Final Forecasting

Originally Published: March 1, 2004
By By Pete Tiernan
Bracketology Research Analyst
Anyone who's ever filled out a bracket for an NCAA tournament pool knows that it gets harder to pick winners with each succeeding round. For one thing, the number of "gimmee" games (like 1 vs. 16 seed mismatches) disappears as the matchups get tighter. For another, as the teams you pick to advance get knocked off, there are fewer games in which you even have the actual matchup on your sheet.

As the various possibilities pile up, the odds the schools you've picked for the Final Four will in fact make it there grow staggering. Even if you played it safe and chose all the top seeds, chances are that only two will get there. Since the tournament expanded to 64 teams in 1985, two No. 1 seeds have made it to the Final Four 11 of 18 times. Three No. 1 seeds have reached the semifinals in just three of 18 tournaments. And all four one seeds have never made it to the Final Four.

As hard as it is to forecast the Final Four, it's critical to do a halfway decent job of it if you want to win the office pool. That's because most pools award higher points for correct predictions in each successive round. Some pools just add a point per round, so that a correct Final Four pick is four points. Others double the points, making a correct Final Four prediction worth eight points. The higher premium on the later round picks, coupled with the bewildering difficulty of getting those picks right, goes a long way toward explaining why people who don't know Duke from Duquesne always seem to come out of nowhere to win the office pool.