Alright so we have established that those creatures who live closest to the seafloor are more likely to be fossilised. That's cool. That is logical.
However that does not address my objections to your two assertions.
All right. Let's make a few assumptions and see where logic leads us. Let's suppose that this species of fish has exactly 10 fossils that science will eventually find. Since a disproportionate number of these fossils will have been made from fish swimming deeply in shallow waters, can we not assume that, by chance, scientists are more likely to find one of those fossils first. Can we not?
Then scientists will draw conclusions based on that fossil, conclusions that may not be accurate.
If that is true, we should see a number of follow-up stories saying something like "fossil sheds new light on" or "new fossil surprises scientists"
For example,
http://www.scientificamerican.com/article/fossils-shed-new-light-on-human-gorilla-split/
This article basically says that what was concluded from previous fossils may be wrong. I can find hundreds of these types of stories all over the web. There must be thousands more that will never be published.
http://www.nytimes.com/2005/06/03/science/t-rex-fossils-surprise-she-was-ovulating.html?_r=0
"Among its rock-hard fossils, the scientists had already isolated soft tissues, including blood vessels and cells lining them -- a most improbable discovery after 70 million years."
I'd say so! In fact, it might cause a skeptic to be skeptical about whether the fossil really is 70 million years old.
You see, I'm of the opinion that every time a story comes up that surprises, baffles, or modifies what was thought before, that's
de facto proof that what was thought before was wrong. That means that scientists routinely make assumptions and inferences that will be discarded or modified in the future.