The acute to chronic workload ratio (ACWR) is your training load over the last 7 days divided by your average weekly load over the last 28 days. A value of 1.0 means this week matched your recent norm, 1.5 means it ran 50 percent above it, and 0.7 means you trained well below what you are used to.
Coaches adopted the ratio in the mid-2010s as an injury-risk flag, and a band of 0.8 to 1.3 became known as the training "sweet spot." Since 2017 a run of statistical critiques, and one randomized trial that found no benefit, have taken most of that promise apart. What survives is a plainer idea. A sudden jump in training relative to what you have been doing deserves attention, and the ratio is one crude way to see that jump. The performance physiology guide shows where load fits among oxygen delivery, thresholds and heart rate zones.
01How the ratio is calculated
The idea started in cricket. Hulin and colleagues followed 28 elite fast bowlers across 43 individual seasons and compared each bowler's one-week workload with his 4-week rolling average, a comparison they called training-stress balance. Bowlers whose acute internal workload exceeded 200 percent of their chronic workload had 4.5 times the injury risk of bowlers in the 50 to 99 percent band. Two years later Hulin's group ran the same comparison on 53 elite rugby league players, used the name "acute:chronic workload ratio," and reported the highest injury risk at ratios of 2.11 and above.
The conventional form uses any consistent measure of load, such as GPS distance, balls bowled, session RPE (perceived effort multiplied by minutes) or a heart rate score like TRIMP.
ACWR = (load over last 7 days) ÷ (total load over last 28 days ÷ 4)
This version is called coupled because the most recent week appears in both the numerator and the denominator. An uncoupled version divides the last 7 days by the average of the weeks before them, so the two windows never overlap. Both forms appear in the literature, and they give different numbers for the same training.
Rolling averages and exponentially weighted averages
A rolling average treats a session from six days ago exactly like one from this morning, then drops it on day eight. Williams and colleagues proposed exponentially weighted moving averages (EWMA) in 2017 to fix this. Each day's value moves part of the way from yesterday's average toward today's load.
EWMA today = load today × λ + (1 − λ) × EWMA yesterday
They set λ = 2 / (N + 1), with N at 7 days for acute load and 28 days for chronic load. That gives an acute λ of 0.25 and a chronic λ of about 0.069. Applied to three fictitious athletes who all scored 1.43 with rolling averages, the EWMA method returned 1.25, 1.41 and 1.55, because it weighted where in the month each athlete's hardest sessions fell.
Murray and colleagues tested both methods in 59 elite Australian football players over two seasons. The two methods produced significantly different ratios in the 1.0 to 1.49, 1.50 to 1.99 and above 2.0 ranges. Both linked ratios above 2.0 to higher injury risk, and the EWMA version explained significantly more of the variance in injury.
EWMA has its own flaw. Wang and colleagues showed that the first value in the series carries outsized weight. With the Williams decay constants over 28 days, the starting load contributed 1.9 times as much to chronic load as the load on the day of calculation. In their worked example, EWMA ratios started from different initial values took around 50 days to converge.
One week, four ratios
The method changes the answer enough to move a week across every published threshold. Take an athlete who has trained at a steady 150 load points a day for months, then doubles to 300 a day for one week.
| Method | Acute | Chronic | Ratio |
|---|---|---|---|
| EWMA, Williams constants (N = 7 and 28) | 280 | 209 | 1.34 |
| EWMA, Titan's time constants (7 and 42 days) | 245 | 173 | 1.41 |
| Rolling average, coupled (7 days over 28 days) | 300 | 187.5 | 1.60 |
| Rolling average, uncoupled (7 days over prior 3 weeks) | 300 | 150 | 2.00 |
Depending on the method, the same training sits just above the sweet spot at 1.34, in Titan's High zone at 1.41, in Gabbett's danger zone at 1.60, or close to the ratios of 2.11 and above where Hulin's rugby league players got hurt most often. Acute and chronic values here are daily averages, and the calculations are ours. A threshold published for one method never transfers to another.
02Where the 0.8 to 1.3 sweet spot came from
Tim Gabbett's 2016 review in the British Journal of Sports Medicine made the ratio famous. His training-injury prevention paradox held that athletes accustomed to high loads get injured less than athletes on low loads, and that "excessive and rapid increases in training loads" explain a large share of non-contact soft tissue injuries. Pooling data from cricket, Australian football and rugby league, he wrote that ratios "within the range of 0.8–1.3 could be considered the training 'sweet spot'," while ratios of 1.5 or more "represent the 'danger zone'." The same review advised limiting weekly load increases to under 10 percent. In one cited dataset, week-to-week increases of 15 percent or more carried injury risks of 21 to 49 percent.
