How Close to Failure Should You Train for Best Results?

Stopping 1–2 reps short of failure produces similar muscle-growth results to training to failure — with less fatigue — for most sets. Training to absolute failure adds meaningful fatigue without a proportional benefit for most lifters, particularly on technically demanding compound movements. The proximity-to-failure research base shows that the exercise you're doing and the load you're using both change this equation, so the practical answer isn't "always go to failure" or "always leave reps in reserve" — it's learning to match your stopping point to the specific demands of each movement.

Key Finding

Training closer to momentary muscular failure consistently increases neuromuscular fatigue, perceptual fatigue, and metabolic stress across both exercises and loads studied. However, the magnitude of that fatigue response is not uniform — moderate loads generate substantially more acute fatigue than heavier loads, and lower-body compound movements are more demanding than upper-body horizontal pulls. Within-session movement quality also tends to decline as sets approach failure.

The takeaway is not that failure training is bad. It's that the cost of failure training is highly context-dependent, and that cost needs to be weighed against the total volume and exercise selection in your program.

Study Details

Note on the evidence: The acute fatigue findings below are drawn from research examining how proximity to failure, load, and exercise selection affect immediate and short-term neuromuscular, perceptual, and metabolic responses. Acute-fatigue studies of this kind do not measure hypertrophy. Where this article discusses long-term muscle growth, those claims are attributed separately to longitudinal work (see the hypertrophy note and citation below).

The general design of this line of acute research is a randomised crossover, in which each participant completes multiple sessions varying two key variables:

Participants return roughly 24 hours after each session for follow-up assessments of neuromuscular function and perceived muscle soreness. The crossover design means each participant experiences all conditions, which reduces the noise introduced by individual differences in training background or recovery capacity.

(If you are citing exact participant numbers, session counts, or specific exercise pairings, verify them directly against the source paper — do not treat unconfirmed specifics as established findings.)

Results

Several findings from this line of research are directly relevant to how lifters should structure their training:

Fatigue scales with proximity to failure. Neuromuscular fatigue, perceptual fatigue, and metabolic stress all increase as sets move closer to momentary muscular failure. This holds across exercises and loads.

Exercise type matters. Lower-body compound movements produce more fatigue than upper-body horizontal pulls at equivalent proximity-to-failure conditions. This aligns with what most experienced lifters observe anecdotally.

Load interacts with fatigue in a counterintuitive way. Moderate loads (~65% 1RM) can generate more acute fatigue than heavy loads (~85% 1RM), likely because moderate loads allow more total repetitions per set when taken to or near failure, accumulating greater metabolic stress.

Movement quality tends to decline near failure. Within-session kinematics degrade as sets approach failure across most conditions, suggesting that proximity to failure is a meaningful variable for maintaining technical integrity — though the degree varies by exercise and load.

More total volume is completed closer to failure. Training to or near failure results in more total repetitions in certain conditions. This is relevant for lifters using volume as a primary driver of adaptation — but it comes with the fatigue trade-off noted above.

Limitations

A few honest caveats are worth noting before applying these findings directly to your program:

  1. Acute data, not long-term adaptation. Acute and 24-hour follow-up assessments tell us about immediate fatigue responses, not long-term adaptation. Whether training consistently closer to failure produces better or worse hypertrophy over 8–16 weeks is a separate question, addressed by longitudinal research (see below).

  2. Limited range of exercises. Findings from a small number of exercises may not generalise directly to movements like the Romanian deadlift, overhead press, or leg curl, which have different fatigue profiles and technical demands.

  3. Trained status matters. The fatigue response to proximity-to-failure training can differ meaningfully between novice and experienced lifters.

What the Hypertrophy Research Actually Shows

Separately from the acute-fatigue data above, a systematic review and meta-analysis by Refalo and colleagues examined the influence of proximity to failure on skeletal muscle hypertrophy across longitudinal training studies. Its broad conclusion: training closer to failure is not clearly superior for muscle growth, and stopping short of failure can produce similar hypertrophic outcomes while accumulating less fatigue. This is the appropriate source for growth-related claims — the acute study is not.

What This Means for Your Training

Taken together, the acute-fatigue and hypertrophy literature support a consistent principle, echoed by the ACSM position stand Progression Models in Resistance Training for Healthy Adults (2009): training to failure is not categorically superior to stopping short, and the fatigue cost of failure training is real and exercise-specific.

Here's how to apply that concretely:

Match proximity-to-failure to exercise complexity and load. For high-fatigue, technically demanding movements like the barbell back squat — especially at moderate loads — stopping at 2–3 RIR is a reasonable default. The fatigue accumulation from pushing squat sets to failure across multiple sets is substantial, and it can degrade movement quality without necessarily delivering more useful volume.

