The question is not whether work capacity is trainable -- it obviously is. The question is how long it takes to convert a very high strength base to work capacity, and how efficiently a lower strength base could be converted to that same work capacity. If Kendrick stopped training for 1RM and devoted his already-developed strength to "Isabel," how long would it take him to beat Malleolo? IOW, what is the most efficient way to train for Isabel: doing Isabel every week, or getting your snatch up to the point where Isabel (as RXed, of course) is a smaller percentage of your absolute strength? IOOW, if you want a high-volume submax effort, is it better to make the effort even more submaximal, or to train for more work at the same level of submaximality?
I think you know which has worked better.