The International Olympic Committee's 2016 consensus statement on load and injury risk, on which Gabbett was an author, included this model, and Wang and colleagues note that it became a guideline for training practice.
03The critiques
Mathematical coupling
Lolli and colleagues showed in 2019 that the coupled ratio builds a spurious correlation between acute and chronic load. Because the acute week sits inside the chronic average, the two numbers move together whether or not any physiological link exists between them. Wang and colleagues added a second consequence. If you did nothing for three weeks and then trained, the coupled ratio equals exactly 4 no matter how much you did, since the denominator is simply this week's load divided by four. A 10 kilometer week and a 100 kilometer week after a layoff get the same score.
What a ratio assumes
A ratio controls for its denominator only if the numerator scales in proportion to it across the whole range. Lolli's group made this point in a second 2019 paper and questioned whether dividing acute load by chronic load normalizes anything. Impellizzeri and colleagues concluded in 2020 that the ACWR fails to normalize the numerator by the denominator even in its uncoupled form. They called it an inaccurate and ambiguous metric, "not consistently and unidirectionally related to injury risk," and found no evidence supporting its use in training recommendations aimed at reducing injury.
Causation
Recommending that athletes keep their ratio inside a band assumes that changing the ratio changes injury risk. Impellizzeri's 2020 paper pointed out that no study had even tried to estimate that causal effect. The second part of their 2020 commentary in the Journal of Athletic Training listed ten methodological problems across the training load and injury literature, including arbitrary time windows, discretized ratios, inconsistent injury definitions and small samples. Their advice was to return to overload and progression and to adjust training by how each athlete responds.
Random denominators work almost as well
The 2021 paper by Impellizzeri, Woodcock and colleagues ran the sharpest test. They took previously published data and replaced each player's real chronic load with a fixed number or a random one. The real ratio gave an odds ratio for injury of 2.45. Dividing acute load by a fixed 1510 gave 1.95, and random chronic loads gave a mean of 1.89. Neither the ratio nor acute load alone predicted injury meaningfully better than a model with no predictors, with c-statistics of 0.574 and 0.544 against 0.5. The authors concluded that ACWR mostly rescales acute load, and recommended dismissing the ratio and the theory behind it.
Thin data behind the thresholds
Wang, Vargas, Stokes, Steele and Shrier reviewed the ratio from an epidemiology standpoint in 2020. The IOC model rested on three studies with 28 athletes, 53 athletes and an unreported number. Ratios at the ends of the range were binned as 0.5 or 2.0, and the rise in risk came almost entirely from points at 2.0. Limiting the data to ratios below 2.0 removed any apparent relationship. The higher risk at low ratios may reflect sparse-data bias, since no biological theory explains why training less than usual would injure you. They also noted that a taper before a marathon or triathlon lowers chronic load and pushes the ratio up on race day, when the rest should have lowered injury risk.
04Where the evidence stands now
Observational reviews still find an association between large spikes and injury. Griffin and colleagues reviewed 22 team-sport studies in 2020, supported the link with non-contact injury and preferred EWMA as the more sensitive method. Andrade and colleagues reviewed 20 studies covering 2,375 injuries in 1,234 athletes the same year. All the athletes were male professional team-sport players. Of those studies, 95 percent used the coupled ratio, and the authors counted 14 different binning schemes, which they said limits the strength of any recommendation. Qin and colleagues pooled 22 cohort studies in 2025. Injury incidence was lowest in the 0.8 to 1.3 band, but the 95 percent confidence interval ran from 14 to 94 percent, and 17 of the 22 studies were in soccer.
The only randomized trial found no benefit. Dalen-Lorentsen and colleagues assigned 34 elite youth football teams, 482 players of both sexes, to plan a full 10-month season around published ACWR principles or to train as usual. Health problem prevalence did not differ, with a relative risk of 1.01 (95 percent CI 0.91 to 1.12).
The defensible reading today has three parts. Large, abrupt increases in training relative to your recent history are associated with injury in team sports. No threshold, including 0.8 to 1.3, has been shown to be safe or to transfer across sports and calculation methods. Managing training to keep the ratio in a band has not been shown to prevent injury in the one trial that tested it.
05Reading a ratio from your own data
Treat the ratio as a description of your recent training. A value near 1.0 says this week resembles your last month. A value well above 1.0 says you ramped up, which is what a planned build block looks like and also what the week back from illness looks like when you pick up your old mileage. A value below 0.8 says you backed off, which is correct during a deload week or a taper and a warning sign only if you did not intend it. None of these values forecasts an injury.
