Sam had lifted for four years and run push pull legs for the last two. Bench, incline, overhead press, three upper body days a week. Around week nine of a pressing block the shoulders stopped feeling fresh. Nothing was injured. They were just "grumpy" every time Sam went overhead, and the overhead press had stalled.
The app Sam used gave the same answer every week: shoulders, trained three times, recovered. True, and no help at all, because "shoulders" is at least three different muscles.
Looking at the shoulder in pieces
Sam switched to tracking recovery per subzone. Valen split the shoulder into front, lateral and rear delts, plus traps, and looked at each one's share of Sam's total shoulder volume over the last 21 days:
| Subzone | Sam's share | Healthy range | Verdict |
|---|---|---|---|
| Front delt | 52% | 18% to 42% | High |
| Lateral delt | 41% | 28% to 58% | In range |
| Rear delt | 4% | 14% to 36% | Critically low |
| Traps | 3% | 4% to 20% | Slightly low |
The front delts were doing more than half the work, because every bench press, incline press and overhead press loads them. The rear delts were getting almost nothing. Nobody had chosen that. It is what happens when a program is built around pressing and the pulling work is all rows and pulldowns that mostly train lats.
What Valen did with it
Valen only gives a verdict once there are at least four sessions inside the 21 day window, so one odd week cannot trigger it. Rear delt is marked as a critical subzone for shoulder balance, so it carried extra weight in the diagnosis.
The corrective session then picked up to four exercises, three sets each, covering the biggest gap in the least time. For Sam that meant Reverse Pec Deck, Face Pull and Reverse Fly, all mapped to rear delts in the exercise library, with some trap and rhomboid work coming along for the ride.
Six weeks later
Sam ran the corrective session once a week and swapped one rowing variation for face pulls. By the end of six weeks the shares looked like this: front delt 38%, lateral 40%, rear delt 17%, traps 5%. Every subzone sat inside its range. The overhead press started moving again, and the shoulders felt less grumpy.
Would that happen for every lifter? No. Which brings us to the honest part.
What this does and does not mean
It is a training balance signal, not a diagnosis. Shoulder pain has many causes: technique, mobility, sleep, load jumps, previous injuries and more. A low rear delt share is one thing worth checking, not the answer to shoulder problems.
The ranges are guidelines. They come from programming principles, for example that pressing-heavy programs tend to leave rear delts behind. Research linking specific muscle imbalances to specific injuries is mixed, so Valen frames this as balance, not risk prediction. No app prevents injury.
If something hurts, see a professional. Not an app.
Check your own numbers without any app
Take your last three weeks of training. Count hard sets that mainly load each delt head: presses for front, lateral raises for lateral, face pulls and reverse flys for rear. Divide each by the total. If rear delts are under about 14% of your shoulder sets, that is worth a look.
Doing that by hand every week is tedious, which is why Valen does it from your log. Every exercise is mapped to the subzones it works, so the breakdown is a by-product of logging, not extra work.
Related: why shoulders never seem to recover and push vs pull imbalance.
Valen is available on Google Play. The recovery map and subzone tracking cost nothing, and the corrective session is part of Pro. Get it on Google Play.
*This case study is illustrative only. Sam is a composite, the numbers are indicative, and the story is based on real experiences of the researchers behind Valen, not on a specific customer. Nothing here is medical advice.