Training effort
RIR Explained: How Hard Should Each Set Actually Feel?
Reps in reserve is the simplest way to control training effort - once you learn to estimate it. What RIR means, why beginners guess wrong, and how to calibrate.
# RIR Explained: How Hard Should Each Set Actually Feel?
"Three sets of ten" tells you almost nothing. Ten reps with a weight you could lift twenty times is a warm-up. Ten reps with your true ten-rep max is a small war. Same prescription, completely different training - and the difference between them is the single most under-coached variable in the gym.
RIR - repetitions in reserve - describes how many technically acceptable repetitions you believe remained when a set ended. RIR 2 means you estimate that two clean repetitions were left; RIR 0 means you reached technical or momentary failure. It is an estimate, not a laboratory measurement, but it adds useful information that sets and repetitions alone cannot provide.
Why not just train to failure every set?
It's a fair question - failure is at least unambiguous.
Training to failure is not consistently superior to stopping short of failure for strength or hypertrophy. It can also reduce repetitions in later sets and increase fatigue, especially in demanding compound lifts. The practical point is not that failure is harmful; it is that you do not need it on every set to make progress.
The practical trade-off is simple: failure can be useful, but its extra fatigue is not always worth paying.
Which doesn't make failure useless - it has two legitimate jobs. On the last set of a low-risk isolation exercise (a lateral raise, an EZ-bar curl, a machine), the cost is small and the effort practice is valuable. And as an occasional calibration tool, it's irreplaceable, which we'll get to.
The uncomfortable truth: you're probably guessing wrong
RIR estimates are imperfect. Research shows that estimates tend to become more accurate as a set gets closer to failure and when heavier loads are used. Higher-repetition sets are harder to judge because discomfort can arrive well before the muscle reaches failure. Training experience may help in some situations, but the evidence does not support assuming that every beginner is inaccurate or every experienced lifter is precise.
The pattern is predictable. On a set of 15-20, the discomfort of reps 10-12 reads as "almost done" when the muscle has six reps left. On heavy sets of 3-5, estimates get much more accurate - bar speed tells the truth loudly. And accuracy improves fast with practice, which is the actual point: RIR isn't a talent, it's a skill with a known training method.
A practical calibration drill: occasionally, on the final set of a stable, low-risk exercise, record your RIR estimate and then continue to technical failure. The difference gives you feedback for future estimates. Use a machine, an isolation exercise or a push-up variation. Do not use heavy barbell squats or deadlifts for this experiment.
What RIR to use, and where
The working ranges that show up in well-built programs, and the reasoning:
Big compound lifts - RIR 1-3. Squats, bench press, deadlifts, barbell rows. Hard enough to drive progress, buffered enough that form holds and the fatigue bill stays payable. The heavier and more technical the lift, the more that buffer earns its keep.
Isolation and machine work - RIR 0-2. Curls, raises, face pulls, leg curls. Low systemic cost, low technical risk, so you can afford to flirt with failure - and on the last set, occasionally marry it.
Beginners - often RIR 3-4 while learning. This leaves room to practise technique and recover between sessions. As movement skill and confidence improve, some sets can gradually move closer to failure.
Deloads and returns from time off - RIR 4-5. Easy weeks feel pointless until you notice how much better the next month goes.
A useful mental model: across a program, sets mostly live at RIR 1-3, the average drifting harder as a training block progresses - week one might sit at RIR 3, the final week at RIR 1 before an easy week resets things.
RIR versus percentages
Percentages of one-repetition maximum provide an objective starting load, while RIR adjusts for day-to-day changes in readiness. Neither method is automatically better. Percentages are useful when load specificity matters; RIR is useful when the same prescribed load feels different because of sleep, stress or accumulated fatigue. Many programs use both.
The practical synthesis most lifters land on: pick starting weights with rough percentages or history, then let RIR steer week to week. Weight goes up when the target RIR gets easy to hit at the top of the rep range. That's double progression with a effort gauge attached, and it runs for months without a spreadsheet.
Putting a number on every set
Once RIR clicks, vague programs become precise ones. "3×10" becomes "3×10 @ RIR 2" - now it means something, and a month later, when the same weight at the same reps sits at RIR 4, you have an objective green light to load the bar.
The Kysero plan builder adds an editable RIR target only where it makes sense: rep-based strength and hypertrophy work. Timed holds, mobility work, warm-ups and other prescriptions that do not use RIR leave the field empty. Whether you use the builder or a notebook, record the estimate consistently and treat it as feedback rather than perfect data.