Raymond’s highly scientific predictions for the 2009 NCAA men’s basketball tournament


Once again, it's time for Raymond to come up with an absurd, arbitrary criterion for filling out his NCAA bracket.

This year, the criterion is the school's graduation rate for basketball players. Lower graduation rate wins. However, if the teams are seeded 11 or more positions apart, I'll give the win to the favorite (just to get rid of highly unlikely upsets).

Update:

  • Correct predictions are in green.
  • Incorrect predictions are in red.
  • (!) marks upsets correctly predicted.
  • (*) marks upsets predicted but did not take place.
  • (x) marks actual upsets not predicted.

Opening Round Game

Alabama State (56%) Alabama State
(56%)
Morehead State (57%)

Midwest bracket

1 Louisville (42%) Louisville
(42%)
Louisville
(42%)
Louisville
(42%)
USC
(37%)
16 Alabama State (56%)
8 Ohio State (53%) Ohio State
(53%) (x)
9 Siena (86%)
5 Utah (67%) Arizona
(20%) (!)
Arizona
(20%)
12 Arizona (20%)
4 Wake Forest (100%) Cleveland State
(71%) (!)
13 Cleveland State (71%)
6 West Virginia (41%) West Virginia
(41%) (x)
West Virginia
(41%)
USC
(37%)
11 Dayton (89%)
3 Kansas (64%) Kansas
(64%)
14 North Dakota State (N/A)
7 Boston College (70%) USC
(37%) (!)
USC
(37%)
10 USC (37%)
2 Michigan State (60%) Michigan State
(60%)
15 Robert Morris (100%)

West bracket

1 Connecticut (33%) Connecticut
(33%)
Connecticut
(33%)
Connecticut
(33%)
Maryland
(10%)
16 Chattanooga (34%)
8 BYU (91%) Texas A&M
(47%) (!)
9 Texas A&M (47%)
5 Purdue (77%) Northern Iowa
(67%) (x)
Washington
(50%)
12 Northern Iowa (67%)
4 Washington (50%) Washington
(50%)
13 Mississippi State (53%)
6 Marquette (100%) Utah State*
(100%) (*)
Missouri
(36%)
Maryland
(10%)
11 Utah State (100%)
3 Missouri (36%) Missouri
(36%)
14 Cornell (N/A)
7 California (30%) Maryland
(10%) (!)
Maryland
(10%) (*)
10 Maryland (10%)
2 Memphis (55%) Memphis
(55%)
15 Cal St. Northridge (8%)

East bracket

1 Pittsburgh (69%) Pittsburgh
(69%)
Tennessee
(38%)
Portland State
(17%)
Portland State
(17%)
16 East Tennessee State (60%)
8 Oklahoma State (92%) Tennessee
(38%) (x)
9 Tennessee (38%)
5 Florida State (100%) Wisconsin
(86%) (!)
Portland State
(17%)
12 Wisconsin (86%)
4 Xavier (82%) Portland State
(17%) (*)
13 Portland State (17%)
6 UCLA (46%) UCLA
(46%)
UCLA
(46%) (*)
Texas
(31%)
11 VCU (53%)
3 Villanova (89%) Villanova
(89%)
14 American (31%)
7 Texas (31%) Texas
(31%)
Texas
(31%) (*)
10 Minnesota (36%)
2 Duke (89%) Duke
(89%)
15 Binghamton (100%)

South bracket

1 North Carolina (86%) North Carolina
(86%)
LSU
(40%) (*)
LSU
(40%)
Clemson
(29%)
16 Radford (80%)
8 LSU (40%) LSU
(40%)
9 Butler (92%)
5 Illinois (80%) Illinois
(80%) (x)
Akron
(63%)
12 Western Kentucky (100%)
4 Gonzaga (67%) Akron
(63%) (*)
13 Akron (63%)
6 Arizona State (38%) Arizona State
(38%)
Arizona State
(38%) (*)
Clemson
(29%)
11 Temple (57%)
3 Syracuse (50%) Syracuse
(50%)
14 Stephen F. Austin (42%)
7 Clemson (29%) Clemson
(29%) (x)
Clemson
(29%)
10 Michigan (46%)
2 Oklahoma (55%) Oklahoma
(55%)
15 Morgan State (45%)

Finals

USC (37%) Maryland
(10%)
Maryland
(10%)
Maryland (10%)
Clemson (29%) Portland State
(17%)
Portland State (17%)

*The Marquette/Utah State game is a draw based on graduation rates, so it went to the tie-breaker: players arrested.

