Tuesday, January 30, 2007

Beyond the Box Score

In the continuing effort to bring you things you don’t get anywhere else, I’d like to give you a sneak peek into what the 2008 season will bring for this site. It will be about measuring things that have never before been measured (at least publicly) in the college game.

A scratch-the-surface example of this is plus/minus data. Just like its hockey counterpart, a player gets a point for each point his team scores while he on the court, and one taken away for each point allowed. Plus/minus has plenty of limitations, especially when it’s confined to just a single game. But ideally it helps us measure the little things that don’t show up in the traditional stats.

Let’s take a look at a couple of intriguing games played last weekend through the plus/minus lens. First up is Baylor at Texas, an 84-79 Longhorn win.

Texas       PF  PA  +/-     Baylor    PF  PA  +/- 
Atchley     47  33  +14     Rogers    61  60  + 1
Abrams      81  69  +12     Dugat     48  48    0
Mason       71  64  + 7     Lomers    25  25    0
Augustin    74  72  + 2     Diene     11  11    0
James       43  41  + 2     Shepherd  21  22  - 1
Durant      72  72    0     Jerrells  68  70  - 2
Winder       3   8  - 5     Hurd       2   4  - 2
Lewis       17  21  - 4     Carter    60  64  - 4
Pittman     12  15  - 3     Bruce     67  74  - 7
                            Bush      32  42  -10

Connor Atchley comes out the big winner, which might make you wonder about the value of such an analysis. But when you look at Atchley’s line, you can see that it isn’t so surprising he was a contributor to success while he was on the floor. Harkening back to the Line o’ the Night days…

                           FG    3pt  FT   Reb
                      Min  M-A   M-A  M-A  O-T  A F  S TO BLK Pts
Connor Atchley         22  3-5   0-2  3-4  4-6  2 2  1  0  2   9 

It was an active night for Atchley, and it’s nice to see it reflected in the plus/minus. Honestly, I didn’t see the game. (I was working on a contraption to spit out this kind of data in a few seconds.) But if I had the tape, I’d go back and examine Atchley’s performance more closely.

Another interesting thing is that even though Kevin Durant scored 34 points, he had a plus/minus of zero. Compare that to AJ Abrams, who also played 36 minutes, scored 22 points and had a +12 rating. The problem with this analysis is that you are basically comparing these guys based on the 4 minutes they were on the bench, and obviously that is a ridiculously small sample.

The game of the weekend in retrospect was Virginia at Clemson, a game won by the Cavaliers, 64-63. Here’s how the plus/minus shook out in that one.

Clemson     PF  PA  +/-     Virginia   PF  PA  +/-
Potter      29  19  +10     Singletary 41  31  +10
Mays        57  53  + 4     Joseph     43  34  + 9
Hammonds    52  53  - 1     Meyinsse    7   2  + 5
Booker      35  36  - 1     Reynolds   62  59  + 3
Sykes        8  10  - 2     Harris     24  22  + 2
Powell      17  19  - 2     Tucker     13  13    0
Perry       25  27  - 2     Mikalauskas 4   5  - 1
Rivers      43  46  - 3     Cain       35  36  - 1
Hamilton    49  57  - 8     Soroye     35  37  - 2
                            Tat        22  27  - 5
                            Diane      34  49  -15

It’s comforting to see what most would consider the best player on the floor, Sean Singletary, come out on top in this one.

Dave Leitao gets the coaching move of the century. With 5:05 to go, and Virginia trailing by 14, Leitao sent Mamadi Diane to the bench. UVA would proceed to the score the next 13 points. Obviously, the run wasn’t totally or even mostly due to Diane’s absence, but his -15 rating was built on watching that run. Leitao did bring Diane back in for Virginia’s final, and game-winning, possession.

The possibilities are endless with this kind of data. Over the next few weeks, I’ll explore some of those possibilities. 

Friday, January 26, 2007

Defending defense

I’d like to start things off by acknowledging the first and maybe only nominee for the 2007 Center for Tempo-Free Stats Luke Winn Trophy for best writing. (By naming the trophy for Mr. Winn, I can make him ineligible without any guilt.) Please welcome the Greensboro News-Record’s Jim Young to our club. Jim uses the power of tempo-free stats to illuminate the point that Duke’s defense may well be improved this season even by losing what many thought was the defensive player of the year.

The case is pretty solid, and it’s a point that needs to be made by somebody besides me. Duke’s style may be ugly and boring this season, but it is more effective than most have been giving them credit for. And frankly, I never thought I’d be a Duke cheerleader and I’m starting to feel nauseous, so I’m going to move things along. But Triad newspaper readers really should give Mr. Young a hearty pat on the back the next time they see him.

I’m going to get back to posting things besides e-mail responses here, I swear. I’ll shoot for February 1st for a return to semi-regular postings of original thought. In the meantime, dear reader Dan provides me with some good stuff.

