You don't have a problem with my analysis of the increase of genetic variability through the accumulation of neutral mutations, then. but you do have two related problems:
1. Haldane's estimate of beneficial mutations
Haldane did NOT estimate the rate of beneficial mutations to be 1 in 300 generations; rather - unless you're referring to some other 300 generation estimate - that's the time for a SUBSTITUTION of a gene that becomes disadvantageous (e.g. due to an environmental change). Haldane did NOT, as far as I know, consider the fate of a new, rare, beneficial mutation. Nor is this the estimate of the rate of appearance, anywhere in the genome, of new beneficial mutations.
Also, I'm not sure what specific point you're making with this conservative estimate of beneficial mutations, anyway. Speciation, for example, only requires neutral mutations (or other genetic change).
2. "fanciful" selection is required to weed out the larger number of harmful mutations
I wasn't aware that selection against harmful mutations was controversial; even creationists usually grant this. Perhaps you could point out an example of what you're talking about when you talk about "tales of ... almost omniscient" selection. I mean, in the scientific literature, presumably population genetics.