How scoring works

EARN EVERY POINT

Calumet turns what you did into two numbers. This page shows what goes in, what keeps it fair, and what would score higher next time.

Two scores, one set of rules

The work, and what it cost you

Every workout gets both. They use the same rules for everyone; they differ only in who they compare you with.

Two scores, one workout1 minute · no sound

Alex and Sam do the same workout. Their Raw Scores stay close because the work was the same; Sam's heart rate ran higher on the standard range, so his reps counted a little more. Personal Score uses each person's own range: Sam worked at 77% of his and Alex at 61%, so Sam earns a higher score and badge. The cards below explain each one.

Raw Score

Measures the work itself without your profile: reps, weight, time and distance, and the fatigue you carried into each set. All are judged against a baseline that is the same for everyone. That's why Raw is the number on leaderboards.

Personal Score

Measures how hard that work was for you. It uses your own heart-rate range, from resting to maximum. Two people can do the same sets and get different Personal Scores. It's the one you beat next time, and it sets your level.

What raises a score

Strong sets. Smart rests. Higher scores

  1. The work

    More reps, heavier weight, longer holds, further and faster. A stronger performance scores higher.

  2. Fatigue

    Short rest and back-to-back exercises on the same muscles make every rep worth more. The fatigue meter carries across the whole workout.

  3. Heart rate

    Your watch streams heart rate live. Time spent working hard, in the upper zones of the Effort Spectrum, adds to the fatigue meter. Fatigue counts from your first set. Warm-ups help you, but don't add points.

  4. Keeping it up

    Fatigue fades during long rests, so one hard burst followed by coasting scores less than steady, sustained work.

  1. Low
  2. Medium
  3. Ideal
  4. High
  5. Max

One set, worked out

Where the +51 came from

An example with round numbers: the last of three sets of back squats, 10 reps after 90 seconds of rest.

  • The work: 10 reps at a weight close to what the standard person can lift for 10.
  • The fatigue meter: two earlier sets on the same muscles, a short rest, and a heart rate that stayed high, so each rep counts 24% more.
  • To score higher next time: one more rep, a little more weight, or a shorter rest. After every workout, the app's Why this score panel shows what drove each set.

Fair by design

Every point is earned

Measured, not guessed

Your score comes from what you did and what your watch measured, never from a "how hard did that feel?" rating. Every point you earn is real, and nobody can talk a score up.

One standard for everyone

Raw Score still rewards effort, including the fatigue you push through, but it measures everyone against the same baseline, with no adjustment for age, sex or body type. That's what lets anyone challenge anyone.

No junk volume

Light reps with long rests barely move the fatigue meter. If chasing points wouldn't make you fitter, it shouldn't score.

Rookie to Legend

Sixteen Badges to Earn

After each workout, Calumet awards a badge from your Personal Score. Matching an average gym-goer on the same workout earns right in the middle at Fighter. Lift more, move faster, push harder, and become a Legend.

Champion

Tier four puts you inside the Calumet C.

Open data, our own model

Where the standards come from

The average gym-goer behind every badge is built from public data and published research, in Calumet's own model. No commercial strength-standards table is used.

  • How strength changes with bodyweight and age. Fitted by Calumet to data from the OpenPowerlifting project, https://www.openpowerlifting.org. You may download a copy of the data at https://gitlab.com/openpowerlifting/opl-data. (Public domain; dataset of 31 October 2026.)
  • Where average sits. Placed at the average lifts of adults who train (Isenmann et al. 2026, Scientific Reports, doi:10.1038/s41598-026-60646-z, CC BY 4.0), with the spread of bench press strength in healthy adults aged 18 to 30 (Brown et al. 2022, Medicine & Science in Sports & Exercise).
  • Reps and one-rep max. Converted with the repetitions-at-%1RM estimates of Nuzzo JL, Pinto MD, Nosaka K, Steele J, "Maximal number of repetitions at percentages of the one repetition maximum", Sports Medicine 2024;54:303-321, doi:10.1007/s40279-023-01937-7, licensed under CC BY 4.0. Changes: Calumet uses the main model's point estimates only and interpolates between them.
  • Equipment. Dumbbell and Smith machine bench press from Saeterbakken et al. 2011 (Journal of Sports Sciences) and Cotterman et al. 2005 (Journal of Strength and Conditioning Research); plank and side plank times from Strand et al. 2014 and McGill et al. 1999.
  • Cardio. Energy costs from the 2024 Adult Compendium of Physical Activities (Herrmann et al., Journal of Sport and Health Science), pacompendium.com. Numbers Calumet adjusts are Calumet's, not the Compendium's.
  • Everything else. Where no published figure exists, Calumet sets an estimate and records why, and the app marks badges that rest on one.

See your score

Calumet is in testing on Android. Leave your email and we'll send your invite.