Hey Ken,

I know you aren’t a huge fan of predictions, but based upon UConn’s resume to date—and the fact that they lost by double digits to a less-than-impressive Louisville team that shot 31%—I wonder what you think the Huskies prospects are in the Big East and for the postseason.  It’s interesting they remain fairly high in your rankings but come in at No. 88 in the RPI.  They’ve beaten nobody thus far; I guess the question is, “how likely is it that they will and will that be enough?”

Thanks,
Dan

UConn goes down in the kenpom.com bust hall of fame alongside 2005 Arizona State and 2006 Washington State. They probably have to get to 20 wins to have any chance of getting in. But honestly, given how their offense crumbles at any sight of having to run a half-court set against a modicum of resistance, the Huskies are now my early pick to win the NIT. It seems like a lot of the blame has fallen on AJ Price, but it’s not really his fault. I thought I saw Jeff Adrien’s picture on a milk carton the other day. I know it would help him out if somebody was an outside threat, but he doesn’t have much ability to score on his own.

Dan adds…

p.s.  I’d love to put in a strong request for your take on/a breakdown of the upcoming UNC-Arizona game this weekend. I’m sure you’ve noticed, but Arizona has the nations best adjusted offense rating and, amazingly, has done it against the best opposing defense rating.  (not to mention the teams they’ve played have the no. 2 offense, additionally= brutal).  Meanwhile, UNC for all the hype they have gotten for their offensive potency (No. 4 AdjO), have really slipped under the radar thus far as a stellar defensive team (No. 2 AdjD—against the 17th-best opponent’s offense rating).  Somehow, in the breakneck pace that seems inevitable, I think UNC’s defense MAY be the difference but will likely get completely ignored in the analysis in favor of unrepresentative offensive statistics.  Of course, Arizona needs a win pretty badly right now and has had the type of horrible luck thus far that makes you think they are more than due for the type of close win that has been torturously escaping them this season.

I don’t know that I need to do much more analysis than that. We have Arizona, the most underrated team in the most overrated conference, but with defensive issues. And we have UNC, a defensive juggernaut. On a related note, it was interesting to hear how UNC shredded Clemson’s press last Saturday. In fact, they scored 77 points in 79 possessions - their second worst offensive rating of the season. Carolina’s defense/Clemson’s offense was the story behind that game.

Getting back to Arizona - it’s not so unusual that they have the #1 offense against the #1 defensive SOS. Those offensive figures are adjusted for that schedule. What it does mean is that the impressive raw numbers that guys like Radenovic and Budinger are putting up have a higher degree of difficulty associated with them than most or perhaps all other players in the nation. And they are coming off a pasting of Arizona State without Marcus Williams.

The Cats are the second best team in the Pac-10, and I think in mid-to-late March that will be obvious. This game is winnable, but if they lose, watch how they lose. None of their losses have been disgraces, they’re sitting near the bottom of the luck standings, and their defense is probably underrated by my stats because they know they don’t need it on every possession with the scoring punch they have. Don’t get me wrong, they have limitations defensively, but having one of the best offenses in the nation is worth something. Where some see fatigue, I see bad bounces and a tough schedule. They have a late-season run written all over them.

On the UNC side, any wagers on whether we’ll hear about how Roy Williams wants his team to be more focused on defense? I don’t doubt that he does, but to imply that the Tar Heels are lacking defensively is nowhere close to reality. Much like the 2005 team, this one is balanced. This season’s version has been slightly better on defense than offense if you want to split hairs. But that will be our little secret.

My prediction: I’d be surprised if this wasn’t the most entertaining game of the weekend.

Locally, BYU is virtually giving tickets away in an effort to defend its 26-game home winning streak against Air Force. If you’re at the game, I’ll be the one not wearing white.

Tuesday, January 23, 2007

Fun With Birthdays

As part of some other work I am doing, I have had to compile a birth date database. Naturally, I need to look at the extremes, out of both curiosity and entertainment. Near as I can tell, the elder statesman of college hoops is Tennesee Tech’s Francis Kabore, who checks in at 28, at least 18 months older than anybody else.

Kabore’s age appears to be legit, but there are some others out there who seem to be fudging things. For instance, the youngest player is allegedly William Bullard from Texas A&M Corpus Christi. According to the info on the Islanders’ web site, Bullard turned 17 just a couple weeks before practice started. That’s unusual enough - except that Bullard happens to be a junior. Actually, Bullard’s name was recently removed from the Islanders’ roster. Perhaps he quit the team to focus on finishing up his degree in three years. The “real” youngest player might be Charles Jenkins, a freshman guard at Hofstra who’s listed as not turning 18 until the last day of February.

Oddities also abound from UNLV, where the school lists two players as being 26, but I’ll need to see some ID. One is listed as having graduated high school as a 22 year old when you do the math.