D-value projections

[jonpin]

Biggest disclaimer first: This is an UNOFFICIAL calculation of the D-values for the 2010 SCT performed solely by me.

Other disclaimers: Teams from Regions 0 (Europe), 9 (Ill/Ind), 12 (Midsouth) are not listed at all because I have no idea how to calculate the data in mixed fields. Teams from Region 4 (Mid-Atlantic) are not listed for obvious reasons.
There is, of course, a strength-of-schedule aspect to the formula, whereby some values are multiplied by (Schedule TPPTH) / (Field TPPTH). Since this is a universal scaling, there is no harm done in assigning a value to Field TPPTH, and I have done so (namely 4.000). The actual average TPPTH is 4.004 in D-1 and 3.965 in D-2 (surprisingly close since I chose my value of 4 after looking at one region).

* denotes automatic bid for region champion. # denotes automatic bid for undergraduate champion.

Division I Top 30

Code: Select all

Min-B 347.48 10*
Min-C 347.48 10
Tor-A 342.33 7 *
Chi-B 332.42 8 *
Irvin 324.68 15*
Yal-A 323.52 1 *
Brown 323.52 1
Chi-A 306.98 8
Fla-A 304.62 6 *
Min-A 287.56 10
----------------
Car-A 287.56 10
Mich. 279.12 7 #
FlaSt 269.29 6
Stan. 266.23 15
UCSD. 266.23 15
WUS-A 259.56 11*
Rens. 251.87 1
Chi-C 231.32 8 #
Colum 230.95 1
Dar-A 224.66 1
----------------
Mizz. 223.52 11#
Ala-A 188.19 5 *
Grin. 181.09 10
McMas 180.94 7
Okla. 176.77 11
SoFla 170.44 6
SCar. 170.44 6
Georg 170.44 6
UCLA. 165.54 15
WOnt. 164.65 7
----------------
UWash 126.91 14* (40th)

There are, to my knowledge, three other automatic earned bids: One in Region 0 and two in Region 12. In addition, I imagine Region 4 will produce an automatic bid. Thus, no team below this cutoff can possibly earn a bid without teams above them declining.

Division II Top 25

Code: Select all

Brown 493.62 1 *
Clem. 479.20 6 *
Yale. 390.57 1
Harv. 390.57 1
Prin. 359.25 1
Ott-B 343.60 7 *
Wat-A 343.60 7
Chi-E 335.00 8 *
Chi-D 333.83 8
GaT-A 329.46 6
----------------
CaT-A 326.73 15*
Mic-B 325.62 7
Colum 322.18 1
Carl. 319.60 10*
Corn. 296.19 1
GaT-B 291.30 6
Ala-C 275.23 5*
Wis-A 262.59 8
No'we 262.59 8
MicSt 262.59 8
----------------
NDm-A 262.59 8
Tufts 261.39 1
Prov. 252.77 1
UCSD. 250.22 15
Tru-A 246.30 11*
----------------
AzS-A 244.07 13? (26th, declined automatic bid)
UWa-B 214.17 14* (33rd)

There are, to my knowledge, two other automatic earned bids: One in Region 9 and one in Region 12. In addition, I imagine Region 4 will produce an automatic bid. Thus, no team below this cutoff can possibly earn a bid without teams above them declining.

[Rothlover]

Thanks for the doubtless time this took to come up with. It really gives you a feel for the clumping of teams, sans the obvious non-entered regions. This is much enjoyed by me personally.

[jonpin]

Obviously if you've got any questions, you can ask me. I may be able to answer some tomorrow in between rounds of the NJ states, but beyond that I'm occupied for the weekend, so it might be Monday before I can explain anything.

Luckily the two things that caused the most confusion within these regions occurred below the Q-line, namely:
URochester (D-1) left before playing two middle-bracket playoff games but for a brief time looked like they might still be in the field. I wasn't sure how NAQT would deal with forfeits counting against teams.
Region 10 (D-2) had an eligibility issue that messed with the standings, leaving the relative finish of teams 3 through 5 somewhat in doubt.

[Important Bird Area]

Note that you can go ahead and calculate D-values for the teams in combined fields that played the set appropriate for their own division. (The problem is how to rank a team that played DI among DII the teams and vice versa.)

