Hoops Nerd

The Hoops Nerd's day job has been effing with the blogging schedule. Not cool - but here are the all-new late February Hoops Nerd Ratings.

The Hoops Nerd Ratings are explained here. As of the approximate beginning of conference play in early January, the preseason projections are disregarded and current season data is the sole input. Offense and defense are, by this point in the season, regressed separately and then reunited to form the Hoops Nerd Ratings. The Hoops Nerd Ratings are expressed in plus or minus points per 67 possessions.

The basic premise is that the quality of offense and the quality of defense in any given Conference is relatively consistent from year to year. Based on a few season's worth of data (Ken Pomeroy's adjusted offensive and defensive efficiencies, of course), we can form an expectation for the average quality of, for example, an ACC defense, or an Ivy League defense. We can also form expectations of the distribution of offensive or defensive prowess amongst the teams in a given Conference, again based on a few season's worth of data. The Hoops Nerd Ratings use z-scores to transpose current season data onto our expectation model, taking into account both average and distribution, for both offensive and defensive efficiencies for each team. The value in this approach rests largely with its ability to rein in aberrant results on either side of the ball, recognize the legit outliers, and use of historical context as the base to which regression will ultimately pull.

1. Kansas (Jan 30 Rank: 1)
Hoops Nerd Rating: +24.3
Rock! Chalk! Grossly Underappreciated by the National Media Jayhawk! Both Ken Pomeroy's ratings and Jeff Sagarin's Predictor ratings say that they are the best team in the nation, and your humble Hoops Nerd concurs, yet the AP and ESPN/USA Today have Kansas slotted 6th and 7th, respectively. The difference really is in the objectives of the different types of systems: systems like Pomeroy and Predictor are designed to quantify performance, whereas the media rankings are more an attempt to attribute varying degrees of merit and deservedness to each team, often based on teams' records (regardless of margin of victory) and an illusory/fictional transitive thingy ("Stephen F. Austin beat Oklahoma, and Nicholls St. beat Stephen F. Austin, then Nicholls St. is better than Oklahoma!"). That is, the former tells which team is best, whereas the latter tells you which team deserves to be ranked first.

2. Duke (2)
+23.5
Like that of Kansas, Duke's national AP ranking (7th) belies its true quality. Recent consecutive road losses to Wake Forest and Miami (FL) may be what sinks the Blue Devils' shot at a #1 seed in THE tournament: Lunardi currently has Duke projected as the last #2 seed (i.e. 8th overall) and Bracketology 101 has them as the third #2 seed. The team will have opportunities to play its way back into a #1 seed, with a showdown vs. North Carolina on March 8th and of course the ACC Tournament.

3. Tennessee (4)
+22.7
The Vols have dominated the SEC for most of this season, and as such have been looked upon kindly by the Hoops Nerd Ratings. As can be read in the detailed description above, the Hoops Nerd Ratings transpose current play relative to conference onto an expectation model based on historical results - exactly the kind of approach that would have believed in the Vols all along. Since November 26th, each installment of the Hoops Nerd Ratings has ranked the Vols higher than the corresponding week's AP Poll did. On 11/26, these ratings had Tennessee ranked 6th, the AP had them 11th, on 12/3 it was 3rd here and 10th in the AP, on 12/10 it was 3rd here and 12th in the AP, on 12/31 it was 6th here and 8th in the AP, on 1/8 it was 4th here and 8th in the AP, on 1/15 it was 5th here and 6th in the AP, and on 1/30 it was 4th here and 7th in the AP. That trend finally reversed this week, as the AP went APeshit and vaulted the Vols to 1st. The Hoops Nerd Ratings are still happy to take full credit for having been (heretofore) on the vanguard of this particular Orange Revolution, however.

4. North Carolina (3)
+22.5
Ty Lawson may return in time for Saturday's game vs. Boston College, but one would assume that Roy Williams would have to be assured that the timetable for Lawson's return would have to be conservative enough so as not to risk further injury. The Tar Heels are a virtual lock for a very high tournament seed, and any further regular season success at this point is subject to diminishing returns.

5. Wisconsin (7)
+20.5
With a reasonable expectation of winning at least two of their final three regular season games (vs. Michigan St., vs. Penn St., at Northwestern), the Badgers could potentially secure a #2 seed in the Tournament. The thought in Madison is that this year's iteration could be superior to last season's more heralded and nationally-ranked Badger team - and those thinkers are correct.

6. Indiana (10)
+19.1
There is nothing that I can say here about the recent goings-on in Bloomington that couldn't be read in much more detail elsewhere, so I will simply quote the following passage that appeared in the Hoops Nerd Ratings column on January 30th of this year, before the recent scandal broke: "Kelvin Sampson wrote this letter imploring fans to be civil at Assembly Hall, no word yet on whether Sampson plans to text the letter to potential recruits..." The Hoops Nerd is nothing if not prescient.

7. Marquette (9)
+19.0
The Golden Eagles propensity to lay thuderous beatdowns on quality opponents and then inexplicably lose a game or two tends to confound power rankings and rating systems, subjective and objective alike. Witness Basketball Prospectus power ratings man Joe Sheehan: Sheehan's rankings column (called "The List", of which I am a big fan) attempts to rank teams using a somewhat subjective wins/losses/eyeball test system, but with empirical underpinnings. Sheehan most recently ranked Marquette 22nd, a ranking that belied his "numbers only" ranking of Marquette as 12th. For a man (like Sheehan) trying to balance the input of his eyes with the input of his Excel sheet, Marquette presents a significant challenge. Luckily for me, it's all Excel sheet here on the Hoops Nerd Ratings.

8. Georgetown (5)
+18.9
February losses to Syracuse and Louisville have put a crimp in the #1 seed plans of the Hoyas, but of course with Marquette, Louisville and the Big East Tournament on the horizon, then could regain their once-lofty ranking.

9. Clemson (13)
+18.5
The Tigers haven't played in a week (having last seen game action February 19 vs. Florida St.). Clemson will need to finish strong against four teams from the upper-middle-tier of the ACC (Miami, Maryland, Georgia Tech, and Virginia Tech) in order to lock down an at-large bid to the Tournament. Two more wins ought to put them into the coveted Glockner Lock Status.

10. UCLA (16)
+18.5
The Bruins appear headed for at least a #2 seed and possibly a #1. The Hoops Nerd is looking forward to the March 8th game vs. Stanford (the same day as Duke vs. North Carolina).

Hoops Nerd Ratings Teams 11 - 30

11. Louisville (18), +18.3
12. Xavier (12), +18.1
13. Michigan St. (8), +17.7
14. Notre Dame (17), +16.9
15. Kansas St. (21), +16.8
16. West Virginia (6), +16.7
17. Texas (27), +16.6
18. Florida (14), +16.3
19. Memphis (15), +16.1
20. New Mexico (26), +15.5
21. Connecticut (19), +15.4
22. Pittsburgh (22), +15.4
23. Drake (11), +15.0
24. Arkansas (NR), +14.8
25. Vanderbilt (NR), +14.7
26. Washington St. (25), +14.7
27. Purdue (NR), +14.6
28. Butler (29), +13.9
29. Mississippi St. (23), +13.7
30. Stanford (NR), +13.7

New this Week: Arkansas, Vanderbilt, Purdue, Stanford
Disappearing this Week: Mississippi, Ohio St., Arizona, Virginia Commonwealth
On the Cusp: Miami (FL), Texas A&M, Ohio St., Gonzaga, Arizona