To figure out a playoff, I’d first look at what I would have done in the past.
2011: 4 teams: LSU, Oklahoma State, Alabama, Stanford
2010: 2 teams: Auburn, Oregon
2009: 4 teams: Alabama, Texas, Cincinnati, either Boise State or TCU
2008: 4 teams: Florida, USC, Utah, Oklahoma
2007: 8 teams: Ohio State, USC, LSU, Virginia Tech, West Virginia, Oklahoma, Georgia, Missouri
2006: 2 teams: Ohio State, Florida
2005: 2 teams: USC, Texas
2004: 4 teams: USC, Oklahoma, Auburn, Utah
2003: 4 teams: LSU, USC, Oklahoma, Michigan
2002: 2 teams: Ohio State, Miami
2001: 4 teams: Miami, Oregon, Nebraska, Florida
2000: 4 teams: Oklahoma, Florida State, Miami, Washington
1999: 2 teams: Florida State, Virginia Tech
1998: 4 teams: Tennessee, Florida State, Ohio State, UCLA
Out of 14 years, 5 worked with 2 teams, 8 would have been best with 4 teams, and 1 was so messed up it needed 8 teams.
Here is where we hit a snag though. If the 4 team playoff is chosen by computers and coaches polls, we would have gotten some awful choices.
2008 is a particularly egregious example. USC and Utah should have been in the playoff, but the actual top 4 was Florida, Alabama, Oklahoma, Texas, giving us 2 pairs of teams that had already played. Note that Utah absolutely massacred Alabama that year.
I think a selection committee might have chosen correctly, so I would love to give a 4 team playoff with a committee a shot. If we stick with computers and polls, I’d rather have 8 teams because computers and polls are downright horrible.