Should You Train to Failure? What the Evidence Actually Says
Proximity to Failure, RIR, and When Grinding Out That Last Rep Actually Costs You
In This Article
The short answer: Training to failure is not required for muscle growth and is often counterproductive, especially for beginners and compound movements. Research shows that stopping 1-3 reps short of failure (called RIR, or Reps in Reserve) produces similar hypertrophy outcomes with significantly less fatigue and injury risk. Failure has a place, but it is a tool, not a default.
- What Failure Actually Is
- What the Research Says
- Compound vs. Isolation
- When Failure Helps
- Beginners and Effort
- The Recovery Cost
- Using Protocol Data
- FAQ
- Key Takeaways
Read key takeaways →
What "Training to Failure" Actually Means
The phrase "training to failure" sounds simple, but researchers and coaches use it to mean three distinct things. Conflating them produces confused training advice and worse outcomes.
Most gym conversations about "going to failure" describe a mix of all three, which makes interpreting training advice nearly impossible. Research studies on failure training typically enforce momentary muscular failure with direct supervision. Unsupervised gym training almost never reaches this threshold on compound movements.
The RIR Framework
Reps in Reserve (RIR) is the research-backed way to quantify proximity to failure without actually reaching it. An RIR of 2 means you stopped 2 reps before you would have reached technical failure.
- →0 RIR: Technical failure. Could not complete another clean rep.
- →1 RIR: One rep left. Very close to limit. High fatigue cost.
- →2 RIR: Two reps left. Challenging but recoverable. Research sweet spot.
- →3 RIR: Three reps left. Moderate effort. Appropriate for volume days.
- →4+ RIR: Warm-up territory. Not sufficient for hypertrophy stimulus.
The RPE (Rate of Perceived Exertion) scale maps directly to RIR. An RPE of 10 is failure. RPE 8 is 2 RIR. RPE 7 is 3 RIR. These two systems describe the same thing from different angles and are used interchangeably in the literature.
What the Research Actually Says
The failure vs. non-failure debate has been studied directly, and the results are less dramatic than gym culture suggests.
Brad Schoenfeld at CUNY Lehman College has published extensively on proximity to failure and hypertrophy. His work, along with a 2019 meta-analysis by James Krieger, consistently shows that sets taken close to failure (1-3 RIR) produce hypertrophy outcomes comparable to sets taken to true failure, especially when total training volume is matched. The hypertrophy stimulus appears to come from sufficient mechanical tension on the muscle, and that threshold is reached several reps before absolute failure.
Key Studies on Failure vs. Non-Failure
Lasevicius et al. (2019)
Both failure and non-failure training produced similar hypertrophy when volume was equated. The failure group showed no additional muscle growth despite significantly higher perceived effort.
Santanielo et al. (2020)
In trained men, training to failure and stopping 2 RIR short produced equivalent increases in muscle thickness and strength over 8 weeks. Failure training produced greater muscle soreness and longer recovery times.
Schoenfeld and Grgic (2019)
Comprehensive review concluding that training near but not to failure is sufficient to maximize hypertrophic adaptations across most conditions. Failure provides marginal benefit at a disproportionate fatigue cost.
Krieger meta-analysis
Higher volume (more total sets) predicts greater muscle growth. Because failure training reduces how many quality sets you can recover from, non-failure training often produces higher total weekly volume and better long-term hypertrophy outcomes.
Common Misconception
"If you are not going to failure, you are leaving gains on the table." This is one of the most persistent myths in gym culture. The research does not support it. Muscle hypertrophy is driven by mechanical tension and metabolic stress, both of which are adequately stimulated at 1-3 RIR. Training to failure on every set adds fatigue and recovery cost without proportionally increasing the hypertrophy signal. The gains are not in the last grinding rep; they are in the quality sets you can recover from and repeat week after week.
The Bottom Line
Train most sets to 1-3 reps short of failure. Invest the saved recovery into more total sets.
More high-quality sets at 2 RIR produces better long-term muscle growth than fewer failure sets with worse recovery. Reserve actual failure for isolation exercises on your last set of the session, not as a default for every set.
Chris Beardsley, applied hypertrophy researcher
For the complete strength training framework that incorporates these intensity principles, see the Strength Protocol.
Why Failure Is Riskier on Compound Movements
Not all exercises carry the same risk when taken to failure. The distinction between compound and isolation movements matters enormously, and it changes the risk-benefit calculation of pushing to the limit.
