One of the arguments I see around here often resolves to "Why don't we see humans evolving?"
If you mean "why don't we see humans speciating", the answer is simply that we have too high a gene flow and too large a reproductive population. Both of these pop.gen. parameters are vastly higher than the historical values.
If you mean "why don't we see humans evolving" then of course the answer is that we do.
how often should we expect to see something like that happen?
A new nearly-neutral allele is expected to spread throughout the entire global population of human beings in something considerably more than 32 billion generations (that is, if my recollection of the 4N equation is correct). Whether this has anything to do with speciation is anybody's guess. That's why speciation tends to occur in small isolated populations, where the time taken for alleles to fix is considerably lower.
This is the main reason why we don't expect to see a new species of Homo any time soon.
Of course we may want to consider an allele that has a selective benefit for the human population. that would spread much faster. I'm only talking here about a neutral allele - one that makes no difference whether you have it or not. Now imagine a selectively beneficial allele in a small isolated population, and the numbers start to get reasonable (kind of).
would we recognize an evolved human if there was one?
You wouldn't. There was nothing special about the common ancestor of all living human beings. She was just lucky. She looked exactly the same as any other female Homo sapiens.