It wouldn't be necessary for a life bearing asteroid to hit Earth in order to seed Earth. Rather, a collision of a life-bearing asteroid (perhaps since frozen) in space would create a large number of fragments, any one of which could seed Earth. This strikes me as the most certain part of the scenario. After all, this is how meteorites are created.
I really don't think it is. For any favorable condition you might find in asteroids, early planets can probably match it, and a bunch of other possible conditions besides, with the additional benefit of not requiring a stage where the nascent life has to survive an impact at orbital speeds, subsequent flight through cold irradiated vacuum, and then re-entry through a dense planetary atmosphere. I'm not saying it's impossible, but I'm saying it needs a lot more evidence to take it seriously.
Either way, the notion of asteroids being more hospitable to OoL than planets, with their complex and varied environments and chemistry, would require quite the extraordinary evidence in my book.