Compound movements
Avoid failure. Squats, deadlifts, barbell bench press, and overhead press involve multiple joints and require bracing, balance, and technical precision that degrades under maximal fatigue. A failed squat is a fall. A failed deadlift often means a rounding lower back. The injury risk is not theoretical.
Isolation movements
Failure is acceptable. Dumbbell curls, lateral raises, leg extensions, and cable flyes involve single joints and are typically performed in positions where failure just means putting the weight down. The injury risk is low and failure training is both safer and more practical here.
There is a second problem specific to compound exercises that is rarely discussed: systemic cardiovascular fatigue often arrives before the target muscle reaches true failure. On a set of squats, your lungs and cardiovascular system may give out before your quads actually reach momentary muscular failure. You stop because you cannot breathe, not because the muscle is exhausted. This means the "failure" is not even producing the intended stimulus on the target tissue.
Practical Guideline
A useful working rule based on the evidence:
- →Compound lifts (squat, deadlift, bench, row, press): stop at 2-3 RIR on all working sets.
- →Isolation movements (curls, raises, extensions): 1-2 RIR on most sets, failure acceptable on final sets.
- →Machine exercises with fixed movement paths: failure is generally safe and occasionally useful on final sets.
This is not about being conservative with effort. It is about directing maximal effort toward the sets and exercises where it produces the most stimulus with the least risk. Grinding out a 20-rep squat to true failure produces systemic exhaustion, spinal compression under fatigue, and a days-long recovery requirement. Isolating the quads with leg press to failure produces a localized training effect with a fraction of the systemic cost.
When Failure IS Useful
Failure is not always wrong. There are specific contexts where pushing to the limit produces a meaningful benefit that justifies the added cost.
Final set of an isolation exercise
Taking the last set of curls, lateral raises, or a cable movement to failure adds a high-intensity stimulus with low recovery cost. Since the exercise is single-joint and safe to fail, the risk is minimal and the added intensity signals the muscle to adapt.
Advanced lifters hitting a plateau
When an experienced lifter has used every other progression variable and adaptation has stalled for several weeks, strategic failure training on isolation exercises can break through a plateau. This is a deliberate short-term intervention, not a default approach.
Testing true 1RM capacity
Establishing a true one-rep max for a lift is, by definition, going to failure on a single rep. This is a testing context, not a training context, and should happen infrequently. The information has value for programming but the set itself should not be replicated weekly.
Low-volume high-intensity methods
Training systems like rest-pause, drop sets, and mechanical drop sets often involve reaching failure as a feature of the method. These are legitimate tools for creating metabolic stress with fewer total sets. They carry higher per-session fatigue but can fit into a periodized plan.
The pattern across all legitimate failure use-cases: it is a deliberate, specific, infrequent tool applied to appropriate exercises. Not a philosophy. Not every set. Not every session.
How Beginners Should Think About Effort
Beginners face a specific challenge: they are not yet good at gauging their own proximity to failure. Research consistently shows that untrained individuals underestimate how many reps they have left, partly because the discomfort of hard training is unfamiliar and partly because motor patterns are still being established. A beginner who thinks they are at 1 RIR is often at 4 or 5 RIR.
This has two important implications. First, true failure in beginners is often reached accidentally through poor form before the muscle is genuinely exhausted, which means the injury risk of attempting failure is higher. Second, beginners respond so readily to training stimulus that near-failure effort is not required; significant adaptation occurs even at moderate intensities because the neuromuscular system is adapting as much as the muscle tissue.
What RPE 8 Actually Feels Like
For beginners calibrating effort, RPE 8 (approximately 2 RIR) should feel like:
- →The last completed rep was hard. You had to push.
- →You could have done 1-2 more reps, but they would have been a real grind.
- →Your form held. You were not fighting to keep the bar in the right groove.
- →You could repeat this set after 2-3 minutes of rest.
- →Your breathing was elevated but not completely maxed out.
The goal for beginners is to develop consistency, technique, and the ability to gauge effort accurately before worrying about optimizing failure protocols. A beginner who trains 3 days per week at RPE 7-8 with good form and consistent progressive overload will outperform a beginner chasing failure on every set, not just because the adaptation is similar but because they stay injury-free and keep training.
For a practical framework on how to track whether that training is actually producing results over time, see How to Track Progressive Overload in Your Training.
The Recovery Cost of Failure Training
The most underappreciated argument against routine failure training is not about gains, it is about recovery. The additional fatigue from taking sets to failure is disproportionate to the additional hypertrophy stimulus produced.
