Massey's Composite of 100 plus ratings has UCF #8....the computers don't care about all the human thoughts on rankings.
Massey Composite has the first four as Bama...Clemson...Georgia...Notre Dame...
The humans put in Oklahoma over Georgia....but UCF was not close to top 4...but would have made an 8 team playoff as the #8 seed.
UCF notwithstanding (I didn't bring them up), I've already said that I am for an expanded playoff based solely on inclusion. More participants makes for more meaningful games,
more better drama, and more interest. Before the CFP, there was 1 meaningful game and 39 exhibitions. The 4 team playoff was half baked in its implementation. 5 "CSMA" conferences (6, I believe in its development, before the Big East was relegated) to fill 4 spots does nothing to quell the controversy of the BCS. It only served to push it 2 spots further down the rankings, adds only 2 more meaningful games, and maintains 38 exhibitions, which are increasingly less meaningful as star upperclassmen skip them in order to keep their bodies fresh for the NFL spring festivities.
Now to answer your question...
Does the CFP Selection committee exclusively use Massey? If not, why not? Take it a step further, why even have a selection committee at all if Massey is the most accurate predictive model?
Rankings aside, there will always be a human element. Humans develop these algorithms. Humans determine the data inputs. When all is said and done, humans want to maximize return.
Humans put in Oklahoma over Georgia because of the eye test and reputation. Oklahoma is P-5 and won their CCG. Georgia did not. Oklahoma (8) is not close to the top 4 either and only 0.01 ahead of 9th rank UCF in the Massey Ratings as of December 15. Going by your logic, Ohio State (#5) should be in over OU. They won their title game, are 12-1, #5 in the Massey rankings, and 0.06 ahead of Oklahoma.