Wednesday, June 13, 2007

Duke Rebounding 2008

From my perspective, rebounding is the most misunderstood aspect of basketball. The most obvious evidence is that rebounding margin is still one of the 4 or 5 major stats that an analyst will show at half-time for any game. The idea this conveys to most people (incorrectly) is that whoever gets the most rebounds is the better rebounding team. Usually analysts will also show the stat to reinforce just how much one team is dominating another.

Of course, rebounding margin might be a sign of one team dominating the other, but not for the reasons some might think. Why are rebounds important? At the most basic level, it's the same reason why rushing is considered a fundamentally important aspect of any offense in football: ball control. The team who has more chances to score will probably win. So to that end, I'll tend to agree that yes, the team with the more rebounds has a better chance at winning, in theory. However, rebounding margin is one of the lesser stats correlated with winning, while field goal percentage is the most.

But why should number of raw rebounds matter the least in deciding who is the better rebounding team in a head to head match-up? What if one team is shooting 60% and the other shooting 30%? Obviously, the first team will have many more opportunities for rebounds (since, typically, a defense has about a 60-70% chance of getting a defensive rebound) than the first team. Very good teams without exception have many more rebounds than their opponents because very good teams shoot a much better percentage from the field than their opponents. Simple as that.

So specifically related to Duke: I actually am one of the few people who think rebounding will be the least of our concerns next year. In fact, I will contend that we will be a good rebounding team. In order to argue this, though, I need to do some groundwork.

What's interesting to note is that after Shelden Williams graduated, a lot of critics suggested that Duke would have trouble rebounding without him in the paint. Of course, in reality, the 07 Devils were a much better rebounding team than the 06 Devils. In 06, our team rebounded about 29.61% of offensive rebounds, and 60.01% of defensive rebounds. This is incredibly bad. The slightly shocking part of it all is that Shelden Williams was such a dominant rebounder (12% OR, 22.6% DR by himself). So, of course, the common pundit might suggested that losing Williams from an already abhorrent rebounding team will simply make the team even worse on the boards.

In 07, however, we were one of the best rebounding teams in the ACC. We rebounded 35.03% of offensive rebounds and 67.12% of defensive rebounds, in spite of losing one of our best rebounders of all time. How does that work? Well, one notion that I will try to dispel is the idea that our change of pace affected our rebounding somehow. In 06, we also averaged about 71.6 possessions a game, which dipped to about 66.4 in 07. The suggestion might be, then, that because we played a fairly slower game, we had more people to crash the boards and a greater ability to grab rebounds on both sides.

Taking a quick look at the stats for ACC teams this year, what I find is that there is really no discernible correlation between tempo and rebounding.

Team Tempo OR% DR%
North Carolina 73.92 39.43 70.77
Maryland 73.64 36.66 63.93
Wake Forest 72.57 30.75 66.83
Virginia 71.67 33.49 66.58
Georgia Tech 70.62 41.36 67.26
Clemson 68.76 37.85 66.47
Florida State 68.22 35.01 63.89
North Carolina State 67.56 27.74 66.08
Duke 66.38 35.03 67.12
Miami 66.24 38.48 66.37
Boston College 66.16 38.07 62.65
Virginia Tech 65.57 32.71 65.78

What I find most interesting about this list, actually, is not just how little tempo has to do with rebounding, but how little offensive rebounding has to do with defensive rebounding. Boston College and NC State are especially interesting. NC State is at once the worst (by far) offensive rebounding team in the ACC, but still was a decent defensive rebounding team. BC was one of the better offensive rebounding teams in the league, but the worst defensive rebounding team. This, at the very least, really hammers home the point that the two are completely different stats and should be entirely separated. At any rate, tempo does not appear to be a substantial factor in rebounding. Duke was the 3rd best defensive rebounding team behind the much faster UNC and the faster Georgia Tech. That being said, I am still not willing to preclude the possibility that tempo has some effect on rebounding.

Now, looking at a list of the top 15 rebounders in the ACC (these are all players with over 40% of minutes - so at least 16 minutes a game or so) it seems that rebounding is mostly a function of personnel.