Failure Training: Costs vs. Benefits
DOMS (delayed onset muscle soreness) is particularly relevant here. Santanielo et al. (2020) found that failure training produced significantly greater muscle soreness 24-48 hours post-session without a corresponding increase in muscle thickness gains at 8 weeks. More soreness, same hypertrophy. The extra pain is not a signal of superior training; it is a signal of greater damage that requires more time to repair before the next stimulus can be applied.
The compounding effect matters over a training lifetime. An athlete who trains at 2 RIR can often add an extra session per week, perform better in each session, and sustain higher volume over months and years. Cumulative volume is one of the strongest predictors of long-term muscle gain. Anything that reduces sustainable weekly volume, including excessive failure training, works against the long game.
The Long-Game Principle
A muscle cannot grow while it is recovering from excessive damage. If failure training extends recovery from 48 hours to 72 or 96 hours, you are losing a training opportunity that compounds over months. Over a year, 10-15% more training sessions at equivalent quality produces more muscle than identical sessions with higher failure frequency.
To see whether your training is recovering normally, check How to Tell If Your Training Is Actually Working for the specific wearable signals that distinguish adaptation from accumulated fatigue.
Using Your Recovery Data to Know When to Push
The failure vs. non-failure question is not just about what to do on a given set; it is also about when to increase effort on any given day. Your wearable data gives you a direct proxy for whether pushing harder will produce adaptation or just add fatigue to a system that cannot absorb it.
HRV at or above baseline
Push hard. Your nervous system is recovered and the training stimulus will be absorbed. This is the session to approach your RIR ceiling, add a set, or push isolation work closer to failure. The adaptation return is highest here.
HRV 5-10% below baseline
Train as planned. Execute your session normally but do not seek extra sets or extra intensity. Stay at your planned RIR target and do not attempt PRs. Recovery is slightly impaired; the adaptation signal can still be delivered without adding unnecessary fatigue.
HRV 10-15% below baseline
Reduce intensity. Add a rep to your RIR estimates. If you planned to train at 2 RIR, target 3-4 RIR today. Consider swapping a heavy compound session for lighter isolation work. The body is signaling that full recovery stimulus will exceed absorption capacity.
HRV 15%+ below + elevated RHR
Active recovery only. This pattern suggests significant physiological stress, whether from accumulated training, illness onset, or severe sleep debt. Any intensity approaching failure will produce fatigue accumulation without adaptation. A walk, Zone 2 session, or full rest day returns more gains than grinding through.
The key insight is that "when to push" is not a fixed weekly schedule; it is a dynamic decision informed by your recovery state. An athlete who consistently times their hardest sessions to their highest-HRV days, and backs off when recovery is suppressed, produces better long-term results than one who follows a fixed intensity prescription regardless of readiness.
Protocol
Your data tells you when to push
Protocol surfaces your HRV, recovery score, and training readiness every morning. Know exactly when to train hard and when to hold back.
Frequently Asked Questions
Is training to failure ever required for hypertrophy?
No. The research is clear: sets taken to 1-3 RIR (Reps in Reserve) produce hypertrophy outcomes that are statistically equivalent to sets taken to true failure when total training volume is equated. Lasevicius et al. (2019) and Santanielo et al. (2020) both found no significant difference in muscle thickness gains between failure and non-failure groups at matched volumes. The stimulus threshold for hypertrophy appears to be reached well before the final grinding rep.
How do I know what 2 RIR actually feels like?
At 2 RIR, the last completed rep should feel genuinely hard, form should still be intact, and you should be confident you could do 1-2 more reps if you had to, but it would be a real struggle. Beginners typically overestimate how close they are to failure; research by Zourdos et al. shows trained individuals are significantly more accurate at estimating RIR than untrained ones. Beginners should add 1-2 reps to their intuitive estimate until they develop calibration over several months of consistent training.
What about forced reps and spotting? Are they useful?
Forced reps, where a spotter assists you past failure for additional partial reps, produce a genuine additional stimulus and have been studied as an intensity technique. They carry the same recovery cost problem as failure training, amplified. For most lifters, the additional hypertrophy signal from forced reps is modest and the recovery cost is high. They are a legitimate advanced technique but should be used sparingly, perhaps once per mesocycle on a priority muscle group, not as a regular practice.
Should I train to failure if my goal is strength rather than hypertrophy?