The ratio also inherits every error in the load number underneath it. On a wrist wearable that number is usually built from heart rate. In a lab study of seven wrist devices, the Apple Watch had the lowest overall heart rate error, and six devices had median errors below 5 percent during cycling. Error was higher during walking, in men, at higher body mass index and with darker skin tone. Strength sessions, where heart rate says little about mechanical load on tendons and joints, are the weak spot for any heart rate based ratio.
How Apple Watch and Garmin report it
Apple's training load "compares the intensity and duration of your workouts over the last 7 days to what you've done over the previous 28 days" and classifies the result "on a scale from well below to well above." You can edit the effort rating on each workout. Apple publishes no formula and no numeric ratio.
Garmin defines acute load as a weighted sum of excess post-exercise oxygen consumption over recent days and shows a load ratio with explicit bands. Below 0.8 is Low, 0.8 to 1.4 is Optimal, 1.5 to 1.9 is High and 2.0 or higher is Very High. The ratio appears after 2 weeks of training. Garmin's Optimal band is wider than Gabbett's sweet spot, which shows how loosely the published thresholds carry into consumer devices.
06How Titan's short-term and long-term load relate
Titan's load ratio belongs to the same family, computed from its own training load model with different windows and weights. The app never calls it an ACWR, and the differences matter when you compare it with published cutoffs.
Daily load comes from time in heart rate zones during workouts, with each minute weighted by its zone number, so a Zone 4 minute counts 4. Workouts without heart rate get a per-minute value by activity type. How Titan measures training load compares this weighting with Banister's TRIMP curve.
Short-term load is an exponentially weighted average of daily load with a 7-day time constant. Long-term load uses a 42-day time constant. Titan's app code sets the daily weight as α = 1 − e^(−1/τ), which moves short-term load about 13.3 percent of the way toward each day's load and long-term load about 2.4 percent. That is gentler than the Williams convention, where the 7-day weight is 0.25. The load ratio is short-term load divided by long-term load, shown to two decimals and left blank on any day long-term load is 0. Both averages include today's load, so Titan's ratio is coupled in the same sense as the EWMA versions in the research. The 42-day long-term window is half again as long as the 28 days used in the team-sport studies, so long-term load reacts more slowly than a classic chronic load.
Titan sorts the ratio into four zones.
| Zone | Load ratio |
|---|---|
| Low | Below 0.8 |
| Optimal | 0.8 up to 1.4 |
| High | 1.4 up to 1.6 |
| Risk | 1.6 and above |
The lower bound matches Gabbett's sweet spot. The Optimal band ends at 1.4, the same as Garmin's and above Gabbett's 1.3. The Risk zone starts at 1.6, between Gabbett's danger zone at 1.5 and Garmin's Very High at 2.0. No study has validated any of these cutoffs for Titan's arithmetic, or for anyone else's. Use the zone as a prompt to look at your recent training.
Why the ratio reads high when you are new
Titan counts days without data as rest days with a load of 0 and starts the averages 84 days before the first day on the chart. If Apple Health holds no workouts from before you started training, long-term load climbs slowly from zero. Train at a steady 150 points a day from a standing start and the ratio reads 2.41 after 3 weeks, 1.58 after 6 weeks and 1.16 after 12 weeks, although your training never changed. These figures come from running Titan's formula on that steady schedule. Early in your history, a High or Risk zone mostly reflects a short record.
Where to find it in Titan
The Trends tile on Today shows your latest Training Load Ratio with an arrow when it moved by more than 0.01, next to a sparkline of the last 7 days of daily load. Tap it to open Trends, where the Training load chart plots long-term and short-term load together. Below the chart, Titan prints your latest ratio with its zone, such as "Ratio 1.12 · Optimal." Dragging across the chart shows both loads and the ratio for any day. If you are cutting load before a race on purpose, set your training goal to Tapering, which Titan describes as "Reducing training load to shed fatigue," and expect a Low ratio.
07Using it without over-reading it
The ratio earns its place as a quick check on how fast your training is changing. Use it to catch unplanned spikes, such as a return from illness, a travel week followed by a double catch-up, or a sudden jump in training volume. Pair it with how you feel and with recovery markers, since a high ratio with normal sleep and resting heart rate means something different from a high ratio alongside the signs of overreaching. Plan increases around progressive overload and how your body answers each step. No ratio target has been shown to keep anyone healthy.
