There is a particular feeling at the top of a hard set. The bar slows, your breath catches, and a quiet question arrives: could I have done one more? Most lifters never answer it honestly. They either stop well short, chasing a number on a program, or they grind every set to a trembling halt and wonder why they feel wrecked by Thursday. Between those two habits sits one of the most useful skills in strength training, and it has an unglamorous name.
It is called reps in reserve.
The question hiding inside every set
Reps in reserve, or RIR, is exactly what it sounds like: the number of reps you had left in the tank when you stopped. Finish a set of squats and feel certain you had two more clean reps in you? That was a set at 2 RIR. Stop the instant the bar refuses to move? That was 0 RIR — true muscular failure.
It sounds almost too simple to matter. But that estimate is a direct read on the one variable that actually drives adaptation: how hard the set was relative to what you were capable of. A set of ten with five reps left is a warm-up. The same set of ten with one rep left is a stimulus. The weight on the bar tells you nothing about which one happened. Only proximity to failure does.
This is why two lifters can do identical numbers on paper and grow at completely different rates. The load, the sets, the reps — all matched. What differs is effort, and effort is invisible unless you have a language for it.
Why the last reps are the ones that count
The reason RIR matters comes down to how muscle fibers get recruited. Your body is economical. When a set begins and the weight feels easy, it calls on only as much muscle as the job requires, leaning on the smaller, fatigue-resistant motor units first. The large, high-threshold fibers — the ones with the most growth potential — stay on the bench.
As the set wears on and the easy fibers tire, your nervous system is forced to recruit those larger units to keep the bar moving. This is the size principle of motor recruitment, and it is one of the better-established ideas in exercise physiology. The practical consequence is blunt: the early reps of a set are mostly a toll you pay to reach the reps that matter. The stimulating reps live near the end, in the territory where things get slow and ugly.
This is also why bar speed is such an honest informant. Early in a set, your reps move at roughly the same velocity even as the weight gets heavier, because you are not yet trying hard. It is only as you approach failure that the bar visibly slows no matter how violently you push. That slowdown is the recruitment of your strongest fibers made visible. When you learn to read it, you are reading your own RIR in real time.
You do not actually need to fail
Here is the part that surprises people: you do not have to reach 0 RIR to grow. A growing body of research comparing sets taken to failure against sets stopped a couple of reps short keeps landing in the same place — when the number of hard, close-to-failure reps is similar, muscle growth is comparable. Training all the way to failure does not buy you much extra, and it charges a steep price for the privilege.
That price is fatigue. A set taken to true failure is dramatically more taxing on your nervous system, your joints, and your recovery than a set stopped at 1 or 2 RIR — far more than the single extra rep would suggest. Push every set to the absolute limit and the cost compounds across a session and across a week. Your later sets degrade, your form frays, and the quality of your hard reps drops just when you need them most. You end up doing more grinding for less actual stimulus.
This is the quiet argument for leaving a little in the tank most of the time. For building muscle, training in the neighborhood of 1 to 3 reps in reserve on your working sets captures nearly all of the benefit while leaving you intact enough to do it again two days later. For heavy strength work, you often want even more in reserve, because the goal is to practice moving big loads crisply, not to bury yourself. Failure is a tool you reach for deliberately and sparingly — maybe on the last set of an isolation movement where the stakes are low — not a default setting.
Learning to read your own effort
The catch is that estimating RIR is a skill, and beginners are bad at it. This is not a character flaw; it is a perception problem. New lifters routinely stop with four or five reps left while feeling like they were at the edge, because the sensation of working hard is unfamiliar and the discomfort reads as a red light when it is really just yellow.
The estimate sharpens with experience, and it sharpens fastest when you actually check. Occasionally — not every session — take a set you would normally end at 2 RIR and push it honestly to failure. Count the extra reps. Most people discover they had more than they thought. Each of those reality checks recalibrates your internal gauge a little, and over months your guesses converge on the truth.
Velocity is your other teacher. Pay attention to the rep where the bar speed noticeably breaks — where a rep that should take a beat suddenly takes two. That break tends to show up in a consistent place relative to failure for a given exercise and rep range. Once you know where your bar slows, you have an external check on an internal feeling, and the two together make your RIR estimates far more reliable than either alone.
The case for writing it down
None of this works as a vague intention. "Train close to failure but not to it" is the kind of advice that evaporates the moment a set gets hard and adrenaline takes the wheel. The estimate only becomes useful when it becomes data — when you can look back and see that last week's top set of bench was 8 reps at 2 RIR, and decide, deliberately, whether this week's should be 9 reps at the same effort or 8 reps a notch heavier.
That is the difference between lifting and training. Lifting is showing up and doing something hard. Training is steering effort on purpose, week over week, so that the hard thing keeps producing change instead of just producing exhaustion. RIR is the steering wheel, but only if you record where you pointed it.
This is the unglamorous reason a good log earns its place in your gym bag. Rep lets you note not just the weight and reps but how a set actually felt — how many you left behind — so the number on the bar finally has the context that makes it mean something. Over a few months that history becomes a portrait of your own effort, honest in a way memory never is, and it quietly answers the question you have been asking at the top of every hard set. If you want a place to start keeping that record, it lives at https://rep.lumenlabs.works.