Does working out with a friend help? What the research says

Usually, yes, though mostly in how long and how often you exercise rather than in any magic extra strength. In lab studies of the Köhler effect, people paired with a slightly fitter partner kept going longer: in one trial, women cycling with a virtually present partner exercised about 21.9 minutes versus 10.6 minutes alone. A 2023 meta-analysis of 19 studies (1,912 people) found a significant motivation gain from partnered exercise. In an 11-week trial of 790 students, groups that could compare themselves with peers attended about 90% more classes than groups without comparison, while support alone did not help. The best partner is a little better than you, and the effect works even when they train somewhere else.

By escapethematrix.app · Updated October 27, 2026

"Get a workout buddy" is advice almost everyone has heard, usually backed by a statistic nobody can trace. There is real research on this, and some of it is surprisingly specific: who the partner should be, whether they need to be in the room, and whether competing beats encouraging. This guide sticks to what the studies actually measured, and ends with a simple way to run a fair push-up or squat battle with a friend.

escapethematrix.app publishes RepSquad, Body Vitals and Presence, three iPhone apps. RepSquad is mentioned once, near the end, and disclosed there. Everything else on this page is cited to peer-reviewed or official sources.

The short answer

• Partners make people keep going longer, especially when the partner is a little fitter and your effort counts toward a shared result. This is the Köhler effect, and it has been shown in planks and cycling.
• Comparison seems to beat cheerleading. In one large trial, seeing how peers were doing raised class attendance; a support-only network did not.
• Friends rub off on each other over months and years, and joining a program with friends has been linked with sticking to it.
• The evidence has limits. Most Köhler studies are short lab experiments with students, and some groups showed no effect.

What is the Köhler effect?

The Köhler effect is the extra effort the weaker member of a team puts in when their performance matters to the group. It is strongest in so-called conjunctive tasks, where the team is only as good as its weakest member: for example, "our score is however long the first of us holds the plank." Knowing you are the one who could let the team down seems to make people try harder.

A research group at Michigan State University, led by Deborah Feltz with Norbert Kerr, has tested this in exercise for over a decade, often using exercise video games with a partner shown on screen.

The exercise studies

A 2023 meta-analysis pooled 19 peer-reviewed studies with 1,912 participants and found a statistically significant motivation gain for partnered exercise over exercising alone (Samendinger et al., Kinesiology Review). The size of the gain depended on whether studies measured persistence or intensity, the participants' age, and whether the partner was human or software.

Where it didn't show up

The effect is not automatic. In the 12-session study, women showed no difference between conditions, and nobody differed in the first half of the sessions. The authors also point out that finding a partner with "just the right amount" of superiority is hard: too small a gap and there is nothing to chase, too large and keeping up stops feeling possible. And a software partner helped, but less than one people believed was real.

Competition or support: which helps more?

Zhang and colleagues at the University of Pennsylvania randomised 790 students into four groups for an 11-week program of free exercise classes (Preventive Medicine Reports, 2016). Some groups could see how peers in their online network were doing (social comparison), some could chat and encourage each other (social support), one had both, and one had neither.

• The groups with comparison attended about 90% more classes than the groups without it (mean 1.9 vs 1.0). After adjusting for baseline differences, comparison raised attendance by 62% (95% CI 5% to 119%).
• Social support on its own had no significant effect, and adding it did not help the comparison group.

The authors' conclusion was blunt: social comparison was more effective than social support. The caveats matter too. It was one university, the online networks were set up by the researchers, the peers were not existing friends, and the only outcome was attendance.

Friends rub off on each other

Aral and Nicolaides analysed about 1.1 million runners in a global social network over 5 years, using weather as a natural experiment (Nature Communications, 2017). When rain kept one friend from running, the other changed too, which let the authors separate influence from simply having similar friends. Their example: an extra kilometre run by a friend led to about 0.3 km more for the runner, and an extra 10 minutes by friends led to about 3 minutes more. One twist: less active runners influenced more active runners, but not the reverse. People seem keen to stay ahead of the friend just behind them.

Sticking with it

Partners may matter most for showing up week after week. In a weight-loss trial of 166 people (Wing and Jeffery, 1999), 95% of those who joined with friends and also got social-support training completed treatment, against 76% of those who joined alone with standard treatment; 66% versus 24% kept their weight loss fully from month 4 to month 10. That compares two combined conditions, so it cannot say how much came from the friends alone.

Couples research is smaller. In a 14-day diary study of 95 young adults in relationships, days spent exercising with their partner brought higher positive mood during exercise and across the day, and higher relationship satisfaction, but no difference in the amount of exercise (Sackett-Fox et al., 2021). A 12-week pilot with 9 older couples found couples exercising together walked a little more (86.5 vs 66.1 minutes a week), but the difference was not statistically significant (Hornbuckle et al., 2021). So: a partner can make exercise more pleasant, and may help with consistency, but the couples evidence is thin.

