Training to Failure: When It Helps, When It Hurts, and the Research Behind RIR

Walk into any gym and you will find two tribes. One believes every set has to end with a shaking, screaming, spotter-assisted grind, because if the bar did not stop moving, you did not really try. The other treats failure like a dirty word, leaving so much in the tank that they never find out what they are capable of. Both are wrong, and the research on this is now deep enough that we can say so with confidence.

Training to failure has been one of the most argued-about topics in the iron game for fifty years, from Arthur Jones and the high-intensity training crowd through modern hypertrophy science. The good news is that the picture has clarified enormously in the last decade. We now have meta-analyses, controlled trials in trained lifters, and detailed fatigue studies that tell us what failure does, what it costs, and how to get the benefit without paying the full price. The answer is more nuanced than either tribe wants to hear, and far more useful.

This guide covers what failure actually is, why lifters ever thought it was necessary, what the growth and strength research shows, how reps in reserve gives you a better dial than the failure-or-not switch, and how to program the whole thing so you build muscle for years instead of burning out in months.

What Failure Actually Means

Before arguing about whether to train to failure, it helps to define it, because the word gets used loosely. In the research literature, the relevant term is momentary muscular failure: the point in a set where, despite maximum effort, you can no longer complete a rep through the full range of motion with proper form. You tried to lift it, and it did not go up. That is concentric failure, and it is the definition most studies use.

This matters because a lot of what people call "failure" is not. Racking the bar because the reps got hard, because your breathing got heavy, or because the burn became unpleasant is stopping short of failure. Genuine momentary failure is a specific, measurable event, and most lifters, especially newer ones, stop well before they reach it. Research consistently shows that trainees asked to stop at a self-selected hard point leave several reps in the tank they did not know they had.

There are also gradations beyond simple concentric failure. Technical failure is the point where your form breaks down even though you could still move the weight with sloppier mechanics. On a deadlift, that might be the rep where your lower back rounds. Training past technical failure toward true concentric failure is where risk climbs sharply on compound lifts, a distinction we will return to. For now, the key idea is that failure is a defined endpoint, not a vibe, and the whole debate is really about how close to that endpoint you need to train.

Why Anyone Cares: The Recruitment Argument

The classic case for training to failure rests on motor unit recruitment. Your muscles are controlled by motor units, and according to the size principle first described by Henneman, your body recruits them in order, from smallest and weakest to largest and strongest, as force demand rises. The largest, highest-threshold motor units control the fibers with the greatest growth potential, and the logic goes that you must recruit those fibers to grow them.

The failure advocates argue that only by pushing a set to its absolute limit do you guarantee recruitment of every last high-threshold motor unit. There is real physiology behind this. When you lift a submaximal weight for reps, the early reps are easy and use mostly lower-threshold units. As those units fatigue, your nervous system recruits progressively larger ones to keep the weight moving. Sundstrup and colleagues (2012) captured this directly with electromyography: during a set taken to failure with lighter resistance, muscle activation started low and climbed steadily, reaching levels comparable to heavy loading only in the final reps as fatigue forced full recruitment.

So the recruitment argument is not nonsense. Full recruitment does appear to require getting reasonably close to failure, which is exactly why junk-volume half-efforts do not build much. The flaw in the argument is the leap from "you need to get close to failure" to "you need to reach failure." Those final high-threshold units get recruited in the last few reps before failure, not only at the moment the bar stalls. Stopping at two or three reps in reserve still takes you into that high-recruitment zone. The recruitment argument justifies training hard. It does not justify training to absolute failure on every set, and the outcome research bears that out.

What the Research Says About Failure and Growth

This is where the evidence has become genuinely clear. The most comprehensive analysis is the 2023 systematic review and meta-analysis by Refalo and colleagues in Sports Medicine, which pooled the controlled studies comparing training closer to versus further from failure for muscle hypertrophy. The finding was a small advantage for training nearer to failure, on the order of a trivial-to-small effect size, and that edge weakened further once training volume was accounted for. In plain terms, getting close to failure matters, but squeezing out those final one to three reps to true failure adds little to nothing for growth.

