The sleeper stretch may be no better than not stretching at all

Athletes in overhead sports commonly develop posterior shoulder tightness and a loss of internal rotation at the shoulder joint, known as glenohumeral internal-rotation deficit or GIRD. Both are thought to contribute to the shoulder pain these athletes suffer, and stretching is the standard first response in gyms and physiotherapy rooms alike. Yet the evidence that stretching actually resolves either problem has been weak to date. This review set out to gather every randomised trial in English and Japanese testing stretching interventions for posterior shoulder tightness and GIRD, and to ask which stretches genuinely hold up.

The authors searched seven English-language and three Japanese-language databases from their inception to December 2015, considering only randomised controlled trials. Risk of bias in each included study was assessed using the Physiotherapy Evidence Database scale, and because the trials varied, the findings were synthesised qualitatively rather than pooled into a single statistic. Ten papers made the cut, eight in English and two in Japanese, ranging from low to high methodological quality. The interventions examined included the cross-body stretch and the active sleeper stretch, with effects assessed immediately and over the short term.

There was moderate evidence for positive immediate and short-term effects of the cross-body stretch on both posterior shoulder tightness and glenohumeral internal-rotation deficit, in asymptomatic young subjects. The sleeper stretch fared differently: moderate evidence suggested the active sleeper stretch might not be more effective than no intervention at all for improving either measure in the short term. No trial established long-term effectiveness for any stretch, and the evidence comes almost entirely from people without symptoms rather than athletes currently in pain.

The sleeper stretch is one of the most widely prescribed shoulder drills in overhead sport, so moderate evidence that it may not beat doing nothing is a genuine challenge to standard practice. The cross-body stretch, by contrast, emerges from this review with usable support, at least for short-term change in symptom-free shoulders. The large open questions are whether either stretch helps shoulders that already hurt, and whether short-term range gains translate into less pain or injury over time. The authors call for methodologically rigorous trials in symptomatic patients to answer both.

If shoulder stretching is part of your routine, this is a reason to favour the cross-body stretch when the goal is easing posterior tightness or restoring internal rotation, since that is where the supportive evidence sits. It is also a reason not to assume the sleeper stretch is doing anything, as it may not outperform skipping it entirely. Keep the limits in view: the evidence covers short-term changes in young, pain-free shoulders, so anyone with a genuinely painful shoulder is outside what these trials tested.