Not forgetting "idiot start". Full revs, first gear, slide you foot off the clutch pedal. And gear changes, sudden load changes.
Take a look at an engine as it moves inside its compartment during a drive-cycle test on a rolling road and you ideas of engine motion will never be the same. Smooth doesn't really exist.
Your 300 ft-lb. "brake torque" figure used to describe engine output is usually less than the instantaneous torque peaks that occur in the cranktrain. However, the difference between peak and mean torques in an even firing N/A V8 are likely not too great.
If your V8 engine is bolted to a transmission (ie. like the classic front engine, rear drive American V8 muscle car) then you must also consider the full reactions produced by the transmission gear torque multiplication.
Finally, if your mounts are compliant, you must consider both engine dynamic forces and engine/trans mass inertia during acceleration/braking/cornering.
Yep - all that. But for maximum torque on the engine mounts, start at the driven tyres and work back. eg for RWD and traction limited launch:
tailshaft torque = (static weight on rear axle) x (% weight transfer to rear) x (tyre friction coefficient) x (rolling radius) / (final drive ratio)
If you want to get anal you also need to account for inertia of the wheels times their acceleration when breaking traction. On the other hand, if a clutch dump is required to break traction, this will REDUCE the torque seen by the engine mounts. (Think of the extreme case of 1:1 gearbox ratio, rev engine, dump clutch and kill ignition simultaneously. Clearly the engine mounts experience no torque - the inertia of the engine rotating components does all the work). This assumes the engine rotates in the smae direction as the tailshaft - not always true for FWD cars.
There are other exceptions - torque tube drive for example.
Engineering is the art of creating things you need, from things you can get.