When a training app suggests an exercise, there is no guarantee it gets done. We looked at which exercises Tsemp users most often replace with something else over a 90-day period. Calf raises, pull-ups and chest-supported barbell rows top the list, and several are swapped out in more than half of the cases. Below are the 15 exercises, the methodology behind the numbers and what the numbers cannot tell you.
The short answer
- Most swapped out: seated barbell calf raise (62.3%), chest-supported barbell row on an incline bench (60.9%) and wide-grip pull-up (60.0%).
- Four pull-up variations appear on the list, along with three different calf raises.
- Lunges, hip thrusts and glute bridges are swapped out roughly one time in three.
- The numbers show behaviour, not causes. They do not tell us whether an exercise was replaced because of equipment, discomfort, preference or something else.
The 15 most frequently swapped exercises
| # | Exercise | Share of occurrences swapped out |
|---|---|---|
| 1 | Seated barbell calf raise | 62.3% |
| 2 | Chest-supported barbell row (incline bench) | 60.9% |
| 3 | Wide-grip pull-up | 60.0% |
| 4 | Cable calf raise | 44.9% |
| 5 | Assisted pull-up | 44.8% |
| 6 | Kettlebell squat | 42.1% |
| 7 | Standing machine calf raise | 40.2% |
| 8 | Overhand-grip pull-up | 39.5% |
| 9 | Dumbbell reverse lunge | 38.7% |
| 10 | Seated barbell overhead press | 37.4% |
| 11 | Cable lateral raise | 36.6% |
| 12 | Dumbbell hip thrust | 34.7% |
| 13 | Glute bridge | 33.5% |
| 14 | Neutral-grip pull-up | 33.3% |
| 15 | Dumbbell Bulgarian split squat | 33.0% |
The figures come from Tsemp users' logged workouts over 90 days and cover recreational strength trainees. They describe training behaviour, not physiology. See the methodology below.
What patterns does the data show?
| Group | Exercises on the list | Range |
|---|---|---|
| Calf raises | Seated barbell, cable, standing machine | 40.2–62.3% |
| Pull-ups | Wide grip, assisted, overhand, neutral grip | 33.3–60.0% |
| Single-leg work | Reverse lunge, Bulgarian split squat | 33.0–38.7% |
| Hip extension | Dumbbell hip thrust, glute bridge | 33.5–34.7% |
| Shoulders | Seated barbell overhead press, cable lateral raise | 36.6–37.4% |
| Other | Chest-supported row, kettlebell squat | 42.1–60.9% |
Calf raises stand out most: all three variations on the list are above 40%. This is consistent with Boostcamp's analysis of one million workouts, in which seated and standing calf raises were among the most frequently skipped exercises. The measures differ – Boostcamp counted skipped sets, we count swaps – but the direction is the same.
Pull-ups take up the most space on the list, with four variations. Even the assisted version, which is easier, is swapped out in almost half of the cases.
Many of the exercises need specific equipment: an incline bench for the row, a cable station, a calf raise machine or a pull-up bar. We do not know whether the equipment was busy. That is one possible explanation, but our data can neither confirm nor rule it out.
What happens around the swaps?
In a separate 30-day extract we saw the following:
| Measure (30 days) | Value |
|---|---|
| Users who swapped at least one exercise | 230 |
| Total swaps | 1,832 |
| Share of suggested weights and reps that users changed | approx. 29% |
| Users who searched for an exercise | 371 |
| Of those, users who abandoned a search at least once | 292 |
Users adjust their plan as they go. Most people who look for a different exercise also abandon the search at least once. That may suggest a good alternative is not always easy to find, but the data does not show why the search was abandoned.
Methodology
| Data source | Logged workouts in the Tsemp app (iOS and Android) |
| Period, exercise list | 90 days |
| Period, other figures | 30 days |
| Measure | Share of the times an exercise appeared in a training program in which the user replaced it with a different exercise |
| Inclusion criteria | Exercises swapped out by at least 10 different users and with at least 40 occurrences in programs |
| Population | Tsemp users in Finland – recreational gym-goers, not competitive athletes |
The user base is described from the app's sign-up questionnaire (4,141 accounts): 51% women and 49% men, median age 26, and 87% train in a gym. 42% have muscle gain as their goal, and 42% fat loss.
What the numbers do not show
- Causes. We do not know why an exercise was swapped out.
- The replacement. The list does not show what the exercise was replaced with. A swap can easily be to a similar exercise.
- Physiology. The numbers say nothing about how effective or safe the exercises are.
- The exact number of occurrences per exercise. We only publish that each exercise meets the minimum thresholds.
- Generalisability. The users are Finnish recreational lifters in a single app. Other populations and apps may show different patterns.
How to cite this data
You are welcome to use the figures with attribution and a link to this page:
Tsemp. The Exercises People Swap Out Most: Data from Tsemp. October 2026.
Frequently asked questions
Which exercise is swapped out most often? In Tsemp's data, the seated barbell calf raise is replaced in 62.3% of the times it appears in a program. The chest-supported barbell row (60.9%) and the wide-grip pull-up (60.0%) follow close behind.
Why do people swap exercises? The data does not show the reason. Possible explanations include busy equipment, an exercise feeling uncomfortable or simply preferring another variation, but the numbers can neither confirm nor rule these out.
Is it bad to swap an exercise? Not necessarily. Swapping to an exercise that trains the same muscles gives much the same result. The problem only arises if a muscle group goes untrained because the exercise is skipped without a replacement.
How much data is the list based on? The list is based on 90 days of logged workouts and only includes exercises swapped out by at least 10 different users and with at least 40 occurrences. Over a 30-day period, 230 users made a total of 1,832 swaps.
Can I use the figures in my own article? Yes, with attribution to Tsemp and a link to this page. Please also mention the limitations: the figures describe behaviour among Finnish recreational lifters and say nothing about causes.
Sources
- Tsemp. Own training data, 90 days (exercise list) and 30 days (other figures), and sign-up questionnaire (4,141 accounts).
- Boostcamp. State of Lifting 2025. boostcamp.app
Author
tsemp-redaktionen
Träningsredaktionen på tsemp
tsemp-redaktionen skriver om träning, kost och återhämtning – med målet att göra kvalitativ coaching tillgänglig för alla.