BackOctober 28, 202612 min readEmmaCentury

Acute:Chronic Workload Ratio Explained (And Why the 0.8–1.3 Rule Is Shaky)

The acute:chronic workload ratio is the most quoted number in training load management, and also one of the most criticised. Here is how to calculate it from your watch, what the evidence supports, and what to use instead of the magic bands.

Updated 2026-10-28Editorial standards
Acute:Chronic Workload Ratio Explained (And Why the 0.8–1.3 Rule Is Shaky)

You had a big week. Three runs, a long ride, two gym sessions, and a Saturday hike that turned into more than you planned. Your watch shows a training load number that jumped, a recovery score that fell, and a message telling you to take it easy.

Somewhere behind that message is a calculation. If you've gone looking for it, you've probably found the acute:chronic workload ratio, usually shortened to ACWR, usually presented with a graph and a green band between 0.8 and 1.3.

It's the most cited idea in training load management. It's also been taken apart fairly convincingly by sports scientists who study the maths behind it. Both of those things are true at once, and understanding the split is more useful than picking a side.

TL;DR: ACWR compares your recent training load (usually the last week) to your longer-term average (usually the last four weeks). The idea is that a spike relative to your base raises injury risk. The specific 0.8–1.3 "sweet spot" comes from a 2016 review and has weak statistical footing: the ratio is built from two correlated numbers, the bands are arbitrary, and there's no trial evidence that managing training by ACWR prevents injuries. What does hold up is simpler: big jumps in load relative to what you're used to are risky, and a high, stable base protects you. Use the ratio as a rough alarm for spikes, not as a prescription.

YouTube: Related video

What the ratio actually is

Two numbers, divided.

Acute load is what you did in the last seven days. Chronic load is your average weekly load over the last four weeks, which is the same as your 28-day total divided by four.

ACWR = acute load ÷ chronic load

If you've been training at a steady level, the ratio sits near 1.0. A big week pushes it above 1.0. A rest week pushes it below.

The problem starts immediately: what counts as "load"? There's no agreed unit. Different systems measure it differently, and the choice changes the answer.

Method How you calculate it What it captures Weakness
Duration Minutes of training per week Volume only Treats a walk and an interval session as equal
Active energy Calories from workouts Volume plus body size Underestimates easy work, rewards heat and long sessions
Session RPE Minutes × perceived effort (1–10) Volume and intensity, cheaply Subjective, drifts with mood
TRIMP / HR-based Heart rate reserve × duration weighting Internal physiological strain Distorted by heat, caffeine, medication

Your Apple Watch gives you options one and two directly, and you can build three yourself in about thirty seconds after each session. All three will produce different ACWR values from the same training. If someone tells you their ratio is 1.4, the first question is "of what?"

A worked example

Suppose your last four weeks looked like this, measured in session-RPE units:

Week Load
Week 1 (four weeks ago) 300
Week 2 320
Week 3 310
Week 4 (last week) 330

Chronic load = (300 + 320 + 310 + 330) ÷ 4 = 315.

Now you do a big week of 480. Your ACWR = 480 ÷ 315 = 1.52, which lands in the red zone on almost every chart you'll find. Your chronic load also nudges up to 360 for next week, which is exactly why the ratio tends to self-correct after a spike.

Coupled versus uncoupled

There's a detail most articles skip. In the calculation above, the acute week is also included in the chronic average, because last week is part of the last four weeks. This is called the coupled method, and it's what Gabbett originally used.

The alternative is uncoupled: compare this week to the average of the preceding four weeks, excluding this week. Same idea, cleaner separation, different numbers. The uncoupled version is less affected by the maths problems described below, and it's the one to use if you're going to use the ratio at all.

Where 0.8 to 1.3 comes from

The band comes from Tim Gabbett's 2016 paper in the British Journal of Sports Medicine, "The training-injury prevention paradox: should athletes be training smarter and harder?"

The paper's argument was counterintuitive at the time and still gets misquoted. It wasn't "train less". It was the opposite: athletes who sustain a high chronic workload are more robust, and the risk comes from sharp changes relative to that base, not from hard training itself. Gabbett proposed that practitioners keep the acute:chronic ratio roughly within 0.8–1.3 to minimise injury risk, with a "danger zone" at higher spikes.

That framing did two useful things. It moved the conversation from "how much is too much" to "how fast is too fast", and it introduced the idea that the same week of training can be safe or risky depending on what came before it. Both of those ideas survive the criticism.

The critique is not nitpicking

In 2020, Impellizzeri and colleagues published a paper in the International Journal of Sports Physiology and Performance titled, without much ambiguity, "Acute:Chronic Workload Ratio: Conceptual Issues and Fundamental Pitfalls." Their objections are technical, and they matter.

Mathematical coupling. Acute load is a component of chronic load. Dividing a variable by something that contains it creates a statistical dependency that produces spurious correlation. This is not a philosophical objection. It's a known artefact: correlated ratios behave in ways that make apparent associations appear and disappear depending on how the data are constructed.

Arbitrary categories. Turning a continuous ratio into "under-training / sweet spot / danger zone" bands requires cut-points, and the cut-points were chosen from the data rather than from physiology. Different studies have used different bands, which makes the findings hard to compare and easy to over-fit.

No causal evidence. Nearly everything here is observational. Teams that spike their load also tend to be in congested schedules, returning from injury, or changing training phase. The ratio may be a marker of those situations rather than a cause of injury. What's missing is the thing that would settle it: trials where training was modified by ACWR and injury rates were compared to a control group.

Sensitivity to method. Change the load metric, the chronic window, or coupled versus uncoupled, and you get different ratios and different conclusions. A metric that moves that much under analyst choices is a fragile basis for decisions.

