Update on bracket model building progress

I’ve now completed eight models for filling out your brackets:

A – From the Gut
1 – KenPom
2 – Final Four/Champ
3 – Outcome Matching
5 – Pulse Check Stats
6 – Factor PASE
7 – Baseline
10 – ESPN BPI

By Monday night, I hope to include at least five more. My wife (bless her heart) noticed a mistake on the Wave 1 PDF. Under the Baseline model, I accidentally copied the Final Four/Champ bracket. That’s been corrected in this wave.

For those keeping score, so far we have six different champs from the eight brackets. Arizona has three–and Florida, Louisville, Villanova, Duke and Wichita State all cut down the nets in one of the models.

This entry was posted in General News. Bookmark the permalink.

12 Responses to Update on bracket model building progress

  1. John says:

    I’m probably wrong, but according to the final four model, shouldn’t Mich St advance to the final four because they fit the criteria for a 4 seed (who are a terrific 9-2 in reaching the final four).

    They definitely pass the “eye test” of a 4 seed making the final four – similar to UK a few years ago and unlike a shaky K-State team last year.

    Also, what are y’alls thoughts on UNC? I like them but damn they’ve got a tough draw

  2. Matt says:

    The PDF has Virginia in the Final Four even though Mich. St. beats them in the Sweet 16.

  3. Jared says:

    I’m loving all these models. Is there any way that we could open some bracket forums to get some discussion going?

    Thanks again Pete.

  4. Carlos says:

    Where do you guys get these criteria ‘s for seeds? Where on this site?

  5. Scott says:

    Pete-

    I’m confused about the Final 4 / Champs strategy after “Advance any other Final 4 qualifiers as far as possible. I don’t understand how to eliminate key underachieving outliers before their seed expected win totals… If Kansas meets the criteria of an underachieving outlier (which I believe they do), do I stop them short of the Elite 8? but if Syracuse meets the criteria as well (which I believe they do), do I then treat the Kansas / Syracuse matchup as an upset / tossup? Also- do I do the same thing- identify the underachieving outlier in the first round matchups as well? or go straight to upset / tossup criteria?

    Thanks Pete!

  6. Jay says:

    Based on the seed match-up guide and the team stats excel file, below are the teams that I found to meet the criteria for each seed match up (both victim & Cinderella) in round 1. I have excluded the 1-16 match-up. I may have made some errors so corrections are encouraged.

    2 vs 15: no 2 seed victims or 15 seed cinderella
    3 vs 14: 3 seed victim: iowa st. & Creighton – 14 seed cinderella: Louisiana Lafayette & Mercer
    4 vs 13: 4 seed victim: UCLA & San Diego St. – 13 seed cinderrella: New Mexico St.
    5 vs 12: 5 seed victim: Cincinnati – 12 seed cinderella: ND st meets both criteria, NC St, Xavier & Harvard meet 1 criteria
    6 vs 11: 6 seed victim: OSU & UMass – 11 seed cinderella: none
    7 vs 10: 7 seed victim: none – 10 seed upset: none

    • Ryan Tressler says:

      Creighton is not a 3 seed victim, as they do not meet the scoring margin % requirement (scoring margin/oppg), creighton’s is 17.9%, the benchmark is less than 14.5%.

      San Diego St is not a 4 seed victim as their win% is over .840 (its .879) and their margin % is over 17% (its a whopping 24.7%)

      North Dakota State actually doesnt meet EITHER of the criteria for a 12 seed cinderella

      Dayton actually does meet the criteria for an 11 seed winner, as they score more than 73 ppg (barely at 73.4), their margin % is over 8% (again, barely, its 8.58%), their winning % is over .640 (its .697, so again, barely), and they have at least one senior starter (they have 2)

  7. Michal says:

    On the Outcome Matching model, it would seem that UMass would be upset. They have a higher BPI and KenPom. What’s the reasoning behind not having them be upset? Thanks.

    • ptiernan says:

      Michal – You’re right. I went back and checked the Pythag differences in opponents. I actually missed that UMass’s was negative! I was just looking at the values. Let me fix that quickly. One of the upsets is going to have to go away.

Leave a Reply

Your email address will not be published. Required fields are marked *

* Copy This Password *

* Type Or Paste Password Here *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>