Injury Risk
LiftShift estimates joint stress from 0 to 100 percent per joint so you can spot load-management risks early.
What the score means
The Injury Risk card scores every joint you train from 0 to 100 percent. Lower is safer. The score is an estimate of how much recent training stress a joint is under — it is not a prediction of injury and not medical advice. A spike above 40 percent suggests considering a deload or rebalancing your program.
The score combines three factors, each shown color-coded in the card: workload (this week’s sets vs. your 4-week average, mirroring the acute:chronic workload ratio), recovery (back-to-back days training the same joint), and balance (volume ratio between an antagonist pair, e.g. quads vs. hamstrings).
How it is calculated
- Workload ratio — sets on the joint this week divided by the 4-week rolling average. A ratio above ~1.3 is elevated; higher spikes push the score up sharply.
- Recovery — counts consecutive days the joint was worked. Each back-to-back day increases the recovery component.
- Antagonist balance — compares weekly sets between opposing muscle groups around the joint. Large imbalances (e.g. far more pressing than pulling) add to the score.
Muscles are attributed to joints through the exercise-to-muscle mapping used across LiftShift, with warm-up sets excluded and secondary muscles weighted lower.
How to use it
- Load management — if a joint’s score climbs week over week, reduce frequency or volume on the exercises that hit it hardest.
- Program balance — a high balance component points at an antagonist pair, e.g. add pulling volume if pressing dominates.
- Deload timing — scores above 40 are a reasonable cue to schedule a deload week.
Limitations
- The score is derived only from what you log. Technique, sleep, stress, and existing conditions are invisible to it.
- It estimates relative training stress, not injury probability. A low score does not guarantee safety, and a high score does not mean you will get injured.
- ACWR-style metrics are most meaningful with consistent logging over several weeks. Short histories produce noisier scores.
Learn more
See the How it works guide for the full methodology and the FAQ for common questions.