Update 9am: Fixed the prediction for the game between LSU and Akron. Thanks, Adam, for pointing this out.

My source for graduation data.

Update 10am: Fixed the prediction for the game between Louisville and Arizona. Thanks, Dave O, for pointing this out.

Update 1pm: Fixed the prediction of the game between UCLA and VCU (and the subsequent matchup against Villanova). Thanks, Crawford, for pointing this out.

Comments (28)
  1. Adam V says:

    Go Horns! I always knew our crappy graduation rates would be good for something someday.

    (Wait, shouldn’t LSU beat Akron in the South Sweet 16?)

    [Oops, thanks. Fixed. -Raymond]
  2. vince says:

    hmmm as a Maryland alumnus I’m not sure if this should make me happy or sad.

  3. John says:

    Wow.  I knew the graduation rates generally weren’t very good, but I did not expect to see so many below 40%.  On the other hand many of these teams send a lot of players to the NBA early, so maybe that’s where the disconnect is.

  4. Nathan_works says:

    I am quite amazed that UNC and Duke have such good graduation rates.. I knew of MD, WashPost did a story on them a while back, and Georgetown.. G-town had miserable rates, and quoted the coach as saying he didn’t care..

  5. ton says:

    Stupid players who read on a 5th grade level  or players leaving early causing low graduation rates is really a detriment not an advantage. Scientific indeed umm no…

  6. Neil (SM) says:

    @ton: Humor detector broken today?

  7. Tad says:

    Some of these numbers appear even worse when you consider that these are 6 year graduation rates, I believe. That is if Raymond is using the same data I remember reading recently.

  8. RobO says:

    "players leaving early causing low graduation rates is really a detriment not an advantage"

    Having players that are good enough to leave to the NBA seems like it would be an advantage.

    Though, I doubt most are leaving to the NBA. There are more players represented here by far than in the NBA.

  9. Andrew says:

    Exactly RobO, schools like Chattanooga, American University and Portland State aren’t losing their players to the NBA.

    I remember a few years back when Nolan Richardson was fired from Arkansas that the graduation rate of his players was a steller 0%.  When asked about it, he said it’s not his responsibility to graduate his players.  The sad thing is, he was probably right, but it still didn’t help and he was fired.

    Love him or hate him, Bobby Knight at least graduated his players, he even had the professors give him attendence reports and he’d punish players who skipped classes.  

  10. Wyatt says:

    What about the Florida State cheating scandal?  shouldn’t that be enough to put them over Wisconsin?

  11. davis says:

    2 years ago the graduation rate for Maryland was 0%, although this didn’t seem to help them win the national championship that year.

  12. JamesCurran says:

    Raymond, I love these prediction, but they don’t seem quite arbitrary enough.

    • The Data points you’ve used (now and in the past) are relatively stable.  If used year-after-year, you’d get roughly the same results.  
    • The Data points you’ve used  are absolutes.  Ignoring the 11-seed rule for a minute, the team with the worst graduation rate (or best paid coach) will always win it all, regardless of who they were matched with along the way.

    So, to be truly arbitrary, we’ll need something that will change completely every year.  I suggest "alphabetical by center’s last name".  

    But this still fails my second test, as "Adams" will pretty much beat anyone and take it all.  So we all another twist : "alphabetical by center’s last name, indexed by round number", so that "Jones" beats "Smith" in the first round (because "J" is before "S"), but "Smith" beats "Jones" in the second round (Because "m" is before "o").

  13. Dave O says:

    Maybe I’m reading the rules wrong, but shouldn’t Louisville (seeded 1) beat Arizona (seeded 12)?

    [Good catch. Fixed. -Raymond]
  14. Jim says:

    Somebody should set up the pool for us to bet on Ray’s prediction!

  15. Kip says:

    Have you ever posted on the accuracy of these predictions?  Or is that an exercise left for the reader?

    [Even I don’t know what the accuracy is. I don’t follow the tournament. I just enjoy coming up with absurd criteria. -Raymond]
  16. ton says:

    I hate to spoil things but :

    “This year, the criterion is the school’s graduation rate for basketball players. Lower graduation rate wins. “

    Louisville: 42%

    Arizona:    20%

    20% < 42% => so the original bracket with Arizona advancing was correct according to the arbitrary rules.

