I think some people at any given point in time (and some institutions and some governments) do a better job than others at splitting their resources because in part because they have a better framework at recognizing what may reach that critical threshold faster than others based on a given set of conditions (or have lower switching costs when it comes to deciding to not allocate resources to something after a certain point).

You don't split, but sell when successful then start other thing. That's the serial entrepeneur thing.