A controlled trial reinforces the meta-analysis. Refalo and colleagues (2024) trained resistance-trained adults for eight weeks, comparing sets taken to momentary failure against sets stopped with reps in reserve, and found similar muscle hypertrophy between the groups. When you match the effort into the challenging range and equate volume, the failure group does not out-grow the reps-in-reserve group. This is a repeatable pattern across the literature: the growth curve rises steeply as you approach failure and then flattens, so the last reps buy you a lot of fatigue for very little extra muscle.

The older work from Nóbrega and Libardi and colleagues framed the question well in their 2016 review, asking directly whether resistance training to muscular failure is necessary, and concluding that it is not a requirement for hypertrophy when effort is sufficient. The consensus that has emerged since is remarkably consistent. Proximity to failure is a real variable, and being too far from failure does blunt growth, but you do not need to reach failure to maximize it.

The practical reading of this table is the entire point of the article. The growth column barely changes between 1 to 2 reps in reserve and outright failure, but the fatigue column jumps. You are paying a steep fatigue premium for a marginal, possibly nonexistent, growth return. That premium is manageable if you spend it deliberately and occasionally. It is ruinous if you spend it on every set. For the full picture of how much total work drives growth, our training volume guide covers the dose-response side of the equation that proximity to failure sits inside.

Failure and Strength: A Different Story

If the case for failure is weak for hypertrophy, it is weaker still for maximal strength. Davies and colleagues (2016) ran a meta-analysis in Sports Medicine specifically on training to repetition failure and strength gains, and found no significant difference between failure and non-failure training for strength. Their conclusion was direct: training to failure does not appear to be required for gains in strength, and non-failure training achieves the same result without the discomfort and fatigue.

The broader failure-versus-non-failure meta-analytic work by Grgic and colleagues reached the same place, reporting that training to momentary failure is not superior to non-failure training for strength, and noting that in already-trained individuals, failure is unnecessary to maximize strength gains. Higher velocity-loss thresholds, a way of measuring how deep into fatigue a set goes, also did not reliably produce better outcomes.

There is a clean physiological reason strength cares even less about failure than size does. Maximal strength is heavily a skill: it is the ability to produce force efficiently, coordinate the movement, and express your existing muscle under a heavy bar. That skill is trained by lifting heavy weights crisply and with good technique, not by grinding submaximal weights into the ground. When you train a heavy triple and stop with a couple of reps in reserve, every rep is fast, sharp, and technically clean, which is exactly the practice a strength athlete wants. Grind those reps to failure and the last one is slow, ugly, and reinforces the opposite of what builds strength. This is why powerlifters and strength coaches overwhelmingly program by RPE and reps in reserve rather than chasing failure, and it connects directly to the ideas in our guide on progressive overload.

For strength, clean and heavy beats slow and grinding. Every rep taken to failure is a rep that rehearses failing. Every rep stopped short and sharp is a rep that rehearses succeeding.

Reps in Reserve: The Better Way to Measure Effort

If failure is a blunt on-off switch, reps in reserve is the dimmer dial, and it is the tool that makes all of the above actionable. Reps in reserve (RIR) is simply your estimate of how many more reps you could have completed at the end of a set before hitting failure. Stop a set believing you had three more good reps in you, and you trained at 3 RIR. Stop believing you had none, and you hit 0 RIR, or failure.

RIR is the practical backbone of the modern RPE scale for lifting, formalized by Zourdos and colleagues (2016) and expanded by Helms and colleagues (2016) into a repetitions-in-reserve-based rating of perceived exertion. On that scale, an RPE of 10 means zero reps in reserve, an RPE of 9 means one rep left, an RPE of 8 means two or three reps left, and so on down. The two systems are the same idea expressed two ways, and lifters use them interchangeably.

