I'm sure that the effects of allele change are predictible to some extent. Afer all there are companies making money in genetic engeneering.
I really have two questions: One is, why isn't this the central focus of ID research? The second, and more troubling is, how can you predict the effects on reproductive success in a complex ecosystem?
That is precisely the conumdrum that natural selection addresses. I agree with you that anything other than an omniscient (in all four dimensions) being could not make that prediction.
Now for the nitty gritty. Even if you have the computational horsepower to make predictions of reproductive success, it doesn't change the fact that natural selection works just like Darwin said it does.