I keep railing against the Playoff Predictor and its occasionally insane pronouncements, but only because overall it seems like a great idea: let fans enter their idea for what happens, and they get the odds for their favorite team making the Playoff (or at least, the 15 teams you're allowed to choose from).
Most of its results are reasonable, even obvious, but every iteration has something ridiculous. This time around—post week 3—here is the most glaring problem:
- Washington goes 13-0; Predictor's odds: 66%
No 13-0 Power Five team has missed the Playoff. Only 2 1-loss Power Five conference champs have missed the Playoff so far.
The only way a 13-0 Power Five team misses the Playoff is if there are 5 13-0 Power Five teams. (Notre Dame and BYU have losses already and aren't an issue)
The odds of having 5 13-0 Power Five teams are exceedingly low, much lower than the 34% chance All-State/ESPN gives for a 13-0 Washington to miss the Playoff.
Even a 12-1 Alabama wouldn't be ahead of a 13-0 Washington for the 4th spot. Maybe 5% of the time at most, if Washington struggled with most of its opponents.
I'd like to see how a 13-0 team is left out 1/3 of the time. What are these scenarios? I take it in most of them have 2 SEC teams, so only three 13-0 teams are needed. By the FPI's numbers, the odds of Georgia, Ohio State, and Clemson going 13-0 is around 1%. So apparently the Predictor thinks that a lot of 12-1 teams would finish ahead of a 13-0 Washington, even for that 4th spot.
It wouldn't happen. Washington's odds at 13-0 should come out to at least 95%, and 99% is more realistic.
/End rant
Comments