The value of RIR is that it lets you regulate effort with precision across every set of every session, rather than relying on the crude choice of failure or not. Instead of writing "3 sets of 8," you can write "3 sets of 8 at 2 RIR," which tells you not just how many reps but how hard, and lets you pick a weight that lands you exactly where you want on the effort spectrum. This is the heart of autoregulation: on a strong day the weight that gives you 2 RIR is higher, on a beaten-down day it is lower, and RIR keeps your true training stress consistent even when your daily capacity swings. Our program design guide shows how to build RIR targets into a full training block.

How Accurate Is Your RIR Estimate?

RIR is only useful if you can judge it, and honesty about its limits is what separates good practice from wishful thinking. The research says accuracy is real but imperfect, and it depends on two things: your experience and how close to failure you are.

Zourdos and colleagues (2016) found that experienced lifters gauge their proximity to failure more accurately than novices, and that accuracy is highest near failure. This makes intuitive sense. In the last couple of reps before failure, the bar speed drops sharply and unmistakably, giving you a clear physical signal. Bar velocity slows as you approach failure, and experienced lifters learn to read that slowdown. At 0 to 2 RIR, most trained lifters are quite accurate. At higher RIR values, judging whether you have four reps left or six is genuinely hard, and estimates get fuzzy.

The consistent error, especially in beginners and at lighter loads with high rep counts, is underestimating reps in reserve. Lifters routinely think they are at 1 or 2 RIR when they actually have four or five reps left. In other words, most people stop earlier than they believe, and they train further from failure than they intend. The scoping and validity research on RIR scales notes exactly this drift, with accuracy declining at higher rep ranges and lighter loads where discomfort arrives well before mechanical failure.

Two practical fixes follow. First, calibrate occasionally by taking a set to true failure on a safe exercise, such as a leg extension or a cable curl, and counting how many reps you got past where you would normally have stopped. Most lifters are startled by how many are left. Second, if you are a beginner, assume you are further from failure than you feel, and do not be afraid to push a bit harder on safe movements while you learn the signal. The slowdown in bar speed is your most reliable cue, so pay attention to it rather than to the burn or the effort of breathing.

The Hidden Cost: Systemic Fatigue

Here is the variable that failure advocates chronically underrate. Failure does not just fatigue the set you are doing. It fatigues your whole system, and that cost is paid across your next sets, your next session, and your weekly training volume. This is where the case against constant failure becomes decisive.

The clearest evidence comes from Moran-Navarro and colleagues (2017), who directly compared the recovery time course after training to failure versus stopping short. Lifters performed matched-load protocols, one taken to failure and one not, and the researchers tracked strength and biochemical markers of fatigue afterward. The failure protocol lengthened recovery substantially: markers of both acute and delayed fatigue, along with actual strength performance, took 24 to 48 hours longer to return to baseline compared with the non-failure protocol. Training to failure did not just make that session harder. It compromised the next day or two of training capacity.

Think through what that means over a week. If every session ends in failure, you carry deeper fatigue into every subsequent session, which forces you to either reduce the weight, reduce the volume, or train while systemically compromised. Because total hard-set volume is one of the strongest drivers of hypertrophy, and failure suppresses the volume you can recover from, chronic failure training can actually reduce your weekly growth stimulus even though each individual set feels maximally productive. You trade a marginal per-set gain for a meaningful weekly loss. The fatigue also stacks toward the need for recovery weeks, which is why lifters who chase failure constantly tend to need deloads far more often.

There is a within-session version of this too. Taking your first set of an exercise to failure tanks the performance of every set after it, so you get fewer quality reps on sets two, three, and four. Stopping the early sets a rep or two short lets you maintain reps and load across all your sets, accumulating more total productive volume. On heavy compounds especially, this within-session preservation is a large part of why leaving reps in reserve produces better weekly numbers.

Compound vs Isolation: Train Them Differently

One of the most useful practical distinctions in this whole topic is that failure is not equally costly or equally risky across exercises. The dividing line runs between big multi-joint compound lifts and smaller single-joint isolation movements, and treating them the same is a mistake.

