Acute:Chronic Workload Ratio (ACWR)
ACWR compares recent load with longer-term baseline load to flag unusually fast training-load increases.
Acute:Chronic Workload Ratio (ACWR)
You can think of ACWR as a simple question: “Did I ramp up too fast?” It compares what you did recently (acute) with what you’ve been accustomed to (chronic), and is commonly used to monitor load progression, especially in higher-impact sports like running.
ACWR should be treated as a screening signal, not an injury prediction model. It helps identify aggressive load changes that deserve closer review.
Example card: values are for UI preview only. Interpret using your own data and trends.
Core Concepts
1. Acute load
Usually the average load over the last 7 days (similar to ATL in PMC). It reflects your current fatigue/stress.
2. Chronic load
Usually the average load over the last 28 days (or 42 days, depending on implementation), similar to CTL. It reflects fitness/adaptation.
3. Ratio calculation
A:C = AcuteLoad / ChronicLoad ≈ ATL / CTL(Note: in many PMC implementations this is approximated by ATL / CTL, though details vary depending on rolling averages vs EWMA.)
Interpretation: load-change ranges
A:C works like a traffic light for whether load has changed unusually fast:
| A:C Ratio | Status | Meaning / guidance |
|---|---|---|
| < 0.8 | Low recent load | Load is well below your usual baseline. This may be recovery or detraining depending on context. |
| 0.8 - 1.3 | Controlled change | Often treated as a moderate range where recent load is close to baseline. |
| 1.3 - 1.5 | Warning | Load is rising fast. Watch your body and recovery closely. |
| > 1.5 | High change | Recent load is much higher than baseline. Review pain, sleep, prior injury, and upcoming intensity before adding more load. |
How Trainingload.ai uses ACWR
Trainingload.ai uses ACWR to flag load spikes before plan adjustments. If A:C rises sharply, the AI coach can review whether the next workouts should be kept, softened, or moved.
The ratio is never enough by itself. The more useful decision combines A:C with ATL, CTL, pain feedback, workout type, and the athlete’s injury history.
ACMP and variants
While Gabbett’s original framing often used rolling averages, modern PMC tools (WKO5, TrainingPeaks, Trainingload.ai) commonly compute the ratio using EWMA ATL and CTL:
- EWMA model: weights recent training more heavily and is often more responsive than rolling averages.
- Uncoupled vs coupled:
- Coupled: acute load is included inside the chronic window.
- Uncoupled: chronic load excludes the acute window (only prior days).
- Trainingload.ai defaults to the standard PMC-style EWMA logic: ATL (7d) / CTL (42d).
Practical tips
- Avoid the “weekend warrior” pattern: low chronic load during weekdays, then very high acute load on weekends can spike A:C quickly.
- Be patient after injury: chronic load (denominator) becomes small after time off. Even a small comeback week (numerator) can produce a very high A:C. Rebuild gradually over weeks.
- Don’t chase a high ratio: tolerance varies. For some athletes 1.3 is already too aggressive. Adjust based on history and how you respond.
References
- The Science of the TrainingPeaks Performance Manager (Andrew Coggan)
- Modeling Human Performance (Fellrnr)
- Gabbett TJ. The training-injury prevention paradox: should athletes be training smarter and harder? Br J Sports Med. 2016.
Training Stress Balance (TSB)
Training Stress Balance (TSB) is derived from CTL and ATL and helps estimate freshness, fatigue, and readiness within a training load system.
Training Monotony
Training monotony describes how similar daily training load is across a week and helps identify overly uniform training structure.