?

This guide explains what the sleep cycle graph (technically called a hypnogram) actually shows, what normal looks like across a full night, and how to use this data to actually improve your sleep and recovery.

Key Takeaways

• A normal sleep cycle is about 90 minutes long and contains light sleep, deep sleep, and REM in varying proportions across the night.

• Deep sleep is concentrated in the first half of the night. REM sleep builds across the night and is most abundant in the final cycles before waking.

• The graph is most useful for identifying trends and anomalies over time rather than judging a single night in isolation.

What Is a Sleep Cycle Graph?

A sleep cycle graph, or hypnogram, is a visual representation of which sleep stage your brain is in at any given point during the night. The vertical axis shows sleep stages from light at the top to deep at the bottom. The horizontal axis shows time. The resulting line moves up and down as your brain transitions between stages across the night.

Clinically, hypnograms are produced by polysomnography, a full sleep study measuring brain waves, eye movements, muscle activity, and more. The graphs your wearable produces are approximations based on movement, heart rate variability, and other physiological signals. They're not as precise as clinical measurements, but they're close enough to be useful for tracking patterns over time.

Understanding the stages of sleep is the foundation for reading the graph.

The Four Sleep Stages

Stage 1: Light Sleep (N1)

N1 is the transition between wakefulness and sleep. It lasts only a few minutes. Muscle activity decreases, the body starts to relax, and you may experience hypnic jerks (the sudden twitch-and-jerk sensation as you fall asleep). This stage is easy to wake from. Most wearables don't display N1 separately, lumping it into general "light sleep."

Stage 2: Light Sleep (N2)

N2 is where you spend the most time during a full night: roughly 45 to 55% of total sleep time. Heart rate slows. Body temperature drops. Brain activity shows distinct bursts called sleep spindles, which are linked to memory consolidation and learning. You're less easily woken in N2 than in N1, but still not deeply asleep. Most of the light sleep on your wearable graph represents N2.

Stage 3: Deep Sleep (N3 / Slow-Wave Sleep)

N3 is the deepest and most physically restorative stage. Brain waves slow to large, synchronized waves called delta waves. Growth hormone is released, immune system activity increases, and tissue repair happens here. It's very difficult to wake someone from deep sleep, and people woken from N3 often feel groggy and disoriented for several minutes.

Normal adults spend about 13 to 23% of their sleep in N3. Deep sleep is concentrated in the first two to three cycles of the night, which is why the early part of your hypnogram typically shows the deepest dips. Increasing deep sleep is often the most impactful sleep quality lever for people who feel physically unrefreshed despite adequate hours.

REM Sleep

REM (Rapid Eye Movement) sleep is when most dreaming occurs. Brain activity in REM resembles wakefulness, but the body's muscles are temporarily paralyzed (which prevents you from acting out dreams). REM is critical for emotional regulation, memory consolidation, and creative processing.

Adults typically spend 20 to 25% of their sleep in REM. Unlike deep sleep, REM is concentrated in the later cycles of the night. The REM periods get progressively longer as the night goes on, which is why the last hour before waking is often largely REM. This is also why cutting sleep short (early alarm, disruptions in the early morning) disproportionately reduces REM. Improving your REM sleep often starts with protecting the later hours of your sleep window.

What a Normal Night Looks Like on the Graph

A normal hypnogram across 7 to 8 hours of sleep typically shows:

• 4 to 5 cycles, each roughly 90 minutes long

• Deep sleep early: The first two cycles contain the most N3. The line dips deepest in the first third of the night

• REM late: The last two to three cycles contain progressively longer REM periods. The final cycle before waking is often almost entirely REM

• Brief awakenings: It's normal to surface briefly between cycles. You may not remember these. Wearables sometimes log them as short "awake" periods

• Overall pattern: Deepest dips in the left third of the graph, longest flat or near-awake stretches in the right third, with light sleep filling the middle of each cycle

The graph doesn't need to be perfectly smooth or symmetrical to represent healthy sleep. Variation is normal. What looks unusual on a single night may be perfectly normal for your body and your sleep that particular night.

What Looks Abnormal on the Graph

Patterns worth noting across multiple nights:

• Very little or no deep sleep: Consistently less than 10% of the night in N3. May indicate age-related changes, alcohol use, sleep apnea, or other issues affecting slow-wave sleep

• Very little REM: Consistently below 15% of the night. Common causes include alcohol (which suppresses REM in the second half), antidepressants, and sleep disorders

• Many short awakenings: Frequent brief awakenings throughout the night indicate fragmented sleep, which disrupts both deep and REM stages even when total hours look adequate

• Sleep latency issues: If the graph shows you lying awake for 30+ minutes before the first cycle begins on most nights, sleep latency may be the primary issue to address

• No clear cycling pattern: A graph that shows irregular, shallow stages without the characteristic deep-then-REM progression may indicate a sleep disorder worth evaluating

How to Read Your Wearable's Sleep Graph

Different devices display hypnograms differently, but the underlying logic is the same. A few things to keep in mind when reading yours:

Compare nights, not just individual ones. A single unusual night tells you very little. Look at the pattern over a week or a month. Is the deep sleep consistently low? Does the REM always get cut short? Are there many awakenings consistently on certain nights (weekends, stressful weeks, alcohol nights)?

