Separating top seed contenders and pretenders

Now that the 2014 Seed Guide is behind me, I’m starting in on the research for picking your Final Four and champion. The first step in that analysis is identifying the statistics that separate Final Four contenders from pretenders as every seed position. In fact, I do two types of “contender/pretender” studies to help you fill out your bracket: one for the Insiders “Final Four/Champ” piece and another separating PASE over- and underachievers at every seed position.

For top seeds, that analysis is one and the same. That’s because the average top seed wins 3.4 games per dance. So overachievement for a one seed means getting to the Final Four; anything less is underperformance against seed expectations. To give you some idea of  what this analysis entails, let’s look at the eight factors that separate top seed pretenders from contenders.

Beware of one seeds with any of these conditions:

  • A margin of victory lower than 10.5 points per game
  • A “snake-bit” coach with more than four tourney trips and no Elite Eight runs
  • A coach with fewer than seven tourney trips and either a team that didn’t go to the previous dance or one with no All-Americans
  • A combined rebounding/turnover advantage lower than five possessions
  • An assist-to-field goal percentage higher than 66%
  • Imbalanced scoring with less than 27% of points coming from either the frontcourt or backcourt
  • A strength of schedule weaker ranked than 80
  • A KenPom possession-based ranking lower than fifth, signaling undeserved seed placement

There have been 39 top seeds that owned one or more of these disqualifiers. Only one of them, Rick Barnes’ 2003 Texas squad, made it to the Final Four. That’s a 2.6% semifinal advancement rate—about 16 times worse than the standard rate at which top seeds make the Final Four. What’s more, only 18 of these 39 teams, or 46.2%, even reached the Elite Eight. Overall, these one seed pretenders underachieve at a dismal -1.097 PASE rate, winning just 2.3 games per dance. That’s more than one game per tourney below expectations.

Meanwhile, the 77 top seeds that were free of any of these disqualifiers made the Elite Eight 62 times (80.5%) and reached the Final Four 46 times (59.7%). Altogether, they own an overachieving PASE of +.556—and average 3.9 wins per tourney. Oh yeah…and all the 18 one seed champions are among these 77 contenders.

Last year, knowing these stats would’ve been a big help in filling out your bracket. Two top-seeded squads, Gonzaga and Kansas, had at least one sign of a pretender. Neither of them reached the Elite Eight. Indiana also failed to win three games…even though they weren’t a pretender by the numbers. And Louisville, of course, cut down the nets.

Within the next couple weeks, keep your eye out for two Insider feature pieces that help you sort out overachievers from underachievers at every seed position. “Final Four/Champ” will identify the attributes that separate semifinal candidates from longshots.  “Contender/Pretender” will separate each seed into teams that beat and fall short of their projected win totals.

This entry was posted in Bracket Tools, Key Factors. Bookmark the permalink.

2 Responses to Separating top seed contenders and pretenders

  1. BUCats says:

    Pete – How much do you change these attributes from year to year? I was looking at last year’s set of attributes for one seeds and both Gonzaga and KU didn’t have any blackmarks:

    *A rookie tourney coach leading a team that didn’t go to the previous dance
    *Fewer than eight wins and a losing streak in their last 10 pre-tourney games—with no All-American to reverse momenutm 
    *An offense that generates fewer than 72.1 points per game 
    *An average scoring margin less than 10.6 points per game 
    *A soft defense allowing more than 82 points a game 
    *A “guard-weak” team getting less than 25% of its points from the backcourt. 
    *A rebounding/turnover margin less than four possessions (remember to subtract turnover margin, since a negative margin is favorable) 
    *A strength of schedule weaker ranked than 80. 

    I firmly believe there are a set of attributes that will point out the underachievers, but how much do you think is curve fitting?

    • ptiernan says:

      While I do adjust these, I used the same disqualifiers for one seeds that I used in the Final Four/Champ article from last year. This one may be an earlier version. I think the snake-bit issue DQ’d Few. And I’d have to go back to see what got Kansas. No doubt, though, I do tinker…and there’s definitely a little curve fitting going on. Overall, though, I’d steer clear of lower scoring, lower margin one seeds with inexperienced or unsuccessful tourney coaches. Those, to me, are the big disqualifiers.

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>