How to Measure Training Intensity: RPE, RIR, and Percentage of 1RM

Intensity means two different things

When a coach says a set was intense, they could mean either of these, and the two come apart constantly:

• Intensity of effort. How close to failure you took the set, measured in reps in reserve (RIR) or as a rating of perceived exertion (RPE). A set of 20 with light dumbbells taken to the point where you cannot do another one is maximal effort at a relatively light load.
• Intensity of load. How heavy the weight is relative to your one-rep max, written as a percentage of 1RM. A set at 85% of 1RM is heavy whether you did one rep or five.

RIR and RPE measure the first. Percentage of 1RM measures the second. You can run a set at 60% of 1RM to 0 RIR, or at 90% of 1RM with 3 reps left in the tank, and neither combination is a contradiction.

Most programming arguments turn out to be about which of these you are using, not about who is right.

Reps in reserve

RIR is the number of additional reps you believe you could have completed with good technique when you racked the weight. Stopping a set with 2 RIR means you think you had two left.

It is the most practical way to control effort, because it needs no equipment, no testing, and no current 1RM. It is also less precise than it looks.

When researchers pooled a dozen studies on this, people underestimated what they had left by about one rep on average. Individual studies disagreed quite a bit, so treat that as a tendency rather than a fixed correction.

Your guess does get better in specific situations: closer to failure, on shorter sets, and on later sets of an exercise. It does not get better with training experience. That one surprises people, and it is worth knowing if you have been told your RIR calls will sharpen up once you have a few years in.

What follows from that:

• Treat an RIR call as a band, not a number. Your 2 RIR is realistically 1 to 4.
• Trust it more on heavy, short sets than on sets of 20.
• Calibrate occasionally. Take one set to genuine failure now and then and compare the result to what you predicted at the point you would normally have stopped.

RPE and the RIR-based scale

Modern lifting RPE is a 1 to 10 scale where the number is defined by reps in reserve. RPE 10 means 0 reps left, RPE 9 means 1 left, RPE 8 means 2 left and RPE 7 means 3 left. Lower ratings describe easier effort more broadly, rather than a precise one-rep mapping all the way down. It comes from a 2016 study that put numbers on the scale and it has been the common language for effort ever since.

Many lifters use half points, such as RPE 8.5 for "somewhere between one and two left". That is a convention rather than a research finding, and given how imprecise RIR estimates are, do not read too much into the halves.

Watch out for the other RPE. The original Borg scale runs 6 to 20 and rates general exertion, not how close you are to failure. If a program tells you to work at an RPE of 15, that is the one it means, and it is a different instrument.

The same 2016 study found bar speed tracks RPE closely in both novice and experienced lifters: the nearer you get to failure, the slower the bar moves. That relationship is what velocity-based training runs on, and a tracker is a decent second opinion on your own judgment if you have one.

Percentage of 1RM

Percentage of 1RM expresses load as a fraction of the heaviest single rep you can complete. 100 kg for a lifter with a 125 kg squat is 80% of 1RM.

To use it you need a 1RM, which you can test or estimate. Testing costs a session and some fatigue. Estimating costs accuracy: the formulas disagree with each other about the same set, and none of them is a tested max. Lift's one-rep max calculator will estimate one from a set you have already done and print the percentage weights that follow from it.

Load percentage earns its place when the goal is maximal strength. The 2026 ACSM position stand puts that at 80% of 1RM and above, through a full range of motion. For muscle size it matters much less: growth looks similar across a wide span of loads as long as the sets are taken close to failure.

The percentage chart, and why it lets you down

Charts that map reps and effort onto a percentage of 1RM are the most-shared artifact in strength training. Here is one, computed with the Brzycki equation that Lift's one-rep max calculator uses:

1RM = weight / (1.0278 - 0.0278 x reps to failure)

Read a set as the reps you completed plus the reps you left in reserve. Five reps at 2 RIR is a seven-rep-to-failure effort, which lands at 83.3% of 1RM. Lighter loads are green, near-maximal loads are red.