Heavy compound lifts, such as squats, deadlifts, barbell rows, bench press, and overhead press, are where failure is both most dangerous and most fatiguing. These lifts load large amounts of weight across multiple joints, so when technique breaks down near failure, the failure happens under a heavy bar with your spine, shoulders, or knees in a compromised position. A missed squat or a rounded-back deadlift near failure is how people get hurt. On top of the injury risk, these lifts generate the largest systemic fatigue, taxing the nervous system and the entire body far more than an isolation movement does. For both reasons, compounds are best kept one to three reps shy of failure the large majority of the time.

Isolation and machine exercises, such as leg extensions, leg curls, cable curls, triceps pushdowns, lateral raises, and most machine movements, are the opposite. A missed rep carries almost no risk, because the weight is light relative to your body, the range is controlled, and often a machine or the floor simply catches it. The systemic fatigue cost is low because you are working a smaller muscle mass. This makes isolation work the ideal place to spend your closest-to-failure efforts. You can take a leg extension or a cable curl to genuine failure, recruit everything, chase the growth stimulus hard, and pay very little for it in recovery or risk.

This gives a clean programming heuristic that resolves most of the practical questions. Save your compound lifts and keep them technically sharp by leaving a margin, and unload your appetite for failure onto the safe isolation and machine work, ideally at the end of the session where the residual fatigue does not compromise your heavy lifting. Our guide to managing training injuries reinforces why protecting your form on the big lifts is worth the small stimulus you give up.

How to Actually Use Failure and RIR

Pulling the evidence together into a system, here is how a lifter who wants to maximize growth and strength while training sustainably should actually handle failure and reps in reserve.

Set your default at 1 to 3 reps in reserve

The bulk of your working sets should land in the 0 to 3 RIR window, and a sensible default is around 1 to 2 RIR for most work. This captures essentially the full growth stimulus, as the hypertrophy research shows, while keeping fatigue in a range you can recover from and repeat across the week. If you are running higher volumes, bias toward the 2 to 3 RIR end so the accumulated sets do not bury you. If you are running lower volumes, you can push closer to 0 to 1 RIR because you have fewer sets to recover from.

Protect the big lifts, spend failure on the small ones

Keep heavy compound lifts at 1 to 3 RIR to guard technique and manage systemic fatigue. Reserve your true or near-failure efforts for isolation and machine work, placed toward the end of the session. This is the single highest-value rule in this article, because it gives you the failure stimulus where it is cheap and safe and denies it where it is expensive and dangerous.

Use failure as a periodized tool, not a constant

There is a legitimate place for planned failure. Taking the last set of an exercise to failure, or pushing failure more aggressively in the final week or two of a training block before a deload, can add a useful stimulus precisely because you are about to recover anyway. The mistake is making failure your everyday setting. Use it as a spice, applied deliberately to specific sets at specific times, rather than as the base seasoning on everything.

Autoregulate with RIR day to day

Because RIR describes effort rather than a fixed weight, it lets you adjust to how you feel without losing the plot. Assign an RIR target to each exercise and pick the weight that hits it on the day. On a strong day you will use more weight to reach 2 RIR; on a flat day, less. Either way your true training stress stays where you programmed it, which keeps progress steady through the inevitable fluctuations in sleep, stress, and nutrition. Logging your loads and RIR in a training journal turns this from guesswork into a trackable trend.

Common Mistakes

Most of the ways lifters get this wrong fall into a short list of predictable errors.