[jonpin]

Note that you can go ahead and calculate D-values for the teams in combined fields that played the set appropriate for their own division. (The problem is how to rank a team that played DI among DII the teams and vice versa.)

OK, I suspected that might be the case. That still leaves Illinois A un-ranked for now. Also, that only works if I can tell who is what division. Are there any numbers that disagree from whatever you have so far (in terms of team-above-other-team)?

[dtaylor4]

Note that you can go ahead and calculate D-values for the teams in combined fields that played the set appropriate for their own division. (The problem is how to rank a team that played DI among DII the teams and vice versa.)

OK, I suspected that might be the case. That still leaves Illinois A un-ranked for now. Also, that only works if I can tell who is what division. Are there any numbers that disagree from whatever you have so far (in terms of team-above-other-team)?

For the Indiana SCT, Miami A, Illinois A, and Illinois B are D1, if that helps.

[Important Bird Area]

Also, that only works if I can tell who is what division.

At Oxford: Warwick A and Imperial College London were D2 teams, all others D1.

At Rice? I can't figure this out from naqt.com (even trying to view the field and invoice data for a completed SCT returns an internal server error). We do have records of this somewhere, and R. will be posting these Monday or Tuesday.

[The Ununtiable Twine]

Also, that only works if I can tell who is what division.

At Oxford: Warwick A and Imperial College London were D2 teams, all others D1.

At Rice? I can't figure this out from naqt.com (even trying to view the field and invoice data for a completed SCT returns an internal server error). We do have records of this somewhere, and R. will be posting these Monday or Tuesday.

The D2 field consisted of Rice and A&M, and the D1 field was everyone else, I'm pretty sure.

Are all the projected D-values available, Jon?

[The Atom Strikes!]

Another note: we played the tourney on DI questions.

[The Ununtiable Twine]

Another note: we played the tourney on DI questions.

...which made your performance even more impressive, of course. Sorry I forgot to mention this in my post. Thanks, Henry.

[jonpin]

Also, that only works if I can tell who is what division.

At Oxford: Warwick A and Imperial College London were D2 teams, all others D1.

At Rice? I can't figure this out from naqt.com (even trying to view the field and invoice data for a completed SCT returns an internal server error). We do have records of this somewhere, and R. will be posting these Monday or Tuesday.

The D2 field consisted of Rice and A&M, and the D1 field was everyone else, I'm pretty sure.

Are all the projected D-values available, Jon?

Except for two things: One, I haven't yet included the combined-but-right-division teams in the spreadsheet, and two it will take some time of typing.
Neither of those things will be accomplished today, but I'll try to get a full list online tomorrow night or Tuesday.

[jonpin]

I've begun to do data entry of the D-value calculations onto my website. I will continue this project as time allows tonight and tomorrow and edit this post with updates as more regions become posted.

Posted as of 12:30am/et Tue-into-Wed: Regions 1, 5, 6, 7, 8, 9, 10 (D-I), 11, 13, 14, 15.
Remaining: Region 0 (waiting to see if stats for tiebreakers exist), Region 12 (waiting for stat corrections if possible), Region 10 D-II (waiting for confirmation of NAQT-official order-of-finish), and Region 4 (waiting for it to happen).

[jonpin]

Just bumping this thread to note that the D-value projections are now all entered.

Division I
These teams are above the cut-line on my projections: Minnesota-C, Penn, Brown, Chicago-A, Minnesota-A, Carleton, Florida State, Stanford, UCSD.
Though the official conversion factors have not been announced, I believe Illinois-A and Illinois-B will both receive bids.
The top of the waitlist appears to be: 1 VCU, 2 Virginia, 3 Columbia, 4 Dartmouth, 5 Carnegie Mellon.

Division II
These teams are above the cut-line on my projections: Yale, Harvard, Princeton, Waterloo-A, Chicago-D, Georgia Tech-A, Michigan-B, Columbia, Cornell, Georgia Tech-B, Wisconsin-A, Northwestern, Michigan State, Notre Dame-A.
The top of the waitlist appears to be: 1 Tufts, 2 Providence, 3 UCSD, 4 Maryland, 5 Arizona State-A.

Besides the fact that these are unofficial to begin with, in some cases the margin between teams is small enough that round-off error, let alone typos, could result in a shifting of order.