Two things before you tape it to your gym bag. Every cell comes from one formula, so a chart built on a different one gives different numbers for the same set, and none of them is the true answer. And the arithmetic runs on reps to failure, so a cell is only as good as your RIR call, which may be wrong by more or less than the roughly one-rep average reported in research.

The second weakness is drift. Your percentages are calculated from a 1RM that was accurate on the day you measured it. Get stronger and every prescription based on it becomes too light, silently.

RM targets

The third method is to prescribe a weight by the reps it allows: a 5RM is the heaviest weight you can lift for exactly five reps.

This bundles load and effort into one instruction, which is convenient, and it self-corrects as you get stronger in a way that a fixed percentage does not. Put the two head to head for building strength and both work, with a lot of variation between studies and neither coming out clearly ahead.

The catch is that a true RM requires going to or very near failure to verify, which is fatiguing to do often.

How much effort do you actually need?

Enough to make the set hard. Not necessarily failure.

A meta-analysis of failure and non-failure training did not establish a clear hypertrophy advantage for momentary muscular failure. The non-failure conditions varied, so this is not a precise comparison of 0 versus 1 or 2 RIR, and the 2026 ACSM position stand landed in the same place: training to failure did not reliably change outcomes either way.

For hypertrophy-focused work, 1 to 3 RIR is a practical starting target, not a universal requirement. Easier sets can still contribute, particularly to strength practice or fatigue management. If progress stalls, review effort alongside volume and recovery.

Which method to use

• Setting a starting weight on a new exercise. Percentage of 1RM, from a tested or estimated max. Then adjust based on what you actually complete.
• Deciding whether today's weight was right. RIR or RIR-based RPE. Both describe how demanding the set felt today.
• Running a rep range week to week. A rep range with an effort target. Double progression does not require testing a true repetition maximum each session.
• Peaking for a strength test. Percentage of 1RM, because specificity to heavy loads is the point, with RIR as a safety valve on the day.

Combining two of them is normal and usually better than picking one. A common setup is a percentage-derived load with an effort cap: "80% of 1RM for 3 sets of 5, stop any set at 1 RIR". The percentage sets the load, the RIR keeps a bad day from turning into a failed session.

Recording it

Weight and reps alone cannot distinguish a set of 8 that nearly killed you from a set of 8 with three in reserve. Those are different sets, and next session's decision depends on which one happened.

Add an RIR value to each working set. It takes a second and it is the column that tells you whether a flat-looking log is actually flat: identical weight and reps at a higher RIR can indicate improved performance. Lower RIR means the same work felt harder. How to track progressive overload covers the rest of what to record and how long a window to judge it over.

You can log an RPE value alongside the weight and reps on every set in the Lift app, so the effort side of a set is stored rather than remembered. Last session's numbers show up on the set you are about to do, and an estimated one-rep max runs across your whole history, so you can review load trends. Full chart access requires Premium. See how the Lift app handles progressive overload.

Common mistakes

• Treating an old 1RM as current. Percentages calculated from a max you tested four months ago prescribe weights for the lifter you were four months ago.
• Using RPE without ever calibrating it. Compare your ratings with repeated performance and consistent technique. You do not have to take heavy lifts to failure to use the scale.
• Comparing RPE across exercises. An RPE 8 squat and an RPE 8 lateral raise are the same statement about proximity to failure and nothing else.
• Reading a percentage-to-reps chart as a prescription. This chart is a formula-based estimate, not a measured population average. Individual repetition capacity varies.
• Chasing RPE 10 on everything. The evidence for a benefit is trivial, and the fatigue cost is not.

A note on load and safety: heavier percentages are not inherently dangerous, but a load you cannot control through a full range of motion is not measuring what you think it is. Stop a set for sudden or severe pain, and if pain persists, worsens, or changes how you move, reduce the load or range of motion and have it assessed by a qualified healthcare professional. This article is general education, not individual medical or coaching advice.