• Taking every set to failure. The classic overzealous mistake. It feels productive because each set is maximally hard, but it accumulates systemic fatigue that suppresses your volume and quality across the week, likely reducing total growth rather than increasing it. Failure on everything is one of the fastest routes to a stall.
• Grinding heavy compounds to failure. This is where injuries happen. Pushing a squat, deadlift, or heavy press to true concentric failure means the rep fails with a heavy bar and degrading form. Keep a margin on the big lifts and take your failure appetite to the machines.
• Assuming you are at failure when you are not. Especially common in beginners, who routinely leave four or five reps behind while believing they are all out. If your reps are not slowing down noticeably, you are probably further from failure than you think. Calibrate on a safe exercise.
• Never getting close to failure. The opposite error is just as real. Training at 4 or more RIR on everything leaves growth on the table, because you never recruit the high-threshold fibers hard enough for long enough. Effort still matters; the point is that close-to-failure, not failure itself, is the target.
• Ignoring the within-session cost. Taking your first set of an exercise to failure sabotages every set after it. Leave a rep or two on the early sets so you can maintain reps and load across all of them, banking more total quality volume.
• Treating RIR as exact. RIR is a useful estimate, not a precise instrument, and it is fuzzier at higher values and lighter loads. Use it as a guide, calibrate it periodically, and do not agonize over whether a set was truly 2 versus 3 RIR.

Frequently Asked Questions

Do I need to train to failure to build muscle?

No. The evidence shows that stopping one to three reps short of failure produces essentially the same muscle growth as grinding to failure, as long as the set is genuinely challenging. Refalo and colleagues (2023) found only a small hypertrophy advantage for training nearer to failure, and that edge shrinks once volume is matched. What matters is that your last reps are hard and slow, not that the bar physically stalls. Leaving one to three reps in reserve lets you accumulate the same productive volume with far less fatigue.

What does reps in reserve (RIR) mean?

Reps in reserve is your estimate of how many more reps you could have done at the end of a set before failure. Stopping at 2 RIR means you believe you had two reps left. It is the inverse of the RPE scale from Zourdos and colleagues, where an RPE of 10 equals 0 RIR, or true failure. RIR lets you dial in effort precisely on every set instead of relying on the crude choice of failure or not, which is why it has become the standard way to autoregulate resistance training.

Is training to failure bad for you?

Not inherently, but it carries a fatigue cost that compounds. Moran-Navarro and colleagues (2017) found that training to failure lengthened recovery, with strength and fatigue markers taking 24 to 48 hours longer to return to baseline than stopping short. Failure also raises the risk of technical breakdown on heavy compound lifts. Used sparingly on safe exercises it is a good tool. Used on every set of every session it accumulates systemic and joint fatigue that eats into your next workouts and your weekly volume.

Should I train compound and isolation exercises to failure differently?

Yes. Heavy compound lifts like squats, deadlifts, and barbell presses create large systemic fatigue and become dangerous when form collapses near failure, so keep them one to three reps shy of failure most of the time. Single-joint isolation exercises like curls, lateral raises, leg extensions, and cable work are far safer to take to or near failure because a missed rep carries little risk and the systemic cost is low. Leave a rep or two on the big barbell lifts, and save your closest-to-failure efforts for machines and isolation work.

How accurate are people at judging reps in reserve?

Accuracy improves with training experience and is best near failure. Zourdos and colleagues found experienced lifters gauge proximity to failure more accurately than novices, and most people are most precise at low RIR values like 0 to 2, where the reps visibly slow down. Accuracy drops at higher RIR values and lighter, high-rep sets, where lifters commonly underestimate how many reps they have left and stop earlier than they think. Calibrate occasionally by taking a safe exercise to true failure, and remember that most beginners are further from failure than they feel.

How many reps in reserve should I leave on most sets?

For most lifters most of the time, training in the 0 to 3 RIR range on the majority of working sets captures nearly all the growth stimulus while keeping fatigue manageable. A reasonable default is to keep heavy compound lifts around 1 to 3 RIR to protect technique and manage fatigue, and push isolation and machine work to 0 to 1 RIR where it is safe. Beginners should lean toward leaving more in reserve while they learn to gauge effort and groove technique, since they tend to overestimate how hard they are actually working.