How to set up a fair push-up or squat battle

The Köhler research suggests a few rules that make a friendly battle both motivating and fair:

1. Pick a partner who is a bit better than you. Not twice as strong; a moderate gap is the sweet spot in the research.
2. Agree one rep standard. For push-ups: body straight, chest down until the elbows reach about 90 degrees, full lockout at the top. The push-ups and heart health guide has a fuller checklist.
3. Count the same way. Different counting methods turn a battle into an argument. In our review study of rep-counter apps, 25.2% of unhappy reviews of auto-counting apps said the counting was unreliable, so agree how you will count before you start, and make any corrections visible.
4. Choose the format. "Best single set" rewards a max effort; "total reps over a week" rewards showing up. The second fits the adherence evidence better.
5. Set a deadline and rest days. A 24-hour or 7-day window gives both people time to recover between attempts.
6. Don't chase every set to failure. Hard sets that stop a rep or two short appear to build muscle about as well as all-out sets (see how close to failure should you train). Save the true max for the battle set.
7. Track your own number, too. Your push-up count by age is a rough guide; beating your own count from last month is the result that matters.

Remote partners count too

Several of the Köhler studies used partners who were not in the room, only shown on a screen, and the effect still appeared. A friend in another city who posts their score before you start your set fits that model well. A partner you know is real seems to work better than an anonymous or software one.

When a partner might not help

• If the gap is too big. Always being the weak link, by a lot, can stop feeling motivating.
• If competition pushes form out of the window. Rushed, shallow reps are not better reps, and an injury ends the battle for both of you.
• If you would rather train alone. Some people simply prefer it, and solo training works.

This guide is not medical advice. If you have chest pain, dizziness, a heart condition or are new to exercise, talk to a doctor before testing your limits.

Running a battle with an app

Disclosure: we make RepSquad. RepSquad counts reps with the iPhone camera, on-device, and has Rep Battles: challenge one friend or a group on any exercise, everyone does their set on their own time within 24 hours, 3 days or 7 days, and you can rematch in one tap. Groups work as a private league joined with a 6-character code, with best-single-set or total-reps challenges. If someone corrects their camera count after a set, the battle standings show the change next to their score. Battles and groups are free with Sign in with Apple, and push-ups and squats are free to count; some other exercises need an in-app purchase (App Store). You can also try counting a set in the browser with our push-up counter.

Sources

1. Irwin BC, Scorniaenchi J, Kerr NL, Eisenmann JC, Feltz DL. Aerobic exercise is promoted when individual performance affects the group: a test of the Kohler motivation gain effect. Annals of Behavioral Medicine. 2012;44(2):151-159. PubMed.
2. Feltz DL, Kerr NL, Irwin BC. Buddy up: the Köhler effect applied to health games. Journal of Sport and Exercise Psychology. 2011;33(4):506-526. PubMed.
3. Feltz DL, Forlenza ST, Winn B, Kerr NL. Cyber buddy is better than no buddy: a test of the Köhler motivation effect in exergames. Games for Health Journal. 2014;3(2):98-105. PubMed.
4. Forlenza ST, Kerr NL, Irwin BC, Feltz DL. Is my exercise partner similar enough? Partner characteristics as a moderator of the Köhler effect in exergames. Games for Health Journal. 2012;1(6):436-441. PubMed.
5. Max EJ, Samendinger S, Winn B, Kerr NL, Pfeiffer KA, Feltz DL. Enhancing aerobic exercise with a novel virtual exercise buddy based on the Köhler effect. Games for Health Journal. 2016;5(4):252-257. PMC.
6. Samendinger S, Hill CR, Ahn S, Feltz DL. The Köhler motivation gain effect with exercise tasks: a meta-analysis. Kinesiology Review. 2023. doi:10.1123/kr.2022-0047. University of Miami record.
7. Zhang J, Brackbill D, Yang S, Becker J, Herbert N, Centola D. Support or competition? How online social networks increase physical activity: a randomized controlled trial. Preventive Medicine Reports. 2016;4:453-458. PMC.
8. Aral S, Nicolaides C. Exercise contagion in a global social network. Nature Communications. 2017;8:14753. PMC.
9. Wing RR, Jeffery RW. Benefits of recruiting participants with friends and increasing social support for weight loss and maintenance. Journal of Consulting and Clinical Psychology. 1999;67(1):132-138. PubMed.
10. Sackett-Fox K, Gere J, Updegraff JA. Better together: the impact of exercising with a romantic partner. Journal of Social and Personal Relationships. 2021;38(11):3078-3096. doi:10.1177/02654075211012086.
11. Hornbuckle LM, Rauer A, Winters-Stone KM, et al. Better together? A pilot study of romantic partner influence on exercise adherence and cardiometabolic risk in African-American couples. Journal of Racial and Ethnic Health Disparities. 2021. PubMed.
12. escapethematrix.app. What frustrates people about push-up and rep-counter apps: 1,543 App Store reviews analysed (October 2026).
13. US App Store listing for RepSquad (checked 21 October 2026).

This guide summarises published research. It is not medical advice. If you have symptoms, a heart condition or are new to exercise, talk to a clinician before testing your limits.

Frequently asked questions