The NCAA Tournament Bracket Mathematical Formula revealed

So it’s that time!!!  March Madness is back and College Basketball mayhem will soon occur beginning tonight.  Well, by Thursday at least.  Time for the office pools, being distracted and lazy at work, and not doing your job to the best of your ability because you have stattracker on your computers.  This year has been interesting.  Many teams have been jumbled in the top ten all season long.  I think we all expected Syracuse to be in the tournament, but I think it’s sufficient to say that they were a total surprise finishing the year off as a #1 seed.  I didn’t get a lot right in my pre-season rankings (I do admit however, that I had UNC ranked lower than all of the other publications, although I did not expect such a horrid season…I will press on) but I had Kentucky ranked a lot higher than most publications.  There were many that didn’t even have them in the top 5.  I had them at #4 to begin the season.  Calipari may be a cheater, but the guy can recruit and coach.  Every place he has been (minus his short NBA stint), he has risen the program from the ashes…UMASS, Memphis, and now Kentucky.  He then proceeded to leave those programs in dissaray.  But Kentucky is a dream job.  And gosh, he looks and talks smooth.  He’s got that New York swagger.  A lot of people may compare this team to his 2008 Memphis team, with John Wall being the heir apparent to Derrick Rose, Joey Dorsey and Chris Douglas Roberts being Demarcus Cousins and Patrick Patterson.  They went to a Final Four.  Many experts see Kentucky going to the Final Four and cutting down the nets.   They got some athletes.  They’re good.  And I do see them as a top 5 team.  But in my bracket, I have them losing to Wisconsin in the Sweet 16.  This is why…

I’ve spent the last couple of weeks working on a mathematical formula in best determining the most well rounded teams with the best chance in winning games.  It will not be perfect and you can criticize me if you feel the need.  Bill James has also been criticized.  And Theo Epstein hired the guy.  This formula is not the savior of College Basketball bracketology, but I see a strong correlation in my point system and the best teams.  Some of my statistics appear skewed, but I made ammends to them.  My top 10 teams based on these numbers are Wisconsin, Kansas, Duke, Syracuse, Ohio State, West Virginia, Notre Dame, Maryland, Marquette, and BYU, in no particular order.  Some of the teams out West like BYU, Utah State, New Mexico, and UNLV had very strong numbers.  And although I did base my formula on RPI, Strength of Schedule, and Conference Power ratings,  these numbers were skewed due to ultra weak conferences and schedules.  Other teams like Minnesota, Temple, St. Mary’s, Missouri, Pitt, Purdue, Georgetown, Michigan State, and Old Dominion also received high numbers.  On average, the entire Big Ten displayed a very strong TO/Assist ratio rating as well as Offensive and Deficiency rating, thus the reason Wisconsin, Ohio State, Purdue, Michigan State, Minnesota, and Purdue rank so high.  They simply take care of the basketball.  And that’s a very important quality.  Kentucky has great numbers in Offensive (17th)  and Defensive Efficiency (21st), but they rank 125th in the nation in To/Assist ratio and were only 5th in their Conference in that category following Florida, Tennessee, Ole Miss, and Alabama.  Some say talent is a better indicator of Championship teams.  Some would say experience.  But being an ACC fan, I remember the ’94 UNC team.  Carolina had just won the ’93 National Championship and still had Derrick Phelps, Erik Montross, Donald Williams, and Brian Reese (4 starters on the previous team), with additions of Jeff McInnis, Serge Zwikker, and dynamite freshmen, Jerry Stackhouse and Rasheed Wallace.  They lost in the 2nd Round.  Still good all around, their TO/Assist ratio was not as good in ’94 when compared to ’93.  The same can be said for the 2006 Duke team.  J.J. Redick, Shelden Williams, Lee Melchionni, and Sean Dockery were all seniors.  Demarcus Nelson was a sophomore, with highly touted freshmen contributors Greg Paulus and Josh McRoberts (who needless to say, underachieved).  This team was #1 for much of the year and I had pegged them to win it all.  But guess what?  Although they didn’t rank 125th in TO/Assist ratio (as Kentucky), it was still was not sparkling.  And they got knocked out…

So needless to say, if you don’t turn the ball over and excel in precision passing and sharing the basketball, your team is at better odds to win.  Basketball 101.  Now Offensive Efficiency.  This statistic is not merely how much you score, but it is is the average amount of points scored by a basketball player per shot taken. This includes missed field goals as well as free throws.  Or…Possessions = .96 * (FGA − ORb + TO + (.44 * FTA).  The .44 accounts for the fact that when a player scores a basket and is fouled, they shoot a free throw, which is not a possession.  The .96 multiplier adjusts for team rebounds. Because these are not considered offensive rebounds, the formula slightly overestimates the number of possessions per team without the multiplier.  Defensive Efficiency is formulated relatively close to this, but in reverse, applying it to the opponent. 

So, going through the bracket, I compared my final numbers.  And I’m not stupid, I do believe RPI, Strength of Schedule play a part in determining seeds, so I also included those numbers as well as Average Scoring Margin (how much you beat teams by) and Away Power Ratings (how well you play away from home).  This season’s Duke team did not win an away game vs. a Top 25 team, but they also didn’t lose a game at home.  And everyone plays away in the NCAA tournament, so it balances out.  This may seem a contradiction and may make the Away Power Ratings irrelevant to you, but there is a difference in Duke playing @ Kansas and Duke playing @ Madison Square Garden vs. Penn State. 

So that being said, tomorrow I will have my picks of winners.  I don’t think it will surprise everyone too much.  There are no drastic surprises.  A few minor ones.  The ACC does not fare well besides Duke and Maryland.  Big 10 surprises.  One region actually has no upsets based on seedings AT ALL!  One region has minor upsets going on all over the place in the 2nd and 3rd rounds.  But these are like 12 over 5, 10 over 7, and 9 over 8.  Nothing crazy, but there will be surprises.  I wish I knew Onuoku’s status from Syracuse.  That could change a thing or two.  The search of finding the most efficient team begins here.  Stay tuned for tomorrow.