Most lifters don't quit the gym on a bad day. They quit on the day after a small one. Your shoulder pinches on a set of overhead presses, you think "I'll take a few days," and a few days becomes two weeks. Two weeks becomes a month in which you tell people you're "between programs." The injury might have healed in nine days. The habit took far longer to come back, and some people never get it back.

The cruel part is that the pause usually feels like the responsible choice. This article is about a better question than "should I rest or push through?" That question is a trap, because it only has two doors, and both are bad.

The two bad doors

The first door is ignoring pain: grit your teeth, add weight, and treat discomfort as weakness leaving the body. Sometimes this works. Sometimes it turns a tweak into a months-long problem.

The second door is total rest: stop everything until it feels perfect. It feels safe, but it has costs that are easy to miss. Tissue that isn't loaded gets less capable of handling load, not more. Strength falls off, and confidence falls off faster. And you slide into what pain researchers call the fear-avoidance cycle, described by Vlaeyen and Linton and studied mostly in back pain. You hurt, you catastrophize ("I've done something terrible to my spine"), you avoid movement, you deteriorate, and the next twinge feels even more threatening.

The back-pain literature is fairly consistent on this point. For most non-specific back pain, staying active generally beats prolonged bed rest. The lesson isn't "always push on." It's that rest isn't automatically neutral or safe.

Pain is an alarm, not a ruler

The belief underneath most bad decisions is that pain measures damage: more pain, more harm. Modern pain science, associated with researchers like Lorimer Moseley, says it's closer to an alarm system. Pain is the brain's output when it judges a body part to be under threat. That judgment is influenced by tissue state, but also by stress, sleep, expectation, and fear.

This is why a papercut can hurt sharply while a serious problem can be quiet, and why the same sore knee hurts more on a day you're anxious and sleeping badly. It doesn't mean pain is imaginary. It means a pain number alone can't tell you whether to stop. You need a rule that uses the pain information in context.

The pain-monitoring model

Physiotherapists use one such rule, most closely associated with Karin Silbernagel's work on Achilles tendinopathy. The studied version is specific to tendon problems, but the logic is useful for gym decisions generally. In the original protocol, patients kept exercising while pain stayed at or below roughly 5 out of 10 during activity, and the key checks happened afterward:

  • Does the pain settle back to baseline within about 24 hours?
  • Is morning stiffness or soreness the next day no worse than before?
  • Is the trend over weeks flat or improving, not climbing?

In that trial, patients who exercised under this model, continuing to load the tendon within those limits, did well compared with what you might expect from stopping altogether. Treat that as support for the principle, not a guarantee for your particular body part. A tendon problem is not a fracture, a torn muscle, or a nerve issue.

The principle is what matters. You're not asking "does it hurt?" You're asking "how does it behave over the next day?" An irritation that spikes and calms is a conversation. One that lingers, builds, or wakes you at night is the body shouting.

Red flags that skip the experiment

The monitoring approach is for the ordinary aches of training: a cranky elbow, a grumbling knee, a tight low back. It isn't for everything. See a clinician promptly if you have:

  • A pop or tearing sensation followed by sudden weakness or visible swelling or bruising
  • Numbness, tingling, or weakness spreading down an arm or leg
  • Pain that is constant, severe at rest, or worse at night
  • Pain after a significant fall or impact
  • Chest pain, dizziness, or anything that doesn't feel musculoskeletal

None of this replaces a professional who can examine you. If an issue is stubborn past a couple of weeks, get it looked at. A good physio will often tell you what you can do, which is the information you actually need.

Train what isn't broken

Here's the quiet advantage of the model: an injury is rarely a whole-body event. A sore shoulder doesn't stop your legs from squatting. Tender elbows don't mean you can't do hip hinges. Most injuries leave a large menu of movements available, and using that menu is how you keep the habit alive.

There's also a research curiosity worth knowing. Training one limb can slow strength loss in the other, a phenomenon called cross-education, and in at least one immobilization study, people who trained the free arm kept more strength in the casted one. It's no substitute for rehab, but it's a reason "I can't train" is rarely literally true.

On the injured area itself, the usual adjustments are surprisingly mild: lighter load, a shorter range of motion, a slower tempo, a different grip or stance, or a variation that feels clean. Many lifters find a pain-free version of the lift within a few minutes of experimenting.

The identity problem nobody mentions

Something else happens during an injury layoff, and it's not physical. If lifting is part of how you see yourself, as someone disciplined, someone who shows up, then stopping feels like stepping out of that person. People describe it as low-grade grief. The surest way to protect against it is to keep a thin thread of the routine going, even at a ridiculous 40% of normal volume. You're still a person who trains. You just trained around something this month.

Your next moves

  • Rate it tonight. Write down where it hurts, what movement triggers it, and a pain number from 0 to 10. A vague worry becomes a measurable thing.
  • Run the 24-hour check. After your next session, note the pain that evening and the next morning. Returned to baseline means the load was probably tolerable. Worse than before means scale back next time.
  • Pick three pain-free movements. List exercises that don't provoke it, and build your next workout from them rather than skipping the gym.
  • Test one modification. Try one change to the painful lift: 20 to 30% less weight, a shorter range, or a slower lowering phase. If it stays under about 4 out of 10 and calms down afterward, keep it for the week.
  • Set a see-someone date. Choose a deadline, such as 10 to 14 days, after which a stubborn or worsening problem goes to a physio or doctor. Decide it now, while you're calm.

Where a log earns its keep

The pain-monitoring model runs on information you can't reliably hold in your head: what you lifted, how it felt that evening, how it felt the next morning, whether it's trending up or down over weeks. Memory is also a poor witness when you're worried. A fearful brain remembers the worst twinge and forgets the ten fine sessions. That's why we built Rep as a fast, clean log where you can note a set, a weight, and a quick comment about how something felt, then look back across weeks and see the actual trend instead of the story fear tells. It's a one-time purchase, no subscription. If you want to try it, you can find it at rep.lumenlabs.works. Whatever you use, keep the thread going. The injury is usually shorter than the silence you let grow around it.