The result of this increased stability is an increase in the virus’s ability to latch onto the protein it targets on the surface of human cells. This is true even though the mutant form of spike doesn’t have an increased affinity for its target protein; there’s simply more of the functional form around.
The study comes with some very important caveats: one, this work wasn’t done with an intact coronavirus. Instead, it was done by putting key coronavirus proteins into a different (and harmless) virus. Second, and more critically, this work wasn’t done in living organisms, but in cells. There’s a lot going on inside the body that makes it a very different environment from the layer of warm liquid that cultured cells live in. And as with all draft papers, keep in mind that peer review may cause the final, published results to look significantly different from the ones here.
Those caveats handled, the evidence does seem to be tending toward this mutation making a difference in the virus’s spread. We’ve gone from having a suggestive pattern of genetics to having a reasonable hypothesis about the mechanism underlying them and some biochemical support for that mechanism. Even if none of these bits of evidence is strong enough to make the case decisive, there’s beginning to be some weight behind the idea.