FanPost

Linear Statistical Prediction of Sweet 16



Adhering to the professional mantra, "KISS:  Keep it Simple, Stupid" I elected to run a very elementary statistical regression on the Sweet 16 match-ups based on offensive and defensive vectors of winning teams, factoring parity in as an assumption (presumably flawing the model), and failing to account for losing teams' vectors.  I wonder how well it lines up with the coming reality?

East Region

Sweet16: (1) UK 81, (12) Cornell 73.75 ; (2) WVU 71.75, (11) Washington 67.75

Elite 8:  (1) UK 74.625, (2) WVU 70.5

 

South Region

Sweet16:  (1) Duke 66.5, (4) Purdue 58 ; (3) Baylor 70.75, (10) St. Mary's 70.5

Elite 8:  (1) Duke 67.5, (3) Baylor 61.625

 

Midwest Region

Sweet16:  (9) UNI 72, (5) MSU 72 -Going with UNI due to injury - ; (6) Tenn 65.5, (2) OSU 68

Elite 8:  (9) UNI 65.4167, (2) OSU 69.3333

 

West Region

Sweet16:  (1) Cuse 69.25, (5) Butler 63 ; (6) Xavier 67.5, (2) KSU 67

Elite 8:  (1) Cuse 70.71, (6) Xavier 64.5833

 

Final Four:

(1) UK 70.28, (1) Duke 68.7825

(1) Cuse 69.235, (2) OSU 67.1165

 

National Championship:

(1) UK 73.95565, (1) Cuse 71.92675

 

Analysis:

*The winning teams' performances to date favor the better seeds progressing further into the tournament, and further supports most analysts' views on the favorites.  If we eliminate the assumption of parity or regress against losing teams' vectors, then this would completely destroy the model in a way I'm too lazy to comprehend, since I don't get paid to do this (and am neglecting selling things, which would get me more money, in order to write this... i.e. it's really my own personal investment in my entertainment interests).

 

*Factors such as injury/return-from-injury of star players should have far more impact than the simple numbers produced, and will completely render this model even more statistically irrelevant than it is today (as does the preeminence of high caloric fast food burrito chains).

 

*I was inconsistent with some decimal points there.  In the words of the famous Michael Bolton (no relation), "I always do something stupid like that, I'm always messing up some mundane detail."

 

*My gut check picks on the Sweet 16 are mostly in line (75%) with what the numbers provided:

          Agree:  UK, WVU, Duke, UNI, OSU, Cuse

          Disagree:  Baylor, Xavier

 

*For the Final Four, I have a hard time arguing against the case that's made here, as it seems reasonable... but it also seems to be close enough to say it's a complete crap-shoot. 

 

 

CONCLUSION:  Overall, this tells me that all of these games are built to be extremely tightly contested match-ups, which leads me to make the following statement (CBS, if you're listening, I'll gladly accept an endorsement check):   THIS will be a MUST-WATCH TOURNAMENT.

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,

Join Troy Nunes Is An Absolute Magician

You must be a member of Troy Nunes Is An Absolute Magician to participate.

We have our own Community Guidelines at Troy Nunes Is An Absolute Magician. You should read them.

Join Troy Nunes Is An Absolute Magician

You must be a member of Troy Nunes Is An Absolute Magician to participate.

We have our own Community Guidelines at Troy Nunes Is An Absolute Magician. You should read them.

Spinner

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