    [On the other hand, 12 − 1 ≥ 11, so the “unlikely upset” rule applies. -Raymond]
  17. tsrblke says:

    @Nathan_works,

    Duke is known for recruiting surprisingly acedemic basketball players.  Most schools get to choose one or the other, but because the world thinks "Duke is da bestest!" they seem to be able to choose both (higher applicant pool basically.)

    But they still don’t deserve a #2 seed (over-rated)

  18. Crawford says:

    In the first round of the East bracket, shouldn’t #6 UCLA (43%) beat #11 VCU (53%)?  UCLA would still beat Villanova (89%) in the second round and lose to Texas (31%) in the third.

    [I need to stop filling our brackets when tired. Thanks. -Raymond]
  19. ton says:

    Oops for whatever reason I mainly concentrated on the first rule the change was correct…

  20. John says:

    I went to UCF.

    UCF stands for "U Can Finish" and "Under Construction Forever", both of which are true.

  21. configurator says:

    How come there are exactly 64 teams?

    Also, why do some states have the word ‘State’ in their name, like ‘Ohio State’?

    ** Not an American, unfortunately **

  22. Tom says:

    @configurator

    Some universities are partially financed by a state, and they are usually called University of <state name> (i.e. Univeristy of Ohio) or <state name> State University (i.e. Ohio State University), which are often shortened to Ohio or Ohio State respectively.

  23. Neil (SM) says:

    @Configurator: In many states there are (at least) two separate universities that use the State’s name.

    In Pennsylvania, for example, there is the Pennsylvania State University (Penn State), a publicly-funded school, and also the University of Pennsylvania (Penn) which is an unrelated private school.

    To make things more confusing, there are other states like Michigan where there is Michigan State, and U of Michigan, and both are public schools.

    So basically the "State" part of the name or nickname will usually used to differentiate from other similarly-named institutions.

  24. Neil (SM) says:

    @Configurator: In many states there are (at least) two separate universities that use the State’s name.

    In Pennsylvania, for example, there is the Pennsylvania State University (Penn State), a publicly-funded school, and also the University of Pennsylvania (Penn) which is an unrelated private school.

    To make things more confusing, there are other states like Michigan where there is Michigan State, and U of Michigan, and both are public schools.

    So basically the "State" part of the name or nickname will usually used to differentiate from other similarly-named institutions.

  25. Adam V says:

    @configurator: Actually, there are 65 now. From WP:

    * 1939–1950: eight teams

    * 1951–1952: 16 teams

    * 1953–1974: varied between 22 and 25 teams

    * 1975–1978: 32 teams

    * 1979: 40 teams

    * 1980–1982: 48 teams

    * 1983: 52 teams (four play-in games before the tournament)

    * 1984: 53 teams (five play-in games before the tournament)

    * 1985–2000: 64 teams

    * 2001—present: 65 teams (with an "opening round" game to determine whether the 64th or 65th team plays in the first round)

    Generally, though, the tournament is that large because it has to be big enough to accomodate every conference’s champion, plus a large number of "at-large" teams who the committee feels have earned a spot. There are thirty-one teams that get in just by winning their conference, which leaves thirty-four "at-large" spots.

    As for why some universities have "State" in their name… couldn’t tell you.

  26. Richard G says:

    @configurator There are 65 teams (yes, 65, check the opening round game) because that’s how many teams qualify for the tournament.

    There are 347 institutions that are represented by teams in men’s Division I basketball; the top 65 reach the tournament based on their full season’s performance.

    The names are the names of the universities they represent – many universities are named for the state they are located in (and often the state funds the university), and in many cases there is more than one such university, and they are named XXX and XXX State.

    The names that are conventionally used in sport are not necessarily how that university is know to academics – the best example there is that University of California, Berkeley is known as Cal in sports, but generally Berkeley in academics.

    Note that this is nominally-amateur sport played by undergraduate students.

  27. Mark says:

    Since there are effectively 64 teams, why not use an LFSR?

  28. disappointed says:

    I like the concept of the tie-breaker for Marquette v. Utah State, but not the execution. The kid on the Utah State team was arrested (primarily) for being a minor in possession of alcohol, something that is undoubtedly a lot easier to get arrested for in Utah than in Milwaukee–after all, minors in Wisconsin are allowed to consume alcoholic beverages if accompanied by their parents.

    And in the NCAA’s football world, an arrest such as the Utah kid’s would be pathetically small potatoes.

Comments are closed.