Most Sports Science Research is False: The Replication Crisis Hits Sport

In this episode, we’re joined by Dr Joe Warne, founder of the Sports Sciences Replication Center (SSRC) and senior author of a landmark new study that should send shockwaves through the world of sports and exercise science. Joe and his team undertook an enormous replication effort—repeating the methods of previously published research to see if the results hold up. The outcome? Just 28% of the studies could be replicated. Even more striking, the effect sizes dropped by an average of 75% when replicated.

This isn’t just a dent in confidence—it’s a crater. Scientific journals are trusted as sources of truth, yet Joe’s work shows that the majority of findings in this field don’t hold water. Perhaps even more concerningly, only 14% of original study authors were willing to share data or collaborate on replications.

We dig into how the SSRC pulled off this monumental study, why so many findings failed to replicate, and what it reveals about the deep structural issues in the field. From academic incentives to publish fast and flashy, to the natural resistance researchers have to being challenged and falsified, we explore how these forces fuel unreliable and unreproducible science.

Finally, we ask the tough question: what should coaches, athletes, and practitioners do when research can't be trusted? Joe shares his candid thoughts on how science must change—and what needs to happen to restore faith in the evidence base.

This is a must-listen conversation on one of the most pressing issues in science today.

Links

• More of Joe - if you join Discourse by making a small monthly pledge here, you can be like the cool kids and hang out with Joe in our VIP Community!
• Meta analysis in Strength and conditioning
• A webinar Joe Warne did on these issues

The three replication papers:

• The proposal that outlines the selection of papers in more detail
• The 28% paper that outlines the findings
• Reflections from Joe and some of the team on the process and findings

Hosted on Acast. See acast.com/privacy for more information.