You have more flexibility on rows and upper-body pulls. Horizontal pulling movements, particularly at heavier loads, tend to carry a lower fatigue cost. This gives you more room to push effort on accessory work.

Moderate loads demand more careful proximity management. If you're running a higher-rep, moderate-load block (think 65–70% 1RM), be conservative with proximity to failure. The combination of moderate load and high proximity to failure tends to produce the most acute fatigue — which matters if you train multiple sessions per week.

Use RIR as a tracking variable, not just a feeling. The value of the repetitions-in-reserve framework is that it gives you a consistent reference point across sessions. Tracking your training with specific RIR targets — rather than vague effort ratings — lets you see whether your fatigue management is actually consistent over time. This is exactly the kind of session-level data that Kenso is built to capture: logging your sets, reps, load, and effort (including RPE) so that patterns become visible across weeks, not just within a single session.

Fatigue compounds. One set near failure is manageable. Multiple sets near failure on a compound movement, repeated across sessions per week, accumulate quickly. If your program has you doing 8–10 sets of squats per week across two sessions, the proximity-to-failure decisions you make in session one have direct consequences for session two.

Kenso's rule-based double-progression engine helps with longitudinal load management. Rather than optimising each session in isolation, it uses your logged training history to generate weight and rep recommendations — and to trigger deloads — so intensity decisions reflect where you actually are in a training block, not just where you feel like pushing on a given day.

The Bigger Picture

The question of how close to failure to train has generated a lot of debate, and this research adds useful nuance. The answer isn't a single number that applies universally — it's a variable that should be adjusted based on the exercise, the load, the total volume in your program, and where you are in a training cycle.

For lifters who track their training carefully, this is good news: the data you're already collecting — sets, reps, load, and effort — is exactly what's needed to make these decisions well. For those training on feel alone, it's a reminder that fatigue from proximity-to-failure choices accumulates in ways that aren't always obvious in the moment.

Consistency in how you manage effort across a program is at least as important as any single session's intensity.

Frequently Asked Questions

Is training to failure necessary for muscle growth?

No. Longitudinal research (see the Refalo et al. meta-analysis cited below) suggests that training within roughly 1–2 repetitions of failure produces similar hypertrophic outcomes to training to absolute failure, with lower fatigue accumulation. Training to failure may be useful selectively — for example, on final sets of lower-risk exercises — but is not required across all sets or exercises.

What does "repetitions in reserve" (RIR) mean in practice?

RIR refers to how many more reps you estimate you could complete before reaching muscular failure. A set completed at 2 RIR means you stopped when you believed you had 2 reps remaining. It's an effort-regulation tool that helps lifters apply consistent intensity across sessions without always taking sets to failure.

Does proximity to failure matter more for some exercises than others?

Yes. Compound lower-body movements like the back squat show greater fatigue responses when taken close to failure compared to upper-body horizontal pulls. Technically demanding movements with higher systemic fatigue cost generally warrant more conservative proximity-to-failure targets.

How does load percentage interact with proximity to failure?

Moderate loads (~65% 1RM) can generate more acute fatigue than heavy loads (~85% 1RM) at equivalent proximity-to-failure conditions, likely because moderate loads allow more total repetitions per set. This means proximity-to-failure management is especially important during higher-rep, moderate-load training phases.

How can I track proximity to failure consistently across sessions?

Logging your RIR (or RPE) alongside your sets, reps, and load is the most practical approach. Over time, this data reveals whether your effort targets are consistent or drifting — and whether fatigue is accumulating across a training week. Kenso's iOS training log is designed for exactly this kind of session-level tracking, giving you a structured record that makes effort patterns visible across your full training history.

Citations

Refalo, M. C., Helms, E. R., Hamilton, D. L., & Fyfe, J. J. (2024). Influence of resistance training proximity-to-failure, determined by repetitions-in-reserve, on neuromuscular fatigue in resistance-trained males and females. Journal of Sports Sciences. PubMed — Please confirm the exact DOI/PubMed ID against the published record before publishing; the identifiers in the original draft could not be verified.

American College of Sports Medicine. (2009). Progression Models in Resistance Training for Healthy Adults. Medicine & Science in Sports & Exercise, 41(3), 687–708. PubMed

Note: Verify all citation identifiers against the original published sources before publication. Do not publish DOIs or PubMed IDs that cannot be independently confirmed.

Ready to start tracking your proximity-to-failure targets alongside your sets and reps? Download Kenso on iOS and log your training with intention — session by session, week by week.