:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/30257815/164966823.0.jpg)
Back in 2011, I set out to revolutionize the art of the bracket with a formula I called the CHAMP Rating. That original system went 0-for-4 on picking Final Four teams. A lesser man would have scrapped the whole idea right there. But I am not a lesser man. I pressed on to 2012, where a slightly-tweaked CHAMP Rating correctly predicted Kentucky to advance to the Final Four. SUCCESS! That's enough of a result for further testing! Last year, the CHAMP Rating incorporated defensive stats into the fold for the first time, and predicted all four #1 seeds to the Final Four. Only Louisville made it there, but the CHAMP Rating correctly picked the Cardinals winning it all. CHAMP may be 2/12 in picking Final Four teams, but from a shutout in 2011 to picking the winner in 2013 is what I call progress. So we press on to 2014. Here's what CHAMP is using this year:
- Wins, slightly less weighted than last year. I'm also including bonus points for wins over the RPI top 25.
- In lieu of using both two and three-point FG% separately, I've combined the two into one category by using a team's True Shooting Percentage.
- Rebound rate, which measures the percentage of missed shots rebounded, is back, because it's fancy and I like it.
- Defensive efficiency rating is back as well, because we need some defensive influence. It's based on the expected # of points allowed in 100 possessions. Fewer than 100 points/100 possessions is good. In order to fit it with this points system, I inverted that number around 100, i.e if a team's defensive rating is 95.6, I'm using 104.4 as my starting point. To give the numbers more of a spread to benefit good defensive teams (and to punish bad ones), I'm doubling the number away from 100. So that 104.4 number becomes 108.8. Divide by 10, and 10.88 points get added to the CHAMP Rating. It's complicated, but I'm really smart.
- Assist to turnover ratio is in, because tournament teams need to take care of the ball.
- Free throw percentage, in addition to being slightly included in true shooting percentage, is included on its own on a less-weighted scale. Gotta make free throws to win in March, right Derrick Rose?
(Season wins/2) + (# of wins vs. RPI top 25) + (True shooting percentage/5) + (rebound rate/5) + (Adjusted defensive efficiency, tweaked a bit) + (AST/TO ratio x 8) + (free throw percentage/10) = CHAMP
| Low seeds the CHAMP Rating likes | High seeds the CHAMP Rating doesn't like |
| #7 seed New Mexico (the #10 overall CHAMP team) | #8 seed Colorado (the #52 overall CHAMP team) |
| #12 Stephen F. Austin (#15 overall CHAMP) | #8 Kentucky (#48 overall CHAMP) |
| #9 seed Pittsburgh (#18 CHAMP) | #5 seed Oklahoma (#38 CHAMP) |
| #14 seed UNC Central (#25 CHAMP) | #6 seed UMass (#39 CHAMP) |
| #12 seed Harvard (#28 CHAMP) | #1 seed Virginia (#12 CHAMP) |
| #8 seed Gonzaga (#19 CHAMP) | #2 seed Villanova (#16 CHAMP) |
| #14 seed Mercer (#34 CHAMP) | #5 seed VCU (#36 CHAMP) |
| #4 seed Louisville (#2 CHAMP) | #2 Kansas (#17 CHAMP) |
| #3 Duke (#4 CHAMP) |
|
The Midwest region is scary, with four of the top eight teams in the tournament. The East region is not so much, with only one team (Michigan State), in the CHAMP's top ten.
Finally, here's the CHAMP rating's Sweet Sixteen and beyond:
South
#1 Florida over #4 UCLA
#7 New Mexico over #3 Syracuse
#1 Florida over #7 New Mexico
***
West
#1 Arizona over #4 San Diego State
#3 Creighton over #2 Wisconsin
#1 Arizona over #3 Creighton
***
East
#4 Michigan State over #1 Virginia
#3 Iowa State over #2 Villanova
#4 Michigan State over #3 Iowa State
***
Midwest
#4 Louisville over #1 Wichita State
#3 Duke over #2 Michigan
#4 Louisville over #3 Duke
***
FInal Four
#1 Florida over #4 Michigan State
#1 Arizona over #4 Louisville
#1 Arizona over #1 Florida
***
You're welcome.