If a Sunday final includes a team not expected to be in the field, then the committee has multiple scenarios (see 1999 tournament, Illinois). If every team involved Sunday is already planned into the field, the Sunday games make minimal difference. The only time I remember a team seemingly skyrocketing from a conference tournament performance was Maryland a couple years back, where they went from on the bubble to a 4 seed (which turned out to be unreasonably high, since they almost lost in the first round and did get bumped in the second). EDIT: Part of the reason the seeding doesn't change much is because of the pod system. Once the committee has teams arranged by pod, so that higher seeds don't get screwed out of their location, then to move even 2 teams around could offset entire brackets. Imagine even a scenario where an Ohio State-Illinois Sunday matchup were to decide which was a one seed and which was a 2 seed. However, Illinois had been slated to go to the Auburn Hills pod as a 2 seed, and Ohio State had been slated to go to the Dayton pod as a 1 seed. If Illinois won, then, what do you do? If you move Illinois to a 1 seed and moved them to Dayton with OSU to Auburn Hills, there might be a geographical disadvantage for Illinois/OSU that there wasn't in the previous pod setup. If you kept them in their pods and just put them against the other teams, there might also be geographical disadvantages (say Kent State won the MAC tournament and was slated to be in OSU's pod at Dayton as a 9 seed, but getting moved to Auburn Hills in Illinois' pod puts them closer than Illinois). There are just too many variables present with the pod system to have a game 20 minutes before the presentation affecting something as trivial as seeding.