Regarding the playoff adjustment, might it make more sense to set the centre-point for this at 50%? When we’re watching a game we’re not estimating whether or not a team is more likely than average to make the playoffs; we’re interested in the binary outcome of whether or not they do.
And implicitly, the leverage of any individual game with regards to a team’s chances is higher when those chances are already close to 50%.
Furthermore, you could add an extra coefficient to recognize where the two teams’ chances are highly interdependent i.e. where they are likely battling for the same playoff spot.