Ah so you mean the mathematical unlikeliness that evolution brought forth the complexity we see today, not that evolution happens at all. That makes more sense. I think it's hard to come up with solid arguments either way, because we only have a poor understanding of genetics and hence we can only give very rough estimates on the amount of complexity needed, and how much can be introduced in a given unit of time.
I'm not very well versed on that topic, but as for speciation, that happens very easily. For example even relatively short term separation of populations can stop interbreedability, e.g. by changing mating behavior. Once that happens it's just a matter of a few million years of random walks to grow feathers where there has once been fur.
The complexity argument has problems because in the small error bars of dna replication, it's so easy to introduce a sequence that actively hurts the animal, or is useless. Producing a whole new limb or whatever is way more complicated that we'd like it to be.
The discussion overall is part of a larger move by the intellectual/theological groups to bring back language (and psychology elsewhere) back into the old energy and matter world-view.
I agree it's hard to answer, the James Tour video is a nice place to start because he talks about biological engineering problems that his students are covering for him. They make little nano-robots and the like in part of the presentation. It seems inevitable that we won't see an overall answer for thousands of years.
That most mutations are useless is a feature, not a bug. Changes that neither hurt nor benefit are a major stepping stone to adaptions. You can have hair that looks more and more like feathers without hurting its performance as an insulator. You can have a spine with a few more or a couple fewer vertebrae without having a use for a tail or super flexibility. Phenotypic diversity is how populations adapt.
> Producing a whole new limb or whatever is way more complicated that we'd like it to be.
It isn't that much more complicated, though. Because as it turns out, you don't always have to code for that limb from scratch. You just have to flip a switch.
The key phrase here is "evolutionary developmental biology". We've learned that complex life has a lot of Lego-life quality - a lot of DNA codes for signals that enable or disable creation of whole organs at various locations, stages and cadences.
I don't have a problem with dna coding for legs existing. We don't know how the snake got that sequencing and the music video didnt solve it. It glosses over arbitrary changes in dna sequencing.
James tour's video in my parent post is far more in depth. Good luck.
I'm not very well versed on that topic, but as for speciation, that happens very easily. For example even relatively short term separation of populations can stop interbreedability, e.g. by changing mating behavior. Once that happens it's just a matter of a few million years of random walks to grow feathers where there has once been fur.