The Last Three Reps. How Close To Failure You Actually Need To Train, And How Much Of Your Volume Is Wasted.
Most of the growth in a set lives in its final reps, the ones that feel hard. Train too far from failure and the work barely counts. Push to absolute failure on everything and you bury yourself in fatigue. This article maps proximity to failure, reps in reserve, effective reps and junk volume into a way of training that earns its results.
Opening. The two lifters in every gym
Walk into any gym and you can find two lifters making opposite mistakes with the same earnest face. The first finishes every set with three or four reps still comfortably in the tank, racks the weight, and moves on feeling like they trained. The second takes every set to the absolute limit, grinding the last rep for ten shaking seconds, then drags themselves through a session so brutal they can barely recover from it.
Both have misunderstood the same thing. How hard a set needs to be to actually drive growth. One trains too easy and wonders why the body never changes. The other trains so hard on everything that fatigue swallows the results.
The answer sits in a small body of training science that has matured nicely over the last decade. Where the growth in a set actually lives, how close to failure you need to get to claim it, and how much of the volume people pile up is quietly doing nothing. This is the article that makes the last few reps make sense.
Section one. Where growth lives in a set
Start with the core idea, because everything else hangs off it. Not all reps in a set are equal. The reps that drive the most adaptation are the hard ones near the end, when the muscle is fatiguing and has to recruit its largest motor units to keep moving the weight.
The mechanism is recruitment. Early in a set, when the weight feels light, the body uses its smaller, more fatigue resistant muscle fibres and leaves the largest ones in reserve. As the set wears on and those smaller fibres tire, the body is forced to call in the large, high threshold fibres to finish the job. Those large fibres have the greatest growth potential, and they only get meaningfully challenged in the difficult final reps. This is the idea behind what people call effective reps. The reps close to failure, where the big fibres are working hard, are the ones that count most.
This single principle explains both lifters in the opening. Stop too early and you never reach the effective reps, so the set barely registers. The hard part is not the price of the growth. The hard part is the growth.
Section two. Reps in reserve, the language of effort
To train this on purpose you need a way to measure how hard a set was, and the most useful tool is reps in reserve, usually shortened to RIR. It is simple. RIR is how many more reps you could have done before reaching failure. Finish a set knowing you had two more in you and that set was two RIR. Finish a set where you genuinely could not have done another and that was zero RIR, or true failure.
RIR turns a vague feeling into a dial you can set. A set at three or four RIR is moderately hard and leaves real reps on the table. A set at one to two RIR is genuinely challenging and reaches the effective reps. A set at zero RIR is taken to failure.
The practical value is that it lets you prescribe effort, not just weight and reps. Two lifters doing the same weight for the same reps can be having completely different sessions if one stops at four RIR and the other at one. The number tells you which set actually mattered.
Section three. How close is close enough
So how near to failure do you actually need to be. Here the research has converged on a reassuring answer. For hypertrophy, you do not need to train every set to failure, but you do need to get reasonably close. Training within roughly one to three reps of failure on most working sets captures the great majority of the growth stimulus, without the heavy fatigue cost of going all the way.
This is liberating in both directions. It tells the lifter leaving four or five reps in the tank that they are training too easy and short changing their results. And it tells the lifter grinding every set to a shaking standstill that they are paying a large fatigue tax for a small or even negative return. The sweet spot for most working sets is hard, close to failure, but not maximal on everything.
Strength has a slightly different flavour. Maximal strength is partly a skill of moving heavy loads efficiently, so heavy strength work often lives a little further from failure, leaving cleaner reps to groove the pattern, as we argued in Strength Is A Skill. Hypertrophy work lives closer to the edge. Same dial, set differently for the goal.
Section four. Junk volume, the work that does nothing
This brings us to the least glamorous and most useful concept in the article. Junk volume. Sets that add fatigue and time to your session without adding a meaningful training stimulus.
Junk volume comes in two flavours. The first is sets taken too far from failure to reach the effective reps, the easy sets that feel like work but do little. The second is the opposite, the extra sets piled on after the productive work is done, when the muscle is already fatigued, performance is dropping, and each additional set adds more recovery cost than stimulus. The lifter doing twenty hard sets for a small muscle is not getting twenty sets of growth. They are getting a handful of productive sets and a long tail of junk that they still have to recover from.
This matters because recovery is finite. Every junk set spends recovery capacity you could have given to productive work or to the rest of your life. More volume is not automatically more growth. Past a point it is just more fatigue. The skill is doing enough hard, close to failure sets to drive adaptation, and then stopping before the junk begins.