Player DR%
Jason Cain (UVA) 20.65
Reyshawn Terry (UNC) 20.22
Ra'Sean Dickey (GT) 19.52
Ekene Ibekwe (UMD) 19.06
Kyle Visser (WFU) 18.74
Brandon Costner (NCSU) 18.73
Josh McRoberts (Duke) 18.71
Uche Echefu (FSU) 18.45
James Gist (UMD) 17.68
David McClure (Duke) 17.67
Tyler Hansbrough (UNC) 17.04
Jeremis Smith (GT) 16.6
James Mays (Clemson) 16.41
Al Thornton (FSU) 16.35
Dwayne Collins (UM) 16.16


Each of the three best defensive rebounding teams has 2 players in the top 15. Maryland is sort of the strange one since Gist and Ibekwe were both excellent defensive rebounders yet they still were not a very good defensive rebounding team. The problem for UMD was more that outside of those 2 guys, the other major players (Vasquez, Jones, Strawberry) were nearly non-existent defensive rebounders. This may have been affected by style, since Maryland had a tendency to shoot off on a break with those three guys leaving the frontcourt to clean up rebounds. So maybe Maryland is an example of a high tempo team that suffers on the boards because of its style.

Now it may seem an obvious and stupid conclusion that personnel has a huge impact on rebounding. But I think the conclusion is a little more subtle than that. The idea is not that you need one guy in the paint who can rebound really well or something, it's that you need a team who can come together to grab a lot of rebounds. For instance, just looking at Duke, we had 6 guys who played over 40% of minutes, and of those 6, 4 of them averaged 10% or more of the defensive rebounds (Nelson, Henderson, McClure, McRoberts). Looking at UNC, they also had 6 players who averaged 40% of minutes or more, and they also had 4 of them grab on average over 10% of defensive rebounds (Ginyard, Terry, Wright, Hansbrough). Georgia Tech had 7 guys over 40% of minutes, and 6 of them averaged over 10% DR rate (Crittenton, Young, Smith, Dickey, Peacock, West).

It seems that having one or two really reliable defensive rebounders is key, but then you also need several support guys who can contribute a lot of defensive rebounds as well. So, now that the groundwork is done, what will Duke be like in 08?

Well, the kneejerk reaction is to say that we will be worse off because of losing McRoberts. I actually don't think this is exactly true. I do think we will probably be a worse defensive rebounding team, but not because of McRoberts. Say, for instance, our line-up is Paulus, Henderson, Nelson, McClure, Zoubek next year. We would actually be a very good rebounding team with McClure and Zoubek in the frontcourt and very good rebounding wingmen Henderson and Nelson. I think this line-up will see a fair bit of burn. But factoring in guys like Singler (unproven, so I can't predict how well he'll do), Scheyer (not a good rebounder) or Thomas (could become a decent rebounder, but wasn't his first year) will probably affect our ability to rebound. That being said, though, McClure only played a little over 700 minutes this year (53% of total) with much of the time he spent sitting being filled in by Lance Thomas, and we were still the 3rd best defensive rebounding team in the ACC.

The point is, a lot of Duke fans want to see a return to the faster paced teams of the the late 80s, early 90s and the late 90s, early 00s. This 07 Duke team was tremendously slow by Duke standards. Yet there is also a fear that we will not be able to rebound well enough to get out on a break. I don't really believe that at all. I think we will be, again, a very good rebounding team with the tools to run a faster paced offense (71, or 72 possessions a game would be nice to see run well).

The other point I wanted to make was simply that rebounding is misunderstood and covered poorly by basketball analysts. Just look back at ESPN's top 10 best teams of the last decade: they actually criticized a 2001 Duke team for having a smaller rebounding margin despite being a more efficient offensive team than any other champion on the list and being among the better defensive teams.

Tuesday, June 5, 2007

More in favor of Oden #1

Oregon Live has reported the results from the Orlando pre-draft camp. The link name is sufficient. Oden is incredibly fast, agile, and huge. 7' with shoes and a 7'4.25" wingspan? 9'4" standing reach? Hello!

