Bubble Watch / Bracketology

[longtimefan]

Indiana St out with a top 30 NET. Oklahoma and Providence out with 9 quality wins and no bad losses. St Johns out with 10 quality wins and one bad loss. Virginia the last team in with 10 quality wins and no bad losses.

We were nowhere close to the bubble after all those bid steals.

The losses to West Virginia and OK State killed us.

[bearcatbret]

The losses to West Virginia and OK State killed us.

And blowing that double digit lead against OK last week.

[swilsonsp4]

I realize that the committee members watch a lot of games. What did they see about South Carolina, which finished #51 NET and #49 KenPom, that led them to assume that the Gamecocks not only deserved a bid, but a #6 seed? Their OOC schedule was ranked #337 by KenPom. I can only assume that there were copius servings of wine during the committee's periods of sequester.

[sedz]

I realize that the committee members watch a lot of games. What did they see about South Carolina, which finished #51 NET and #49 KenPom, that led them to assume that the Gamecocks not only deserved a bid, but a #6 seed? Their OOC schedule was ranked #337 by KenPom. I can only assume that there were copius servings of wine during the committee's periods of sequester.

6-5 in Q1 and 5-0 in Q2 is a very impressive resume even with a couple bad losses. 8 quality wins away from home. Even their two bad losses were to top 100 teams. I'd say that's right in line with other 6 seeds, who each had 10 or 11 quality wins but also 10 or 11 losses.

Much of their poor metrics is the result of losing to Auburn by a combined 71 points over 2 games. They have the 11th best strength of record in the country though.

[sedz]

And blowing that double digit lead against OK last week.

We only had 7 quality wins. Even if we eliminate both bad losses and win the OK game, we're still a couple wins short of Virginia for the last spot. We'd also need to pull out say the early January Texas and Baylor games. We weren't close.

This year had probably the strongest bubble ever because of all the bid steals. There are 10 seeds playing in Dayton.

[sedz]

bracketmatrix has compiled their rankings for this year.

Dave Ommen at Bracketville was the best performer of the most well known bracketologists, finishing #5 out of 216 with 45 teams seeded correctly and only 27 missed by multiple seed lines. Perennial high achiever 131sports also did well, finishing in the top quartile at #54. Glazewskitology finished in the top 3 for the second in year in a row, their only two years submitting brackets.

Lunardi again came in below average, ranking #166. He only had 38 out of 68 teams seeded correctly and missed 29 teams by multiple seed lines. Jerry Palm of CBS ranked #213 out of 226, and several of the brackets worse than his are just algorithms. Palm had 35 out of 68 teams seeded correctly and missed 33 teams by multiple seed lines. I have no idea how that guy keeps his job. He's not even a good communicator like Lunardi is.

Torvik's algorithm did very well, seeding 43 teams correctly and missing 28 by multiple seed lines for a top third ranking at #69. That's a computer model that updates in real time outperforming most bracketologists including Lunardi.