Well, for one thing, there are some best-selling novels out there that are crap. It's easy for an unpublished novelist to point to one of those and say, "I write better than that: ergo, since that crap is published, I should be published also."
There is some merit to that argument -- however, there is an important nuance that can't be ignored: the game is not "who can write the best book" : the game is, "who can sell the most books."
Publishing peope talk about 'branding' ad nauseum, but it's true: quality writing is only one of several factors that move books off of bookstore shelves and into beach tote bags. In fact, it may be one of the least important factors. Why? Simple:
A book buyer cannot tell if the writing is any good without reading the book. If she needs to read the book to find out whether it's any good... then what's the point?
The buyer can get around this problem if the author is established. She knows what she's getting when buying another Dan Brown, Sue Grafton, or Estate-Of-Clive-Cussler.
If the author is NOT established, the buying decision can come down to whether the book is on a subject the buyer wants to read about.
It's true! For a while I was reading anything set in the British navy of the Napoleonic era. Other people are nuts for Regency romances, or steampunk, or Viking stories, or cozies featuring sleuths who are also Franciscan monks, or whatever.
And this is why a genre book is easier to sell as a first novel -- it taps a ready market. The bar for writing quality is set lower. And so I postulate further:
A 'genre' book will sell in the established market for that genre, as long as it does not completely suck.
And so the 'pickup basketball to NBA' analogy falls apart. It's not whether we can play better basketball than the status quo: it's whether anyone wants to pay to see us play. And unlike basketball, sheer performance on the court is not the thing that convinces the paying public to part with their cash.