The trap is that junk volume feels productive in the moment. A long session, a high set count and a pump all register as hard work, and the lifter leaves feeling they earned something. But the body does not read effort by the clock or the set count. It reads the quality of the stimulus and the recovery cost it has to pay afterward. A short session of genuinely hard, close to failure sets can outperform a marathon of easy and redundant ones, while leaving you fresh enough to train again sooner. Volume is a means, not a virtue. The lifter who falls in love with the number of sets rather than the quality of them is working hard in the service of very little.
Section five. Autoregulation, training the day you are in
The last piece is what to do with all this on a real day, because you are not a spreadsheet. Some days you are fresh and strong. Some days you slept badly, ate poorly or carried stress in, and the same weight feels like a different weight. Training a fixed number regardless of how you feel ignores this. Autoregulation is the answer.
Autoregulation means adjusting your training in real time to the state you are actually in, usually by training to a target effort rather than a fixed load. Instead of forcing a planned weight, you work to a target RIR. On a strong day that target weight goes up to keep the effort honest. On a flat day it comes down, and you still get a quality, close to failure stimulus without grinding yourself into the ground chasing a number your body cannot pay for today.
This is how experienced lifters last. They stop treating the program as a contract that must be honoured rep for rep and start treating it as a framework they steer by feel and by the bar speed in front of them. The effort stays high. The specific load flexes with the day.
Section six. A defensible effort protocol for the serious lifter
Set the dial by goal. For hypertrophy, take most working sets to roughly one to three reps from failure. For heavy strength work, often leave a little more in reserve to keep the reps clean, since strength is a skill of moving load well.
Use failure as a tool, not a default. Reserving true failure for the last set of an exercise, or for safer single joint movements, lets you reach the edge occasionally without paying the fatigue tax on every set. Avoid routine failure on heavy compound lifts where form breaks down and risk rises.
Cap the junk. Count only your hard, close to failure sets as real volume. When performance within a session clearly drops, set after set, you have reached the productive ceiling. Stopping there beats adding sets you only have to recover from.
Autoregulate the load. Work to a target effort rather than a fixed weight. Let the load rise on strong days and fall on flat ones so the effort stays honest and the quality stays high.
Support the recovery the effort demands. Training closer to failure raises the recovery cost, so the foundations matter more, not less. Adequate protein, creatine monohydrate, sleep, and electrolytes around hard sessions. A sensible pre workout can help on a genuinely flat day, but it does not replace the dial.
Section seven. The mastermind frame
The deepest lesson here is that effort is not the same as suffering, and volume is not the same as growth. The lifter who grinds everything to failure and the lifter who never approaches it have made mirror image errors, both confusing the feeling of a session with its actual effect. The body does not reward theatre. It rewards the right reps, close enough to the edge, repeated across years, with enough recovery left over to come back and do it again.
The serious lifter learns to read the last three reps. They know which sets counted and which were noise. They train hard where it matters, they stop before the junk begins, and they steer by the day they are actually in. Train close enough, often enough, and recover enough to keep doing it, and the body answers across decades.
FAQ
Do I have to train to failure to grow muscle?
No. The research suggests training within roughly one to three reps of failure on most working sets captures the great majority of the growth stimulus, with far less fatigue than training every set to true failure. Getting close matters. Going all the way on everything does not.
What does reps in reserve actually mean?
Reps in reserve, or RIR, is how many more reps you could have done before failure. A set with two RIR means you could have done two more. It turns effort into a number you can target, so you can prescribe how hard a set is, not just the weight and reps.
What is junk volume?
Junk volume is sets that add fatigue without meaningful stimulus. Either sets taken too far from failure to reach the productive reps, or extra sets piled on after performance has clearly dropped. They cost recovery without buying growth.
Is failure training ever useful?
Yes, as a tool rather than a default. Used occasionally, such as on the last set of an exercise or on safer single joint movements, it has a place. Routine failure on heavy compound lifts raises fatigue and injury risk for little extra benefit.
How do I autoregulate without overcomplicating it?
Pick a target effort, such as stopping two reps from failure, and let the weight adjust to hit it. On strong days the weight goes up, on flat days it comes down, and the effort stays consistent. You steer by how the bar moves, not by a fixed number on paper.
Editorial footer
Written for the Supplement Superstore Vault. We sell the supplements that support the work. We do not sell the work. Information here is educational and is not medical advice. Speak with a qualified health professional before changing any protocol, especially during pregnancy, breastfeeding, competitive training or while managing any clinical condition.
