Seeming contradiction in tyrannosaurid cursorial abilities?

[Infinity Blade]

Tyrannosaurids are known for possessing hindlimb characteristics suggesting cursoriality (arctometatarsals, elongated distal limb segments, and absolutely enormous thighs). But it's been suggested that since the M. caudofemoralis was attached to the upper part of the femur, they wouldn't have been able to run for such a prolonged distance, or at least not for as long as hadrosaurs, potential prey, could. It has thus been suggested that tyrannosaurids would have been sprinters, making quick, short bursts of speed that could initially outpace prey like hadrosaurs, yet tiring out before their ornithopod quarry (link).

But cursorial animals aren't exactly known for pursuing their prey for only a short distance. So why then did tyrannosaurids have such an arrangement of M. caudofemoralis attachment to the femur if they had anatomical characteristics in their hindlimbs that otherwise point to cursoriality? Would they have simply used their bursts of speed and taken even better advantage of them with their cursorial features during the comparatively short time they chased after prey (which would make for a really good ambush predator)? Discuss.

[creature386]

Cheetahs are cursorial and known for hunting their prey for only a relatively short time (but probably still a bit longer than non-cursorial predators), so I guess the best way to resolve the contradiction is some golden middle (like in my example, more stamina than non-cursiorial predators, but not that much when compared to other cursors). Remember that nature does not think in boxes, it is very well possible among hunters to have different graduations of hunting stamina.