If the motor were not stopped on the dead pole comm segment, it probably does start to rotate, but as soon as it reaches the dead pole, it stops. The armature has to have some momentum to travel past the dead pole comm segment to one what works. A severe contamination that prevented the brushes from contacting a single segment, or some other ailment that causes the same thing would result in the same symptom, but I've never seen such condition. An out-of-round would need to be extremely severe and again not likely. Another possibility that I have experienced is a thrown comm segment, but that's going to cause more problems than just failing to self-start, and I've yet to see a new motor with a missing segment. Again, the reason a missing comm segment causes the same symptom is it's a dead pole. Dead pole is really the only practical cause. In manufacturing, if the wire insulating varnish doesn't get stripped clean where it connects to the comm tab or any other insulating contamination in the joint between comm tab and wire occurs, then a pole will be disconnected and this problem results.
Today, very few products receive 100% quality control inspection. Product integrity depends on manufacturing workmanship and process control. At best, only a small sample may be quality control inspected/tested. This allows defects to pass through to the customer unless the manufacturing process is near perfect. Some manufacturers rely on build process that is good enough to not create many defects and find it far cheaper to simply replace an occasional defective item that reaches the customer than to provide more quality control. Other manufacturers simply do not care about customers receiving defective products. That's the customers' loss. In this case, it's buyer beware.