The Bottom Line

Training to failure is a tool with a specific, limited role, and the fifty-year war over it was mostly a failure to define terms. The growth stimulus comes from taking sets close to failure, not from reaching failure itself, and the research is now clear that stopping one to three reps short builds the same muscle as grinding to a dead stop. For strength, failure offers no advantage at all, because strength is a skill best trained with clean, sharp, submaximal reps rather than slow grinds.

What failure reliably delivers is fatigue. It lengthens recovery by a day or two, it degrades technique and raises injury risk on heavy compounds, and it quietly suppresses the within-session and weekly volume that actually drives your gains. That makes constant failure training a bad trade: a marginal per-set gain in exchange for a meaningful weekly loss. Reps in reserve is the tool that lets you avoid the trade, dialing effort precisely so you train hard enough to grow without digging a fatigue hole you cannot climb out of.

The whole system fits in a few lines. Keep most working sets at 1 to 3 reps in reserve. Protect your heavy compound lifts by leaving a margin, and spend your near-failure and true-failure efforts on safe isolation and machine work at the end of a session. Use failure as a periodized spice, pushed a little harder before a deload, rather than an everyday default. Calibrate your RIR judgment now and then, because most lifters, and nearly all beginners, are further from failure than they feel. Do that, and you get almost all of the stimulus, almost none of the wreckage, and the ability to keep training hard for the decades it actually takes to build a body.

For the surrounding pieces, see our guides on training volume, progressive overload, and deload weeks -- the variables that decide whether your hard sets add up to long-term progress.

References

1. Refalo, M.C., Helms, E.R., Trexler, E.T., Hamilton, D.L., & Fyfe, J.J. (2023). Influence of resistance training proximity-to-failure on skeletal muscle hypertrophy: a systematic review with meta-analysis. Sports Medicine, 53(3), 649-665.
2. Refalo, M.C., Helms, E.R., Robinson, Z.P., Hamilton, D.L., & Fyfe, J.J. (2024). Similar muscle hypertrophy following eight weeks of resistance training to momentary muscular failure or with repetitions-in-reserve in resistance-trained individuals. Journal of Sports Sciences, 42(1), 85-101.
3. Davies, T., Orr, R., Halaki, M., & Hackett, D. (2016). Effect of training leading to repetition failure on muscular strength: a systematic review and meta-analysis. Sports Medicine, 46(4), 487-502.
4. Grgic, J., Schoenfeld, B.J., Orazem, J., & Sabol, F. (2022). Effects of resistance training performed to repetition failure or non-failure on muscular strength and hypertrophy: a systematic review and meta-analysis. Journal of Sport and Health Science, 11(2), 202-211.
5. Moran-Navarro, R., Perez, C.E., Mora-Rodriguez, R., et al. (2017). Time course of recovery following resistance training leading or not to failure. European Journal of Applied Physiology, 117(12), 2387-2399.
6. Zourdos, M.C., Klemp, A., Dolan, C., et al. (2016). Novel resistance training-specific rating of perceived exertion scale measuring repetitions in reserve. Journal of Strength and Conditioning Research, 30(1), 267-275.
7. Helms, E.R., Cronin, J., Storey, A., & Zourdos, M.C. (2016). Application of the repetitions in reserve-based rating of perceived exertion scale for resistance training. Strength and Conditioning Journal, 38(4), 42-49.
8. Sundstrup, E., Jakobsen, M.D., Andersen, C.H., et al. (2012). Muscle activation strategies during strength training with heavy loading vs. repetitions to failure. Journal of Strength and Conditioning Research, 26(7), 1897-1903.
9. Nobrega, S.R., & Libardi, C.A. (2016). Is resistance training to muscular failure necessary? Frontiers in Physiology, 7, 10.
10. Schoenfeld, B.J., Ogborn, D., & Krieger, J.W. (2017). Dose-response relationship between weekly resistance training volume and increases in muscle mass: a systematic review and meta-analysis. Journal of Sports Sciences, 35(11), 1073-1082.