What Is Estimated 1RM and How Should You Use It in Training?

A set of 80kg for 8 reps tells you more than what you lifted that day. It gives you an approximation of your true 1RM, and when that number is tracked after every session, it becomes a continuous trendline of strength development that a one-time max test cannot provide.

Estimated 1RM is calculated from a submaximal set: the load you lifted and how many reps you completed. You don’t need to lift your actual maximum to get a useful estimate.

The Formulas

Two formulas dominate the literature on estimating 1RM from submaximal performance.

The Epley formula, developed by Boyd Epley and published in his 1985 Poundage Chart, calculates: Estimated 1RM = weight × (1 + reps/30). So 80kg for 8 reps gives: 80 × (1 + 8/30) = 80 × 1.267 = 101.3kg.

The Brzycki formula, published by Matt Brzycki in the Journal of Physical Education, Recreation & Dance (1993), calculates: Estimated 1RM = weight / (1.0278 − 0.0278 × reps). The same 80kg for 8 reps: 80 / (1.0278 − 0.0278 × 8) = 80 / (1.0278 − 0.2224) = 80 / 0.8054 = 99.3kg.

The two formulas produce similar results in the rep ranges where they’re most accurate (3–6 reps) and diverge meaningfully as rep count rises. At 10+ reps, both formulas become unreliable predictors. The physiology behind this: at higher rep counts, the limiting factor in a set is increasingly metabolic (fatigue, cardiovascular demand) rather than pure maximal strength. A person who can squat 100kg for a true 1RM and a person who can squat 100kg for a true 1RM but has better aerobic fitness may produce very different results on a 15-rep estimate, even though their maximal strength is identical.

For the most accurate estimates, keep rep counts between 3 and 6. Sets of 8–10 are usable but less precise. Sets above 10 reps should not be used to estimate 1RM.

How Accurate Is an Estimated 1RM?

Reasonably accurate within a specific range, less accurate at the extremes. Validation studies have found that Epley and Brzycki estimates within the 3–6 rep range predict actual 1RM within roughly 5% for most lifts in most individuals.

Accuracy is better for compound movements (squat, bench press, deadlift) than for isolation exercises (curls, lateral raises). It’s also better for lifters who train in moderate rep ranges regularly than for those who exclusively train very heavy or very light. A powerlifter who rarely exceeds 5 reps may find their 8-rep estimates are less reliable simply because their muscles aren’t conditioned to express their maximum strength output at that rep range.

Individual factors also affect accuracy: limb length, muscle fiber type distribution, and training history all influence how rep count relates to percentage of maximum strength. Two lifters with the same estimated 1RM may have quite different actual 1RMs.

The practical takeaway: don’t treat your estimated 1RM as a precise fact. Treat it as a consistent estimate with known uncertainty. What matters for training purposes isn’t the absolute number. It’s the trend.

Tested 1RM vs Estimated 1RM

Testing your actual 1RM means performing progressively heavier singles until you reach your maximum. The result is more precise for that moment, but it comes at a cost: significant fatigue (both neural and physical), some injury risk with maximal loads, and meaningful disruption to your training schedule. For most recreational lifters, testing actual 1RM once per training cycle (4–12 weeks) is the maximum practical frequency.

An estimated 1RM, by contrast, gets recalculated after every working set. It’s less precise per data point, but it creates a continuous trendline rather than occasional discrete measurements.

The analogy from statistics is apt: a weekly average built from daily measurements is more statistically informative than a single measurement taken every two months, even if each daily measurement carries more noise. The trend is clearer, and deviations from the trend are visible faster.

For tracking strength progress over time, a continuously updated estimated 1RM is more useful in most situations than periodic actual 1RM tests.

Using Estimated 1RM to Set Training Percentages

The primary practical application of estimated 1RM is programming training loads as a percentage of maximum. A training block that prescribes “3 sets at 75% of 1RM” is only executable if you know what your 1RM is.

Common load-percentage guidelines based on the research:

%1RM	Approximate Rep Range	Training Goal
90–100%	1–3 reps	Maximal strength, peaking
80–90%	3–6 reps	Strength, strength-hypertrophy
70–80%	6–12 reps	Hypertrophy
60–70%	12–20 reps	Hypertrophy, muscular endurance
Below 60%	20+ reps	Muscular endurance, technique

These ranges interact directly with training volume and intensity programming. For a deeper look at how load percentages fit into the broader volume vs intensity balance, see the training volume vs intensity guide.

