How Close to Failure Should You Train? Reps in Reserve and Muscle Growth
Ask ten lifters how hard they push each set and you will get ten different answers. Some grind every set until the bar stops moving. Others stop the moment a rep feels heavy. The honest truth is that most people have no idea how close to failure they actually train, and that gap is one of the biggest hidden reasons progress stalls.
Proximity to failure, usually measured as reps in reserve (RIR), has quietly become one of the most important variables in evidence-based training. It often matters more than your exact rep range and more than whether you load 70 percent or 80 percent of your one-rep max. This guide breaks down what the research says, where the line of diminishing returns sits, and how to keep effort honest from week to week.
What "Reps in Reserve" Actually Means
Reps in reserve is simply the number of additional reps you could have performed before reaching technical failure, the point where you cannot complete another rep with good form. Finish a set knowing you had two more in the tank and that is 2 RIR. Grind to a complete stop and that is 0 RIR, or true failure.
RIR is the practical cousin of RPE (rate of perceived exertion). On the common lifting scale, an RPE of 10 means 0 RIR, RPE 9 means 1 RIR, RPE 8 means 2 RIR, and so on. The two systems describe the same thing from opposite directions: RPE asks how hard the set felt, RIR asks how many reps you left behind. If the scale is new to you, our RPE explainer walks through it in detail.
The reason this matters: effort is the signal your muscles respond to. A set carried close to failure recruits the high-threshold motor units and exposes the most muscle fibers to meaningful mechanical tension. A set stopped early, by contrast, may leave a chunk of your muscle barely challenged.
Why Proximity to Failure Drives Growth
Mechanical tension on individual muscle fibers is the primary driver of hypertrophy, and proximity to failure is how you guarantee that tension reaches enough fibers. Early in a set, your body uses its easily fatigued lower-threshold fibers. Only as the set gets genuinely hard do the larger, growth-prone fibers get pulled in to keep the weight moving. Stop too soon and those fibers never get their stimulus.
This is also why load turns out to be less important than lifters once believed. A 2021 meta-analysis found that low-load training (lighter weights, higher reps) and high-load training produced similar hypertrophy when both were taken to a similar high level of effort (PMID: 34822137). The weight on the bar was almost interchangeable. What was not interchangeable was how hard each set was pushed. Our hypertrophy rep range guide covers this load-versus-reps question in more depth.
How Many Reps in Reserve Is Optimal?
Here is where the nuance lives. A 2022 systematic review and meta-analysis examining resistance training to failure versus stopping short concluded that training closer to failure tends to favor hypertrophy, but that you do not need to hit true failure on every set to maximize growth (PMID: 35819335). Sets taken to roughly 1 to 3 reps in reserve captured the large majority of the available stimulus.
The practical sweet spot for most working sets looks like this:
Hypertrophy: 0 to 3 RIR on most sets. Aiming for 1 to 2 RIR is a reliable default that drives growth without burying you in fatigue.
Strength: Often 1 to 4 RIR on your heaviest compounds, since hitting true failure on a max-effort squat or deadlift is risky and costly to recover from.
The cost of failure: Going to 0 RIR on every set generates disproportionate fatigue, lengthens recovery, and can quietly reduce the total quality volume you accumulate across a session.
In other words, the relationship is not "more failure equals more growth forever." It is closer to a curve that rewards getting genuinely close to failure, then flattens, then starts to cost you once excess fatigue eats into your next sets and sessions.
The Most Common Mistake: Training Further From Failure Than You Think
The biggest practical problem is not that lifters chase failure too aggressively. It is the opposite. Multiple studies asking trained lifters to stop at a self-selected "hard" point find that they routinely leave four, five, or more reps in the tank while believing they were close to failure. They simply stop when a set starts to feel uncomfortable.
This calibration error is why two people can run the same program and get wildly different results. If you genuinely train at 1 to 2 RIR and your training partner thinks they do but is actually at 5 RIR, you are getting a meaningfully larger stimulus per set even though the spreadsheet looks identical.
The fix is to occasionally take a set to true failure on a safe machine or isolation movement, just to recalibrate what 0 RIR actually feels like. Once you have felt the real ceiling, your estimates of 1, 2, and 3 RIR become far more accurate.
