Judging reps in reserve is a skill, not a feel — six sessions cut the error by 2.3 reps
In short
Estimating RIR (reps in reserve — how many reps remain before failure) is not an innate feel but a trainable skill. Thirteen younger adults (33±7 years) and thirteen older adults (67±4 years) completed six sessions over six weeks of bench press and leg press at randomized loads of 75–85% of 1RM, calling out their 4RIR and 2RIR points before continuing to concentric failure. Mean absolute error fell by up to 2.3 reps from session one to session six, with no significant effect of age group and no session-by-group interaction — the younger group was neither more accurate to begin with nor faster to improve.
The design is simple. Two groups of thirteen trained six times across six weeks, doing three sets each of bench press and leg press per session. Load was randomized set by set within 75–85% of 1RM, which stops participants from memorizing “this weight is an eight-rep set.” In every set they verbally marked the rep at which they believed they had four reps left, then two reps left, and then kept going to actual concentric failure. That puts the guess and the answer inside the same set.
Why did six sessions change anything?
Because the answer arrived immediately. Most lifters use RIR for years without ever reaching failure, which means the call “two left” is never checked against anything and a wrong sense of effort simply hardens. The 2.3-rep improvement here came from no special method — it came from closing the feedback loop. Randomizing load served the same purpose: it forces a judgment made from today's sensation rather than from memory.
Age was not a factor
That was the study's second question. The 67±4 group was neither less accurate at baseline nor slower to improve than the 33±7 group, with no main effect of age group and no session-by-group interaction. That cuts against the assumption that RIR-based autoregulation is too demanding for older lifters. The authors conclude that after brief familiarization it is a feasible tool for individualized exercise prescription in this population. For how training structure changes with age, see training hard after 50.
What this means for your log
The point of RIR is hitting the right intensity without testing a max every week. But if the estimate is off by two reps, a set planned at 75% is actually run at 85%, and weekly volume drifts with it. So treat RIR as an instrument that needs calibration, not a permanent sense. Once a month, taking one set of a light accessory lift to failure and checking it against your estimate is enough. Pair it with bar speed as a proxy for percentage and velocity loss cutoffs. The big three maxes you log in fairlift only produce a meaningful Muscle Index if they come from calibrated estimates.
Taking every set to failure was an experimental condition for measuring accuracy, not a training recommendation. Never go to failure on squat or bench press without safeties or a spotter — calibrate on machines or light accessory work instead.
Frequently asked questions
What does RIR mean?
RIR stands for reps in reserve — how many additional repetitions you could have completed when you stopped the set. 2RIR means two reps were left. It is used to control training intensity without testing a one-rep max.
Does RIR estimation get more accurate with practice?
Yes. Across six sessions in six weeks, mean absolute error for 2RIR and 4RIR estimates fell by up to 2.3 reps on both bench press and leg press. The authors concluded that RIR estimation is a trainable skill.
Do older lifters estimate RIR less accurately?
Not in this study. A group averaging 67 years was statistically indistinguishable from a group averaging 33 years in both baseline accuracy and rate of improvement, with no main effect of age group and no session-by-group interaction.
How do you calibrate your sense of RIR?
Make the estimate, then continue to actual failure and compare. Stopping at the estimate never reveals the answer, so an inaccurate sense of effort persists. For safety, do this on a machine or a light accessory lift, roughly one set a month.
What happens to a program when RIR estimates are wrong?
An error of two reps turns a set planned at 75% of 1RM into one performed at roughly 85%. Weekly volume and recovery planning drift with it, which is why RIR is better handled as a periodically calibrated instrument than as a fixed feel.
Source: PubMed