Training load ramp rate and injury risk: what does ACWR actually tell you?

This is a different question from functional overreaching or overtraining, which is about whether your recovery metrics — resting heart rate, HRV, sleep — are currently failing to keep up with training. ACWR is upstream of that: it's about how fast you changed your training load in the first place, independent of how your body happens to be responding this week.

What ACWR actually measures

ACWR compares two windows of your own training load:

• Acute load — typically your training load summed over the last 7 days.
• Chronic load — typically your training load summed over the last 28 days, then divided by 4 to express it as an average week.
• Ratio = acute ÷ chronic.

"Load" itself isn't fixed to one measurement — the original research (largely from cricket, Australian rules football and rugby) used session-RPE multiplied by duration, but WHOOP Strain, GPS distance, and total training minutes have all been used as substitutes in later studies. The ratio only means something if you calculate it the same way, with the same load measure, week over week.

The traditional risk zones

Read this table as a rough map of a real pattern — sudden, large jumps in load are riskier than gradual ones — rather than as calibrated thresholds you should treat like a blood test reference range. The next section is why.

How much should you actually trust this number?

This is the section most training-load content skips, and it matters more here than usual.

The ACWR concept (Tim Gabbett and colleagues, building on earlier training-load research) has genuine face validity: a body that ramped up training volume by 60% in a week is plausibly less prepared for that load than one that built up gradually. But a separate group of sports scientists — Franco Impellizzeri's team most prominently, in a series of papers starting around 2019-2020 — identified a specific statistical problem: the standard ACWR calculation includes the most recent week's data in both the acute number and the chronic number (since the chronic average is a rolling window that still contains this week), which creates a mathematical coupling that can produce part of the acute-chronic correlation as an artifact of the arithmetic rather than a real physiological relationship. Their critique also points to inconsistent results across meta-analyses — some studies find high ACWR predicts injury, others find no relationship, and a few find the opposite — which is not what you'd expect from a robust, universal threshold.

A 2025 systematic review and meta-analysis of 22 cohort studies still reported an association between ACWR spikes above 1.5 and higher injury rates and between ratios below 0.8 and undertraining risk, so the pattern hasn't been thrown out — but the coupling critique hasn't been resolved either, and the same review's authors caution that individual factors (training history, prior injury, sport, age) shift how much any given ratio actually means for a specific athlete. The honest summary: sudden, large changes in your own training load relative to your own recent history are a real risk factor for injury across the sports-science literature broadly. The specific number 0.8-1.3, treated as a precise cutoff, is weaker evidence than the phrase "sweet spot" implies.

Rolling average vs. EWMA — the calculation problem within the calculation problem

Even independent of the coupling debate, the original rolling-average method has a smaller, more mechanical flaw: because it treats every day in a 7-day or 28-day window equally, a single big session can cause the ratio to jump sharply the moment it drops out of the window a week or four weeks later — a training-load echo that has nothing to do with what you actually did that day.

If you're tracking this yourself with a spreadsheet, a simple rolling average is fine for a rough read — most people don't need EWMA's precision. If you're coaching a team or managing return-to-play after injury, where a false alarm or a missed spike has real cost, the EWMA method is the one more recent sports-science guidance recommends over the original.

A simple way to track your own ramp rate

1. Pick one load measure and stick with it. WHOOP Strain is a reasonable default if you have a WHOOP, since it already blends duration and intensity via heart rate; total training minutes or distance work fine if you don't. Don't switch measures mid-tracking — the ratio only means something week over week with a consistent input.
2. Log daily load for at least 4 weeks before trusting the ratio. The chronic average needs a real month of data; a ratio calculated from your first week of tracking is comparing this week to itself.
3. Calculate weekly: sum the last 7 days, divide by (sum of the last 28 days ÷ 4). Do this once a week, not daily obsessively — day-to-day noise in a ratio built from weekly sums isn't meaningful.
4. Flag ratios above 1.5 or below 0.8 as a prompt to check in with yourself, not a verdict. Ask whether the jump matches something you did on purpose (a race, a training camp) versus something that crept up without a plan. A deliberate, planned spike followed by a taper is a different risk profile than an accidental one.
5. Weight it more heavily around returns from injury, illness, or a long break. This is where the undertraining side of the ratio — and the original research generally — shows the clearest signal: resuming a normal training week too soon after time off is one of the more consistent injury-risk patterns across studies, coupling critique aside.
6. Don't let a "safe" ratio override actual pain. ACWR is a load-management heuristic, not a substitute for how your body feels — a niggling joint or tendon issue deserves attention and, if it persists, a physiotherapist or doctor's assessment, regardless of what your ratio says that week.

Where this fits with recovery signals

ACWR and overtraining signals are answering different questions and can disagree with each other. You can spike your ACWR for a planned hard week, have HRV and resting heart rate dip exactly as expected, and rebound fine within two weeks — that's the ramp-rate risk showing up as normal, functional fatigue, not overtraining. You can also sit inside the 0.8-1.3 "sweet spot" every week and still develop overtraining signs, because ACWR says nothing about sleep debt, life stress, nutrition, or a nagging old injury — all of which affect how much load your body can actually absorb regardless of how gradually you ramped it. Neither number replaces the other; a fast ramp rate is one common on-ramp to overreaching, not the only one.

This is also where a cross-source view helps more than either number alone. Vita pulls training load (WHOOP Strain, or workout duration and effort from Apple Health if you don't have a band) alongside resting heart rate, HRV and sleep into one trend, so a load spike and a recovery dip that happen together are easier to see than eyeballing separate charts. If you're not sure whether a hard week you just had was a normal planned overload or the start of something that needs a deload, asking Vita's AI coach to look at both your recent training volume and your recovery trend together is a faster read than tracking ACWR by hand.

The bottom line

ACWR is a useful rule of thumb — don't ramp your training volume up dramatically faster than your recent history, and don't assume a long break means you can jump straight back to your old normal week — wrapped in a specific numeric formula that sports scientists have spent several years actively disputing on statistical grounds. Use the 0.8-1.3 range as a rough guardrail for how quickly you're changing your own training load, not a precise threshold with the certainty of a lab reference value, and pair it with how your actual recovery metrics are responding rather than trusting either number in isolation.