100 million years. Such a long time should be enough to reach the maximum capacity of the species.
Compared to the large-scale transformation of one’s own species members into other species, a new species separates from its own and begins to compete with you, reaching a new steady state in the number of species in the new cycle. Perhaps this is more acceptable?
The lack of branching is definitely a concern (to me). Even in cases where the player species is miles ahead of others - i.e. having a nucleus when no other cells do - it never speciates. I think that removes quite a bit of challenge from the game.
If I were implementing more player-species speciation, I would probably stick some special code in the auto-evo that either all speciations or specifically player speciations have severe mutations and a guaranteed color change.
The player species is as likely to be selected as the source for a new species emerging as any other. It’s just that if there are 5 species the chance is 1 in 5, but when there are 100 species then the chance is only 1 in 100. And of course the normal rules still apply: a split species must gain population for auto-evo to actually make it exist (otherwise the attempted created species is just silently discarded).
Of course there’s the differing selection pressure step that can split a species in two, which is explicitly coded to not consider splitting the player into two (so that the player doesn’t randomly lose up to half of their population; people complain about auto-evo hating their species enough as-is). So the end effect is that the player species is less likely to be used as a source for new species. And also most players make changes that auto-evo doesn’t really like so that also means that auto-evo created species would be preferred for auto-evo to make new species from.