What the Latest Discussion Adds
This topic has stayed front and center in the evidence-based community. In a recent roundup of new muscle-building studies, Menno Henselmans reinforced that effort proximity remains one of the most robust predictors of hypertrophy across rep ranges, while Brad Schoenfeld has been highlighting fresh work showing that intensity of effort, rather than the absolute load, appears to be the key signal even for tissues like tendons. The throughline is consistent: how hard you push the set is doing more of the work than the specific weight you chose.
That reframes a long-running debate. The old Mike Mentzer style argument was about whether one all-out set to failure beats multiple submaximal sets. The modern read is more measured: you want most sets close to failure, but spread across enough total volume that you are not relying on a single brutal set to do everything. We break that specific debate down in our piece on one set to failure versus multiple sets.
Reps in Reserve and Progressive Overload
Proximity to failure and progressive overload are two sides of the same coin. RIR tells you how hard a given set is relative to your current capacity. Progressive overload is what happens when that capacity grows and you add load or reps to keep the effort meaningful.
Picture it across a few weeks. In week one you bench 185 for 8 reps at 2 RIR. As you get stronger, 185 for 8 starts feeling like 4 RIR, too easy to drive much growth. That drop in effort is your cue to add weight or reps and pull RIR back down into the productive zone. Tracking both numbers together is how you make sure your "hard" sets stay hard. Our progressive overload guide shows how to structure those jumps over time.
How to Apply This in Setgraph
The whole RIR system collapses without an accurate memory of what you actually did, and that is exactly what Setgraph's set logging is built for.
Log RIR or RPE in your set notes. When you record a set, add the effort level in the Set note field (for example "8 reps, 2 RIR"). Now your history shows not just the weight and reps but how hard each set truly was, which is the missing context most paper logs throw away. The workout log keeps that note attached to the specific set forever.
Use pre-filled set history to keep effort honest. Because the record screen pre-fills your most recent set, you open an exercise and instantly see "last time: 185 x 8 at 2 RIR." That single glance tells you whether to repeat, add reps, or add load to keep this set in the 0 to 3 RIR window rather than guessing.
Track the trend in Analytics. Per-exercise charts show your weight, reps, and volume climbing over scrollable time ranges. If your loads are stalling while your logged RIR keeps drifting upward, that is a clear signal your sets have gone soft and it is time to push closer to failure again.
You do not need to overhaul anything. Add one short effort note per working set, glance at your pre-filled history before each set, and let the charts confirm you are still progressing.
FAQ
Q: Should beginners train to failure?
Generally no, not often. Beginners benefit hugely from leaving 2 to 4 reps in reserve while they learn technique, because form tends to break down badly near failure and the injury risk rises. Newer lifters also grow well from submaximal effort, so there is little reason to take the extra risk early on (PMID: 35819335).
Q: Is training to failure ever useful?
Yes. Occasional sets to true failure help you recalibrate what 0 RIR feels like, and failure is generally safer and more practical on machines and isolation exercises than on heavy free-weight compounds. Many lifters save their last set of an isolation movement for an all-out effort.
Q: What is the difference between RIR and RPE?
They measure the same effort from opposite ends. RIR counts the reps you had left; RPE rates how hard the set felt on a 1 to 10 scale. RPE 8 equals 2 RIR, RPE 9 equals 1 RIR, and RPE 10 equals 0 RIR. Use whichever feels more intuitive to you.
Q: How do I know if I am actually close to failure?
Most lifters overestimate how close they are. The best test is to take an occasional set on a safe machine all the way to the point where you cannot complete another rep. That true reference point makes your everyday RIR estimates far more accurate.
Q: Does training closer to failure mean I should do fewer sets?
Often, yes. Pushing every set to 0 RIR creates a lot of fatigue, so you may need slightly less volume to recover. Many lifters get the best results from keeping most sets at 1 to 2 RIR, which lets them accumulate more quality sets without wrecking recovery.
Effort is the variable hiding in plain sight. Dial in your proximity to failure, log it honestly, and let progressive overload keep those sets meaningful week after week. Start tracking your sets, reps, and effort in one place at setgraph.app.
