Is the drop really comparable?
A performance drop means fewer reps with the same load and similar effort, or a lighter load for the same set. It is not a clean comparison if you changed the variation, range of motion, setup, exercise order, or rest time.
First retrieve these details from your workout log. RIR is the number of reps you think you could still do at the end of a set. It is subjective: use it as context, not as an exact verdict.

What should you do during the workout?
After warming up, do your first work set with the planned load. If you remain within your rep range and the movement is controlled, finish the workout without trying to make up for the dip. If you clearly miss the range or technique breaks down, lower the load by the smallest available increment and keep the remaining sets controlled. Stop the movement if unusual pain appears.
Practical example, not a research threshold: you plan three sets of 8–12 reps at 60 kg. The first yields 7 reps with deteriorating technique. Try 57.5 kg for the remaining sets if that increment is available; record 60 kg × 7 separately, followed by the sets you actually did at 57.5 kg. Do not rewrite the workout as three sets at one load.
When should you change the program?
One weaker workout does not demonstrate muscle loss or a lasting plateau. Today’s performance reflects adaptation, fatigue, and outside factors; monitoring models describe this variation but do not set a universal number of workouts to wait.
The review by Greig and colleagues describes daily fluctuations and the limits of decisions based on one test. As a practical rule, keep the exercise and review the next two or three comparable exposures. If the drop repeats across several lifts, check sleep, recovery, exercise order, and volume before changing one program variable.
For a persistent decline, read the criteria for changing a workout program and the guide to a deload week.
What the evidence shows, and what remains a rule of thumb
A review by Helms and colleagues describes ways to adjust load or volume to daily performance. Findings come mainly from athletic settings and varied protocols. They support adjusting a workout to observed performance; they do not prove that reducing the load by exactly 2.5 kg or waiting exactly three sessions is best for everyone.
The load increment and observation period depend on the exercise, your experience, and your usual workout variability. Keep the raw values in your log: they are more useful than labeling a workout simply “good” or “bad.”
Sources and methodology
Both reviews below were read in full. They mainly cover monitoring and autoregulation of performance across heterogeneous populations and protocols. The example’s numbers and the suggestion to review two or three workouts are practical rules, not validated prescriptions.
Frequently asked questions
Should I remove a set after a weak first set?
Not automatically. If technique is sound, keep the planned structure and adjust load if needed. Reduce the work if later sets also deteriorate.
Does one fewer rep mean I am regressing?
No. Check effort, technique, and context, then compare several sessions of the same exercise. One value cannot establish a trend.