My understanding is that an unlikely event, given time occurs, but once it occurs in meaningfully diminishes the opportunity for subsequent unlikely events to take hold. Similarly, were several unlikely events to have occurred, it is by no means likely that their basic premises were equally viable for a few billion years. That quite coincidentally, of the inventions, it is more likely for a particular gross plan to be fundamentally superior. Within a defined niche (in this case the planet) the best suited is most likely to excluded the others. If the most suited is also continuously diversifying, one could imagine an impenetrable wall of incumbency advantage.
Finally, if one applies notions of drift and founder effect, the tendency is to arrive at one. If I throw many colored stones in a pot, sample some, replicate those and keep doing so, the ultimate result is a homogeneous population. It is at first counter-intuitive, but if forms much of the basis of mitochondrial population tracking e.g.
Pushing back against this natural effect of sampling and replication to lead to one, is mutation. The relative roles of those form the basis of much discussion and dispute, but to me at least it indicates that even if a hundred or a thousand or a million primordial events occurred, in all likelihood that which walks today will trace to one, as observed.