Look for the cycle structure. Can you see the 90-minute cycles? If your graph is a fairly random distribution of stages with no clear cycling, that's worth noting. Clear cycling is the signature of healthy sleep architecture.

Note the time distribution. In a healthy night, the first few hours should show the deepest dips. If your graph is flat throughout or shows deep sleep evenly distributed (or mostly in the second half), that's atypical.

Don't over-index on single metrics. Sleep tracking apps and wearables vary in accuracy. They're better at detecting trends and relative patterns than at producing exact percentages. The graph is a tool for spotting patterns, not a clinical diagnosis.

Also see how deep sleep and REM compare in terms of what each one does for you, since they serve different recovery functions and both matter.

Factors That Change Your Sleep Graph

Understanding what shifts the pattern helps you interpret what you're seeing:

• Alcohol: Suppresses REM in the second half of the night. The graph often shows early sleep stages looking normal, then a disrupted second half with little REM and more awakenings

• Age: Deep sleep naturally decreases with age. By the 60s and 70s, N3 may be significantly reduced or absent even in healthy sleepers

• Exercise: Physical activity, especially earlier in the day, tends to increase the depth and duration of slow-wave sleep

• Stress: High cortisol reduces deep sleep and can increase early-morning awakenings, cutting REM short

• Sleep timing: Going to bed significantly later than usual shifts the cycle distribution. You may get more REM early (since the circadian REM window starts at the same time regardless) and less deep sleep if you sleep shorter than usual

Best Tool for Acting on Your Sleep Data

Reading your sleep graph is one thing. Connecting it to how you plan your day is another. Lifestack integrates with Oura, Apple Watch, Garmin, WHOOP, and other wearables and uses your sleep quality and recovery scores to help you schedule your day in line with your actual readiness.

If your graph shows a fragmented night with low deep sleep, Lifestack can reflect that in your day plan: protecting your best focus window if it still exists, flagging that heavy cognitive work may be harder than usual, and helping you see whether light recovery work or a recovery window makes sense. Energy management starts with knowing where your energy actually is, and your sleep graph is the most direct signal for that. On nights where sleep gives you strong energy recovery, Lifestack helps you put that capacity to work on what matters most.

Lifestack is $7/month or $50/year (7-day free trial on the annual plan).

Frequently Asked Questions

What does a normal sleep cycle graph look like?

A normal sleep cycle graph shows 4 to 5 cycles across a 7-8 hour night, each roughly 90 minutes. The line dips deepest in the first two to three cycles (deep sleep), then becomes shallower with longer REM periods in the final cycles. Brief near-awake periods between cycles are normal. The overall pattern looks like a decreasing staircase in the first half and a shallower, REM-heavy pattern in the second half.

How many sleep cycles per night is normal?

Four to five complete cycles across a 7-8 hour night is typical for adults. Each cycle is roughly 90 minutes. With 6 hours of sleep you get about four cycles; with 9 hours, up to six. The later cycles are lighter and REM-dominant, so shorter sleep tends to cut REM more than deep sleep.

How much deep sleep and REM is normal?

Normal ranges for adults: deep sleep (N3) makes up about 13-23% of total sleep time, REM about 20-25%. In practical terms, for an 8-hour night, that's roughly 60-110 minutes of deep sleep and 90-120 minutes of REM. Both decline somewhat with age. Consistently below 10% deep sleep or below 15% REM on most nights is worth investigating.

Why does deep sleep happen early in the night?

Deep sleep is regulated primarily by sleep homeostasis, the biological drive to recover from wakefulness. This pressure is strongest at the start of sleep and dissipates as the night progresses. REM is regulated more by the circadian clock, which schedules REM peaks in the early morning hours, regardless of when you went to sleep. The result is a predictable pattern: deep sleep early, REM late.

Is it normal to wake up between sleep cycles?

Yes. Brief awakenings between cycles are a normal part of sleep architecture. Most people wake briefly 4-6 times a night and have no memory of it. Wearables sometimes log these as short awake periods on the graph. The distinction between normal between-cycle awakenings and problematic fragmentation is duration and frequency: brief, infrequent awakenings are normal; many awakenings per hour that each last several minutes indicate a problem.