Tako Tuesdays: The CHAMP Rating 2013

NO ONE SAID THERE WOULD BE MAAAAAAAATH!!!! - USA TODAY Sports

If it's broke, fix it.

Before you read this batch of convoluted nonsense, go to yahoo.com and join ATQ's Tourney Pick 'Em!

Group name: Addicted to Quack

Group ID (because apparently that's a thing): 157884

Password: scottmffrost (That's Scott MF Frost, all lowercase)

If, somehow, your cat picks all the right games and you win our pool, you will win a one-of-a-kind, autographed, and framed copy of Dave Hasn't Told Me What the Prize Is Yet T-shirt of your choice (up to a $25 value) from the Duck Store! OMG, aren't you excited?!? You can find the full contest rules HERE.

In 2011, I created the CHAMP rating as a way to use statistics as a way to properly curate a winning NCAA bracket. The original CHAMP rating had all four #1 seeds advancing to the FInal Four; 2011 happened to be the year #11 seed VCU joined #8 Butler, #4 Kentucky, and #3 UConn in Houston. CHAMP Rating 2: Electric Boogaloo featured three #1 seeds and #2 Missouri headed to New Orleans. Only eventual champion Kentucky (whom CHAMP did not pick to win it all) made it there, as North Carolina and Syracuse fell in the Elite Eight, and Mizzou dropped their first game to #15 seed Norfolk State. If you're keeping score at home, that's 1 for 8 on picking Final Four teams, and 0 for 2 on picking a champ.

But I'm no quitter. It just needs retooling. And on Selection Sunday this year, I realized a major flaw in the system: in the first two iterations of the CHAMP rating, there were no defensive statistics involved. I'd basically been ranking the best offensive teams in the country, and completely ignoring the phrase "defense wins championships". Make no mistake, I am not very smart. And so for the first time, defense gets its due in the CHAMP rating.

Here is what this year's CHAMP rating is valuing:

Wins - Last year, Kentucky had the most wins of any team in the tournament, and won it all. Clearly, this is foolproof.

Conference Strength - A new feature, teams get points based on how many total teams from their conference made the field of 68. The thinking behind this: the deeper the conference, the more battle-tested the team.

3-point shooting % - Teams that make their threes can make a deep run in March. Teams that can't have less room for error.

Free throw shooting % - Make them. They're free. Ask Derrick Rose: the charity stripe can be important this time of year.

Assists/field goals made - Last year I used AST/TO ratio to measure a team's cohesiveness and reliability. The problem was, it has the largest spread of any of the stats used, and I felt it had too big an impact on the final standings. By switching to assists per field goals made, I'm valuing teams with active ball movement, that use passing to generate open looks at the basket.

Rebounding rate - In past years, I've used total rebounds per game, which I think skewed the data towards teams that take, and give up, lots of shots. By using rebounding rate (the % of missed shots that a team successfully rebounds), teams that rebound efficiently are now getting the big points.

Field goal percentage - Though it's much less weighted than in past years, on account of also using three-point and free throw percentages.

Steals - Force turnovers, generate momentum and runs. Generate momentum and runs, win tournament games.

Defensive efficiency - The only real stat that can (kind of) measure how well a defense does at the most important part of defense: preventing points. DE measures the number of points allowed per 100 possessions; in order to make the smaller numbers worth more, I took the team DE, which ranged from about 80 to 110 points per 100 possessions, and subtracted it from 200.

Winning streak - I'm a big believer in momentum, and riding a winning streak into the tournament can sometimes provide a boost. Just ask 2011 UConn, they'll back me up.

Using math, I weighted each category, keeping the three shooting categories weighted somewhat less than the rest as to not overvalue shooting. The end result looks like this:

(Wins / 2) + (Total conference bids / 2) + (three-point shooting % x 20) + (free throw % x 18) + (assists per FG made x 20) + (rebounding rate x 25) + (field goal % x 15) + (steals per possession x 5) + ((200 - defensive efficiency) / 10) + (current winning streak / 4) = CHAMP

You can find all the data HERE, if you're willing to dive down that rabbit hole. Some interesting bullet points:

  • The four #1 seeds hold the top four CHAMP ratings, which isn't terribly surprising. But the teams that follow are surprising: #6 seed Memphis, #3 New Mexico, #7 Creighton, #8 Pittsburgh, and #14 Davidson.
  • In addition to Davidson, double-digit seeds Middle Tennessee State and Belmont find themselves in the top 20 of the CHAMP rating.
  • High seeds the CHAMP rating hates: #4 seed Michigan (#26), #2 seed Duke (#27), #2 seed Miami (#40), and #5 seeds Oklahoma State and Wisconsin (#s 42 and 43, respectively).
  • Thanks to their strong defensive and rebounding numbers, Oregon comes in at #23 in the CHAMP rating, the highest rating in the Pac-12. The CHAMP rating has them defeating Oklahoma State, but losing to Saint Louis, who rank #17 in this rubric.
You can view the full CHAMP Rating bracket HERE, and scoff at the idea of Davidson, and their current 17 game winning streak, making the Elite Eight, and without Steph Curry!
X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join Addicted To Quack

You must be a member of Addicted To Quack to participate.

We have our own Community Guidelines at Addicted To Quack. You should read them.

Join Addicted To Quack

You must be a member of Addicted To Quack to participate.

We have our own Community Guidelines at Addicted To Quack. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9347_tracker