Reps in Reserve (RIR) Explained — And Why Your Estimate Is Probably Wrong

Reps in reserve (RIR) in lifting is the number of additional reps you believe you could still perform before hitting true muscular failure on a set. Call a set "2 RIR" and you're saying you had two more reps in the tank. Simple idea. The problem is that most lifters — beginners and veterans alike — are bad at it, and the research says the error usually runs in the opposite direction you'd guess.

I've coached people who swore a set was an all-out grinder when they had four reps left. I've also watched trained lifters stop at "failure" with two clean reps still available. Both are normal. Let's get the definition straight, then I'll show you where your estimate breaks down and what you can actually do about it.

What is reps in reserve (RIR)?

Reps in reserve (RIR) is a self-reported estimate of how many more reps you could complete before reaching momentary muscular failure. It's the backbone of the modern resistance-training effort scale. When you finish a set and could squeeze out three more, that's 3 RIR. Zero reps in reserve means true failure — the next rep doesn't move.

RIR exists because raw load percentages lie. "Use 75% of your 1RM" assumes your 1RM is fixed and your fatigue is fixed. Neither is true day to day. RIR lets you autoregulate: instead of chasing a number on the bar, you chase a target distance from failure. That's why every serious hypertrophy program is now written in RIR or its sibling, RPE.

One thing to kill early. You'll see RIR called a "0-4 scale." That's a simplification, and an incomplete one. The validated tool is a 10-point RPE scale from Zourdos and colleagues (2016), where RIR descriptors map cleanly onto RPE 5 through 10 — that is, 0 to 5 RIR. Beyond about 5 reps from failure, the self-estimates fall apart. Hold that thought, because it matters for how you program.

What does 2 reps in reserve mean?

Two reps in reserve means you stopped a set believing you could have completed exactly two more reps with good form before failure. On the RPE map that's an RPE of 8. It's the most common prescription in hypertrophy training — hard enough to drive growth, far enough back to keep your form, your joints, and your next set intact.

Here's the catch with "2 RIR" specifically. It sits right inside the range where lifters are least reliable. In the Halperin and colleagues (2020) knee-extension experiments, trained participants who stopped at their self-perceived failure actually had reps left — they under-predicted reps to failure by roughly 2.0 reps on average. Read that again. When people feel like they're at 0 RIR, they're often closer to 2 RIR. So if you're targeting "2 RIR" by feel, there's a real chance you're at true failure and don't know it.

How many reps in reserve for hypertrophy?

For hypertrophy, train most working sets in the 1-3 RIR range. Growth is dose-dependent on proximity to failure — the closer you push, the more stimulus per set, up to a point. But you do not have to hit failure to grow maximally, and the fatigue cost of those last reps is real.

The cleanest evidence here is Refalo and colleagues (2024). Eighteen trained adults, eight weeks, leg press and leg extension, one leg to failure and one at 1-2 RIR. Quadriceps growth was essentially identical: +0.181 cm to failure versus +0.182 cm at 1-2 RIR. Same growth. But the failure leg paid more — greater acute neuromuscular fatigue and bigger velocity loss for nothing extra in the muscle.

The Robinson and colleagues (2024) meta-regression across 122 studies adds the nuance most articles miss: hypertrophy scales with proximity to failure, but strength gains are largely independent of it. So the "right" RIR is goal-specific. Building size? The last rep or two carries disproportionate stimulus, so 1-3 RIR earns its keep. Chasing a heavier squat? You can sit at higher RIR — around 4-5 — and keep most of the strength while saving the fatigue. That's a real fork in the road, not a rounding error.

Why lifters are bad at estimating RIR

The short version: your sense of "reps left" is a prediction of an event that hasn't happened yet, made under accumulating fatigue, with a training partner watching. It's noisy by nature. And the error isn't random — it has a direction and a context.

Accuracy is proximity-dependent. The Bastos and colleagues (2024) scoping review found that about a third of studies (10 of 31) showed prediction accuracy decreases the farther you are from failure. Your estimate is decent in the last few reps and gets fuzzy fast above 3 RIR. So the deep-reserve, far-from-failure calls — exactly what light "pump" work asks of you — are the least trustworthy.

Then there's load and body region. RIR estimates are sharper at high loads (80-100% 1RM) and worse on high-rep, lighter sets where metabolic fatigue muddies the signal. Upper-body estimates beat lower-body ones in five studies, probably because your arms feel more than your legs do. And here's the kicker for the "just train more and you'll learn it" crowd: the meta-analytic evidence shows no clear effect of training experience on RIR accuracy. Multi-year lifters still misjudge submaximal sets.

Let me clear up two myths that ruin most RIR discussions:

• "When a set feels like failure, you're at failure." Usually false. Trained lifters who stop at self-perceived 0 RIR average about 2 real reps left. Perceived failure shows up before mechanical failure.
• "Beginners overrate effort, so they're basically at failure already." Backwards for proximity. The documented beginner error is calling a set "2 RIR" when it's really 4-5 — stopping well short. Effort perception and reps-remaining prediction are two different measurements with opposite biases. Conflating them is the single most common analytical mistake in RIR writing.