With an estimated 1RM updated from each session, your load percentages stay current without periodic max testing. If your bench press estimated 1RM improves from 100kg to 105kg across a 6-week block, your 75% training load automatically adjusts from 75kg to 78.75kg, maintaining the intended stimulus relative to your current capacity.

Reading the Trend: Flat, Rising, Dropping

A single estimated 1RM data point tells you where you are now. The trend across multiple sessions tells you whether your programme is working.

Rising estimated 1RM: The programme is producing strength gains. The rate of rise matters too. Beginners should see meaningful increases (1–2% per week) in the early months. Intermediates might see 0.5–1% per week. Advanced lifters measure progress over mesocycles, not weeks. A slowly rising trendline is still rising. That’s progress.

Flat estimated 1RM across 6–8 weeks despite consistent training: The stimulus may have stopped being sufficient. Time to examine volume (add sets), intensity (increase load), or variation (change exercise angle or implement). Before doing any of that, check whether rest periods have shortened, whether sessions have been completed as programmed, and whether non-training factors (sleep, stress, calories) are compromising recovery. The framework for diagnosing what’s actually wrong when a program stalls is covered in how to know if your workout program is working. The guide to breaking through training plateaus covers this diagnostic process in detail.

Dropping estimated 1RM: Accumulated fatigue is masking underlying strength. The same weight is requiring more effort and producing lower rep counts, leading to lower estimates. This is a common pattern in late-block training when volume has been high for several weeks. A deload (1–2 weeks at significantly reduced volume) typically reveals that the underlying strength is higher than the dropping trend suggested. After a deload, estimated 1RM usually bounces back above the pre-deload baseline.

How Workout Lab Uses Estimated 1RM

Workout Lab calculates estimated 1RM automatically after every set using the logged load and rep count. The result appears in your exercise history as a continuous trendline, so you can see your strength trajectory for any exercise across any time window without ever needing to max out.

The practical benefit of this approach: your strength curve for an exercise like squat or bench press becomes visible data rather than a subjective impression. If you feel like “I’ve been getting stronger,” the estimated 1RM chart either confirms or contradicts that impression. If the trendline has been flat for six weeks, you have an objective basis for changing something rather than guessing.

Comparing estimated 1RM across exercises also lets you identify relative strength imbalances. If your squat estimated 1RM has been climbing steadily while your hip hinge estimated 1RM has been flat, the data points to a programming adjustment before the imbalance becomes a performance or injury issue.

The ghost values feature in Workout Lab uses your estimated 1RM data to surface a number to beat each session: a concrete per-set target based on your actual previous performance. That feature is explained in the guide to ghost values and per-set targets.

For tracking your training comprehensively and understanding how estimated 1RM fits into the bigger picture, why tracking your workouts matters is worth reading alongside this guide.

When to Test Your Actual 1RM Instead

Estimated 1RM serves most training needs most of the time. There are situations where testing an actual maximum makes sense.

Before a competition: if you’re competing in powerlifting or another strength sport, knowing your true 1RM for competition planning (selecting opening attempts, calculating attempts) requires a tested number. Estimated values carry enough uncertainty that basing competition strategy on them alone is inadvisable.

When the estimate has been derived from a long rep range: if your working sets have primarily been in the 8–12 rep range for several months, your estimated 1RM has been calculated from data that carries meaningful error. Testing an actual max (or at minimum a 3-rep max) recalibrates the estimate against ground truth.

When a significant training shift has occurred: returning from injury, switching training modes, or a substantial change in body composition can cause the estimated 1RM to drift from actual capacity in ways that aren’t fully captured by submaximal data. An actual test establishes a new baseline.

Outside of these situations, the continuous estimated 1RM trendline from regular training sessions is more informative than periodic testing and requires no disruption to your training schedule.

---

Estimated 1RM is a practical tool for tracking strength without frequent maximal testing. The formulas (Epley and Brzycki) are reliable in the 3–6 rep range and less reliable above 10 reps. As a continuous trendline across every session, estimated 1RM is more informative than periodic tested maxes for most training purposes. Rising trend: the programme is working. Flat trend: examine volume, intensity, and recovery. Dropping trend: fatigue is accumulating and a deload will typically reveal strength that the data hasn’t been capturing.