Monday, March 13, 2017

Complete Final Four Analysis

My Picks:

2. Duke over 1. Gonzaga
1. North Carolina over 4. Purdue

2. Duke over 1. North Carolina

Don't get too cute: Want to pick a 10 seed to the Sweet 16 or a 4 seed to the Final Four? Have some fun, particularly when you are in a pool with a lot of contenders or where upsets are rewarded. But I generally recommend staying away from low seeds for your title game teams, because you really risk having your bracket blow up in your face during the opening weekend. The fact is that the Final Four is likely to be heavy on 1 and 2 seeds, so don't fight it. If some 11 seed makes the Final Four, the odds are that you aren't going to get lucky enough to pick that team anyway.

Look for balance: In general when picking Final Four teams, look for teams that are in the Top 20 of both Pomeroy offensive and defensive efficiency, or at least something right around there. While some unbalanced teams have managed to get through to a championship by being incredible on one side of the ball, history says that you want to pick balanced teams, as they are less likely to be undone by a single thing going wrong. An example team that I would stay away from, for example, is UCLA.

Bracket difficulty matters: When two teams you are deciding between seem like relatively equal opponents, look to see who had the easier path to get there. Which team is more likely to get taken out early? In my bracket, that led me to stay away from Villanova, who are likely going to have to face two very difficult opponents (Wisconsin and then either Florida or Virginia) before they even get to an Elite 8 game. In contrast, North Carolina and Duke were given total cakewalks to the Elite 8. It's not a coincidence that those are the two teams I have in my title game.

Who is the best team?: Gonzaga is #1 in most metrics, though it's fair to assign a slightly larger statistical uncertainty to them than to a team that has faced a more consistently difficult schedule. Villanova is the defending champs and might just be the best team in the nation, though as I said earlier their bracket is a nightmare. The most talented roster of all, and the team that is finally humming at the level that we were expecting all season long, is Duke. There is such a ridiculous amount of talent and athleticism there that I was leaning toward taking them as my national champion pick even before the brackets were released. The fact that they have a relatively soft path to the Elite 8 and that the 1 seed in their bracket has a greater than 50% chance of failing to make the Elite 8 in most computer models seals it for me. Duke is my pick.

3 comments:

Anonymous said...

Can you dig deeper on Duke's "cakewalk", especially compared to Nova? Nova's potential opponents (Wisconsin, UVA/FLA) are a little better than Duke's (Marq/So Car, Baylor/SMU), but not so much so that I would grade Duke's road as being markedly easier, especially with Duke's inconsistency this year. If I'm not mistaken, Duke's most likely Sweet 16 opponent (SMU) is actually ranked ahead of Duke in Pomeroy's ranking.

Jeff said...

There are reasons to be skeptical of SMU's high computer ratings, as I discuss in that region's breakdown. And SMU has to get to that game first, while Villanova is practically guaranteed a KenPom Top 10 opponent in the Sweet 16. And Wisconsin is a team that was borderline Top 10 until a three week goofy funk midseason - they are a far more difficult opponent than Marquette or South Carolina.

Anonymous said...

Illinois is ready to write a very large check. You don't think making any of those coaches Top 5 highest paid NCAA coaches would make them attainable? I think 6 years $30 is a possibility for the right guy. Most of those coaches don't get paid half of that. Some only make about 25% of that. Illinois is a top 20 job in the country easily.