08References
- Andrade R et al. (2020). Is the acute: chronic workload ratio (ACWR) associated with risk of time-loss injury in professional team sports? A systematic review of methodology, variables and injury risk in practical situations. Sports Medicine 50(9):1613-1635. https://doi.org/10.1007/s40279-020-01308-6
- Apple (n.d.). Track your training load on Apple Watch. Apple Watch User Guide. https://support.apple.com/guide/watch/track-your-training-load-apde4c07a6cf/watchos
- Dalen-Lorentsen T et al. (2021). Does load management using the acute:chronic workload ratio prevent health problems? A cluster randomised trial of 482 elite youth footballers of both sexes. British Journal of Sports Medicine 55(2):108-114. https://doi.org/10.1136/bjsports-2020-103003
- Gabbett TJ (2016). The training-injury prevention paradox, should athletes be training smarter and harder? British Journal of Sports Medicine 50(5):273-280. https://doi.org/10.1136/bjsports-2015-095788
- Garmin (2025). Acute load. quatix 7 Series Owner's Manual. https://www8.garmin.com/manuals/webhelp/GUID-6D76A13F-2195-4287-9B0C-2124AECF9717/EN-US/GUID-AEDB0872-C5A1-4378-86D5-2239734B59E8.html
- Garmin (2025). Load ratio. tactix 7 Owner's Manual. https://www8.garmin.com/manuals/webhelp/GUID-AC520B63-3C82-4266-90F6-6E9F22D5F76E/EN-US/GUID-200689D7-F65C-40F0-BB82-3C51236C676A.html
- Griffin A et al. (2020). The association between the acute:chronic workload ratio and injury and its application in team sports, a systematic review. Sports Medicine 50(3):561-580. https://doi.org/10.1007/s40279-019-01218-2
- Hulin BT et al. (2014). Spikes in acute workload are associated with increased injury risk in elite cricket fast bowlers. British Journal of Sports Medicine 48(8):708-712. https://doi.org/10.1136/bjsports-2013-092524
- Hulin BT et al. (2016). The acute:chronic workload ratio predicts injury, high chronic workload may decrease injury risk in elite rugby league players. British Journal of Sports Medicine 50(4):231-236. https://doi.org/10.1136/bjsports-2015-094817
- Impellizzeri FM et al. (2020). Acute:chronic workload ratio, conceptual issues and fundamental pitfalls. International Journal of Sports Physiology and Performance 15(6):907-913. https://doi.org/10.1123/ijspp.2019-0864
- Impellizzeri FM et al. (2020). Training load and its role in injury prevention, part 2, conceptual and methodologic pitfalls. Journal of Athletic Training 55(9):893-901. https://doi.org/10.4085/1062-6050-501-19
- Impellizzeri FM et al. (2021). What role do chronic workloads play in the acute to chronic workload ratio? Time to dismiss ACWR and its underlying theory. Sports Medicine 51(3):581-592. https://doi.org/10.1007/s40279-020-01378-6
- Lolli L et al. (2019). Mathematical coupling causes spurious correlation within the conventional acute-to-chronic workload ratio calculations. British Journal of Sports Medicine 53(15):921-922. https://doi.org/10.1136/bjsports-2017-098110
- Lolli L et al. (2019). The acute-to-chronic workload ratio, an inaccurate scaling index for an unnecessary normalisation process? British Journal of Sports Medicine 53(24):1510-1512. https://doi.org/10.1136/bjsports-2017-098884
- Murray NB et al. (2017). Calculating acute:chronic workload ratios using exponentially weighted moving averages provides a more sensitive indicator of injury likelihood than rolling averages. British Journal of Sports Medicine 51(9):749-754. https://doi.org/10.1136/bjsports-2016-097152
- Qin W et al. (2025). Acute to chronic workload ratio (ACWR) for predicting sports injury risk, a systematic review and meta-analysis. BMC Sports Science, Medicine and Rehabilitation 17(1):285. https://doi.org/10.1186/s13102-025-01332-x
- Shcherbina A et al. (2017). Accuracy in wrist-worn, sensor-based measurements of heart rate and energy expenditure in a diverse cohort. Journal of Personalized Medicine 7(2):3. https://doi.org/10.3390/jpm7020003
- Soligard T et al. (2016). How much is too much? (Part 1) International Olympic Committee consensus statement on load in sport and risk of injury. British Journal of Sports Medicine 50(17):1030-1041. https://doi.org/10.1136/bjsports-2016-096581
- Wang C et al. (2020). Analyzing activity and injury, lessons learned from the acute:chronic workload ratio. Sports Medicine 50(7):1243-1254. https://doi.org/10.1007/s40279-020-01280-1
- Williams S et al. (2017). Better way to determine the acute:chronic workload ratio? British Journal of Sports Medicine 51(3):209-210. https://doi.org/10.1136/bjsports-2016-096589