A 2021 editorial in Frontiers in Physiology made a similar argument, asking directly whether there's scientific evidence for the ratio and concluding that the case is much weaker than its popularity suggests.

So should you ignore it entirely?

No, but you should downgrade what you think it's telling you.

Strip away the bands and the decimal places, and ACWR is a ratio of recent load to habitual load. That's a sensible thing to keep an eye on, for the same reason you notice when someone goes from running twice a week to running six times a week and then gets shin splints. The concept describes something real. The precision is fake.

What I'd keep:

  • Spikes matter. A week of training well above your recent norm is the situation most worth flagging, whatever number you use to flag it.
  • The base protects. A high, consistent chronic load is associated with fewer injuries, not more. Being generally fit and durable is the point, not staying under a threshold.
  • Consistency beats optimisation. The athletes who get hurt are usually the ones whose training is erratic, not the ones whose ratio was 1.35 for a week.

What I'd drop:

  • The idea that 1.31 is dangerous and 1.29 is safe.
  • Any claim that hitting a ratio causes or prevents injury.
  • Using a single decimal to decide whether to train on a given Tuesday.

How to calculate it from an Apple Watch

You don't need a platform. Three approaches, from crudest to best.

1. Workout minutes (crudest)

Open the Fitness app, look at your weekly workout minutes, and jot them down for four weeks. Divide last week by the four-week average.

This is fast and it's better than nothing, but it treats a gentle 60-minute walk as three times the load of a 20-minute hard interval session, which is wrong in a way that matters.

2. Active energy (better for volume)

Health app → Activity → Active Energy, weekly totals. Same calculation. This at least scales with body size and effort, but it still rewards long easy sessions disproportionately and it's inflated by heat.

3. Session RPE (best for training decisions)

After each session, write down duration in minutes and a perceived effort from 1 to 10. Multiply. That's your session load. Add up the week.

A 40-minute easy run at RPE 3 is 120. A 25-minute interval session at RPE 8 is 200. That ordering is much closer to how your body experienced the week than either duration or calories, and it takes ten seconds to record in a note. If you want a structured way to rate effort, the RPE guide covers anchoring the scale.

Whichever method you use, calculate uncoupled: this week against the average of the previous four.

A practical version without the decimal points

Here's what I'd actually do with the concept, and it doesn't require a spreadsheet.

What last week looks like Plain reading What to do
Well below your usual You backed off, deliberately or not Fine for a deload; don't let it become a habit
Roughly your usual Steady state Keep going
Moderately above your usual Productive overload Fine if sleep and HRV are holding
Way above your usual, after months of low volume The classic spike Make this week easier, or at least don't add more
Way above your usual, on a big base Hard block Watch recovery markers, not the ratio

Notice that the action depends on context, not on a threshold. That's the whole argument in one table.

Two rules of thumb that survive scrutiny, mostly because they're modest:

  • Increase weekly load gradually, not in jumps. The "10% rule" is also arbitrary and has been criticised for the same reasons, but the underlying instinct, that sudden large increases are riskier than gradual ones, is well supported.
  • Build for three weeks, back off for one. A planned lighter week every fourth week keeps the chronic load from collapsing and gives you a reset you didn't have to negotiate with yourself.

What to pair it with

The ratio is a load-side metric. It says nothing about how you're responding, which is where your watch is genuinely strong.

The signals worth checking alongside it:

  • HRV trend versus your own baseline. Not the population range, your four-week baseline. A sustained drop while load rises is the pattern that matters.
  • Resting heart rate. A rise of a few beats sustained across several days, alongside hard training, is a classic under-recovery signal. The HRV and resting heart rate guide has a decision tree for this.
  • Sleep duration and consistency. Load tolerance drops fast when sleep is short, and no ratio captures that.
  • How sessions actually feel. If easy runs feel hard for a week, that's data.

Century AI compresses the recovery side of this into a daily score: HRV, resting heart rate, and sleep, read against your own recent history rather than a population average. It won't compute a workload ratio for you, and honestly that's fine. The useful question is "am I absorbing this training", and that's a recovery question.

If you're specifically worried about the injury side, the running mileage guide goes through how to add volume without the classic mistakes.

Where the ratio came from, and why it spread

A bit of context helps explain the enthusiasm. ACWR arrived in team sports, where staff needed a number they could put in front of a coach who wanted to know whether a player was ready. It had to be simple, computable from GPS and session data, and easy to act on. It was all three.

Then it moved into consumer wearables, where the constraints are different. Your watch has no coach, no squad, no fixture list, and no one to have a conversation with about why your ratio is 1.4. It has a notification. That's a much worse context for a metric whose main weakness is that it needs human judgement to interpret.

So use it as a smoke alarm, not a thermostat. A jump in recent load relative to your base is worth noticing. What you do about it should depend on how you feel, how you're sleeping, what your HRV is doing, and what's coming next week.

The bottom line

ACWR is a good idea wrapped in bad maths. The idea, that load changes relative to your base are what cause trouble, is worth keeping in your head. The 0.8–1.3 band, the danger zone above 1.5, and the confident decimal places are not supported well enough to plan your training around.

Calculate it if you find it useful. Use session RPE, use the uncoupled version, and treat any number outside your normal range as a prompt to look at your recovery data rather than an instruction. The athletes who stay healthy aren't the ones with perfect ratios. They're the ones who train consistently, increase gradually, and back off before their body forces the issue.


Century AI helps you understand your body with a daily health score, recovery score, and sleep insights — using the watch you already wear.

Tagstraining-loadacute-chronic-workloadinjury-preventionrecoveryhrvrunningapple-watch

Wearable measurements are estimates, not diagnoses. Read Century’s methodology and medical disclaimer.