Even less so. Strength training prioritizes neural efficiency, motor pattern precision, and explosive force production. Failure training on low-rep strength sets produces significant neural fatigue and form breakdown, both of which directly undermine strength adaptation. Powerlifting programs like 5/3/1 and Sheiko explicitly keep most sets well short of failure. Strength athletes typically train at RPE 7-8 on main lifts and reserve RPE 9 for test days, not regular training days. See the Strength Protocol for the full intensity framework.
Does training to failure get more important as you get more advanced?
Marginally. Advanced lifters have a higher baseline training tolerance, recover faster from high-intensity stimuli, and have better technique that reduces injury risk on hard sets. Chris Beardsley and others argue that advanced athletes may need higher relative proximity to failure to continue stimulating adaptation as the nervous system becomes more efficient. Even so, most elite bodybuilders and strength athletes keep the majority of their sets at 2-3 RIR and reserve failure or near-failure for isolation exercises, late in a mesocycle, or specific intensity techniques. Failure becomes a more viable tool, not a required default.
My HRV is low but I feel fine. Should I still hold back?
This is a common scenario. HRV reflects autonomic nervous system state, which is not always consciously detectable. Feeling fine does not mean the body is optimally primed for adaptation. Research on overreaching consistently shows that subjective readiness lags behind physiological readiness; athletes often feel acceptable 24 hours before a significant performance drop. Using HRV as a leading indicator and adjusting intensity proactively, rather than waiting for felt fatigue to confirm the problem, is one of the most practical applications of wearable data for training. See How to Tell If Your Training Is Actually Working for how to read these signals together.
What to Remember
- →Training to failure is not required for hypertrophy. Sets taken to 1-3 RIR produce statistically equivalent muscle growth when total volume is matched, confirmed across multiple direct comparison studies.
- →Failure training carries a disproportionate recovery cost: greater DOMS, more neural fatigue, and longer recovery windows, without proportional gains in muscle growth.
- →Compound movements (squat, deadlift, bench) should almost never be taken to true failure. Systemic fatigue arrives before muscular failure, and the injury risk is significant.
- →Isolation exercises are the appropriate place for failure: single-joint movements in safe positions on final sets only, not every set of every exercise.
- →Beginners are poor at estimating their own proximity to failure and respond so readily to training stimulus that near-failure effort is not needed. Consistency and technique matter far more.
- →HRV and recovery score are practical proxies for when to push harder. Training at 0-1 RIR on high-HRV days produces superior adaptation compared to grinding to failure when recovery is suppressed.
Related on Protocol
The Strength Protocol
The complete framework for progressive overload, rep ranges, and training intensity.
How to Track Progressive Overload
How to measure whether your training is actually progressing over time.
How to Spot High Cortisol in Your Data
How to tell when to push and when to back off using your wearable data.
Protocol
Your data tells you when to push
Protocol surfaces your HRV, recovery score, and training readiness every morning. Know exactly when to train hard and when to hold back.
See your dashboardReferences
Key Researchers
- Brad Schoenfeld, CUNY Lehman College Leading researcher on proximity to failure and hypertrophy. His reviews have established the scientific consensus that 1-3 RIR produces equivalent muscle growth to failure training at matched volumes.
- James Krieger Meta-analyst focused on volume and failure. His work showed that weekly training volume is a primary predictor of hypertrophy, and that failure training often limits sustainable weekly volume.
- Chris Beardsley Applied researcher and writer synthesizing hypertrophy literature. Advocates for proximity to failure as the core stimulus variable, with failure reserved for specific contexts in advanced training.
Key Studies
- Lasevicius T et al. (2019). Effects of different intensities of resistance training with equated volume load on muscle strength and hypertrophy. European Journal of Sport Science. Found equivalent hypertrophy between failure and non-failure groups when volume was matched. One of the most cited direct comparisons.
- Santanielo N et al. (2020). Effect of resistance training to muscle failure vs. non-failure on strength, hypertrophy and muscle architecture in trained individuals. Biology of Sport. 8-week RCT in trained men showing no significant difference in muscle thickness gains despite greater soreness in the failure group.
- Schoenfeld BJ, Grgic J. (2019). Does Training to Failure Maximize Muscle Hypertrophy? Strength and Conditioning Journal. Comprehensive review concluding that non-failure training to within a few reps of failure is sufficient for most hypertrophic adaptations.
- Zourdos MC et al. (2016). Novel resistance training-specific RPE scale measuring repetitions in reserve. Journal of Strength and Conditioning Research. Established the RIR-based RPE scale as a valid and reliable tool for quantifying proximity to failure during training.
