Good video, as usual. Watching the beginning of it where you were explaining the prep you did, something occurred to me which I hadn't thought about before. What if drilling holes at each end of the crack doesn't actually make any difference but everybody does it because that's the way it's always done? Even if anyone took the time to try to perform a controlled test, it would be extremely difficult to get two pieces which were exactly the same in all respects.
My reasoning on why the holes might not make any difference is this: You described the crack as if it were an object exerting force on the metal, like a flooding river eroding land as it runs, and that's the way we generally think of it. But the crack is not a thing; it doesn't have substance and mass. It is the result of stresses in the metal which pull the metal apart along a fault line. Drilling holes at the current ends of a crack might help if they do something to relieve the stresses in the metal. Otherwise, can holes really be defeating the force of the crack by imposing a void that it can't cross? What if the holes actually only extend and widen the crack, and if we didn't perform the repair on it right away, we might see the crack continue on the other side of the holes?
People in colonial America didn't eat tomatoes because everyone knew they were poison, so sometimes the collective wisdom is only as sound as our unwillingness to test it. On the other hand, if I have any cast iron repairs to make, I'll undoubtedly drill holes at the ends of the crack because that's how it's done, and I could be wrong.