This is the gap Riven is built for. Riven uses the Apple Watch — motion sensors plus heart rate, no camera, no barbell clip — to measure how much your reps slow down across a set, then converts that velocity decay into a 0-100 failure-proximity score. It's an objective second opinion on a number you're otherwise guessing.

RIR vs RPE: how they relate

RIR and RPE are two readouts of the same thing: how close a set is to failure. On the Zourdos scale they're tied together directly. RPE 10 equals 0 RIR (true failure). RPE 9 is 1 RIR. RPE 8 is 2 RIR, and so on down. RPE asks "how hard was that?"; RIR asks "how many were left?" Same coin, two faces.

I prefer RIR with the people I coach because it's a concrete count, not a vibe. "Two reps left" is checkable. "It felt like a 7" is not. But they share the same weakness: both are subjective, and both degrade the farther you sit from failure. Swapping the label doesn't fix the underlying guesswork.

How to measure reps in reserve without guessing

You measure RIR objectively by tracking rep velocity. As a muscle fatigues, the bar — or your wrist — physically slows down, and that slowdown scales with proximity to failure in a way your feelings don't. This is the part most lifters never get told: there's a measurable fingerprint underneath the guess.

The numbers are clean. Velocity loss across a set runs roughly -25% at failure, about -13% at 1 RIR, and around -8% at 3 RIR on post-set measures. Zourdos showed bar velocity correlates strongly with reported RPE — r ≈ -0.88 in experienced squatters. The bar slowing down is the thing your RIR is trying to perceive. The catch, and I won't pretend otherwise: velocity isn't a magic universal cutoff. Across nearly 3,000 measurements, bar velocity explained only about 30% of perceived-RIR variance (PMC12360324), and the relationship shifts by exercise, load, and set number. Velocity and felt RIR are complementary, not interchangeable. You individualize it or it lies to you.

That's where wrist-based velocity tracking earns its place. A linear position transducer bolted to the bar is the lab standard, but it costs hundreds and lives in a rack. Riven reads velocity loss from the Apple Watch you already own and flags the moment your subjective RIR and the velocity curve disagree — for example, when you call a set "2 left" but your rep speed already dropped like you hit failure. Honest caveat: the wrist signal is a proxy. It reads roughly half the magnitude of a $300 transducer at the same fatigue, so it's not lab-grade and I won't tell you it is. But for the real choice — guessing versus an objective on-wrist read of effort — it's not close. It beats guessing, and guessing is what almost everyone is doing.

Practically, here's how I'd build the skill:

1. Calibrate near failure first. Take a few sets to true 0 RIR so you learn what the real edge feels like. Your "2 RIR" recalibrates fast once you've met actual failure.
2. Trust your estimate most on heavy sets. High load, low reps — that's where your read is sharpest. Lean on feel there.
3. Distrust it on light, high-rep work. Deep-reserve, lighter sets are where self-estimates collapse. Use an objective check or just push closer than feels necessary.
4. Watch for the under-prediction drift. If your "2 RIR" program keeps leaving you wrecked, you're probably training to failure without knowing it.

FAQ

What is reps in reserve in lifting?

It's how many more reps you believe you could do before failure on a given set. 3 RIR means three reps left; 0 RIR is true failure. It's a way to gauge and control set intensity without relying on rigid load percentages.

What does 2 reps in reserve mean?

You stopped believing you had two clean reps left before failure — RPE 8 on the standard scale. It's the staple hypertrophy intensity, though it sits in the range where lifters most often under-predict, so by feel you may actually be closer to failure than you think.

How many reps in reserve should I use for hypertrophy?

Most working sets at 1-3 RIR. Growth scales with proximity to failure, but 1-2 RIR matches to-failure growth with less fatigue, so you rarely need to hit 0 RIR on everything.

Is RIR the same as RPE?

They're directly linked. RPE 10 = 0 RIR, RPE 9 = 1 RIR, RPE 8 = 2 RIR. RPE rates effort; RIR counts reps left. Same scale, two readouts — and both degrade the farther you are from failure.

Can you measure RIR objectively?

Partly. Rep velocity drops as you approach failure (about -25% loss at failure, -13% at 1 RIR), so velocity tracking gives an objective proxy. It's individual and context-dependent, but tools like Riven use it to give a second opinion when your felt RIR and your rep speed disagree.

Sources

• Zourdos et al. (2016), Novel Resistance Training-Specific RPE Scale Measuring Repetitions in Reserve, J Strength Cond Res — https://pubmed.ncbi.nlm.nih.gov/26049792/
• Halperin et al. (2020), Just One More Rep! Ability to Predict Proximity to Task Failure, Frontiers in Psychology — https://pmc.ncbi.nlm.nih.gov/articles/PMC7785525/
• Bastos, Machado & Teixeira (2024), Feasibility and Usefulness of Repetitions-In-Reserve Scales: A Scoping Review, Perceptual and Motor Skills — https://pmc.ncbi.nlm.nih.gov/articles/PMC11127506/
• Refalo et al. (2024), Similar muscle hypertrophy following failure or repetitions-in-reserve training, Journal of Sports Sciences — https://pubmed.ncbi.nlm.nih.gov/38393985/
• Exercise type, training load, velocity loss threshold, and sets affect the relationship between lifting velocity and perceived RIR (2024) — https://pmc.ncbi.nlm.nih.gov/articles/PMC12360324/