British Journal of Sports Medicine:

[…] the Nordic hamstring exercise cut injury risk by 49% across 15 trials […]

[…] less than 70% of maximum EMG activity recorded during overground sprinting.

It is not always the spike in the load that causes the hamstring injury. Reduction of match exposure in the 2 preceding matches has been identified as a significant injury risk factor, suggesting chronic under-preparation is equally dangerous.

[…] Athletes who have had previous hamstring strains display significantly lower mean frequency of hamstring activation compared to the uninjured during jumping and landing tasks, with the activation deficit associated with higher landing forces and loading rates.

This means the athlete lands harder and loads the joint more despite subjectively feeling recovered (14).

Weak trunk and gluteal support: footballers who went on to injure their hamstrings showed lower gluteus maximus, erector spinae and oblique activity during swing. The opposite iliopsoas can add more than 25mm of stretch to the biceps femoris long head, while gluteal and oblique contraction reduces it (7).

[…] The Askling H-test earns its own line. It is a ballistic active straight-leg raise that exposes the insecurity an athlete can still feel after strength, range and palpation look clear.

[…] athletes reported marked insecurity on the injured leg despite a normal common clinical examination; the authors also reported that 95% of athletes in their clinical use of the test felt insecurity before final clearance (23).

[…] However its evidence base is limited to two studies led by its own developer, so findings need further validation (41).

Hamstring injury risk isn’t just about hamstring strength. Trunk and gluteal activation during sprinting may matter, while altered muscle activation and landing mechanics can persist after symptoms resolve.

Original research: Prospective study of gluteal and trunk activation in footballers.