Durant's measurements are actually really impressive... I never had any idea he was actually 6'10"+ with shoes, but that really justifies his freakish wingspan (7'4.75"). Given that wingspan, I'm not entirely shocked that he can't put up 185, but even once? I'm about 6'7" with about a 6'10" wingspan and I find benching to be much harder than my shorter friends, but even so, 185 is not particularly difficult. He'll REALLY need to bulk up a bit to take the strain of his future position in the NBA.

Monday, June 4, 2007

Durant??

So I was looking over Draft Express, and noticed they had Durant going first to Portland in their mock draft.

First, let me say that yes, this is yet another blog entry by some random guy about Oden and Durant. And yes, this is yet another blog entry talking about why Oden is the better pick. I understand this is not a particularly controversial position. I just hope I can provide a couple fresh arguments for why Oden is the better pick.

Common Sense

I cannot for the life of me figure out why the Trailblazers would select Durant over Oden. I mean, it's hard to say either decision is the wrong one, especially this early in the game, but I just cannot understand why, at the forefront, a team would think Durant is the better pick.

I'll voice some concerns I have for Durant. First, he's skinny. I know he's young and will get stronger... but I think he's probably got that kind of bodytype that doesn't put on weight very easily. For Durant's first handful of seasons he will be best served as a jumpshooter probably. The size and physicality of NBA interior defenders will wear on him quickly if he drives, and I don't think he has the strength to post up on decent sized NBA power forwards. The great upside about Durant is that the kid is sort of a freak in terms of a mix of size (including knuckle dragging primate arms), skillset and athleticism. When he gets bigger, he will be a force, but does Portland really want another perimeter player offensively? Even when Durant fills out and gets used to the rigors of the NBA season, he will be a perimeter player first, and an interior player second. The answer, of course, is no.

The problem really is that you are probably not going to win with two superstar perimeter players unless one of them is named "Jordan." Look at Denver, for instance. They have a good, solid team built: Blake, Iverson, Carmelo, Nene and Camby as their principle five. Two perimeter superstars, a solid, quick PG who can defend and hit some 3s, and two good interior defenders/rebounders, but no scorers. Now, maybe Denver will mature into a complete and consistent team, but as for now, their dependence on the perimeter game for offensive production severely limits them. Portland would see the same problem if they picked Durant. Randolph is a problem anyway, and they should deal him no matter what, leaving Aldridge as their main inside option if they picked Durant. Aldridge will probably turn into a solid player, but I don't really think he'll carry a team offensively in the middle like Elton Brand or Carlos Boozer or Tim Duncan (notice how I put 2 Duke players first?).

So again, why would Portland pick Durant? If they pick Oden, they can deal Randolph and they are left with an imposing interior defense with immense upside on the offensive side. Brandon Roy has all the right tools to become a legitimate perimeter star with some good coaching and some good teammates. That's called balance, and balance wins.

Stats and Effect on Team

Durant was the golden child of the NCAA this year for his gawdy stats, but it's easy to forget that his team lost in the 2nd round and went 25-10. Some might argue that Texas just wasn't very good without Durant. Well, maybe, but Durant headlined an excellent recruiting class and there were a number of very promising players on that Texas team. In fact, they will without a doubt be a pre-season top 25 team even without Durant.

Meanwhile, Oden's team went 35-4 and lost in the NCAA title game (which was, as we all know, not any fault of Oden's). Some might say that OSU was just a better team than Texas outside of Oden. A lot of people even went as far as to say that Ohio State played better without Oden on the floor. Now at this point we all know that's wrong. In the 10 games before the NCAA tournament, Ohio State had a net +27 points per 40 minutes with Oden on the floor. In the NCAA tournament, with only two exceptions (Florida and Xavier), Ohio State dominated the opposition with Oden on the floor, and... well, didn't with him on the bench. The fact is that the Buckeyes were a rather... average team without Oden near the end of the season as Cook stopped producing. They relied too heavily on Lewis to score, but though he was a pretty solid support for Oden and Conley, he couldn't carry a great team. Cook fell off towards the end of the season after having a solid early season. At any rate, Ohio State was as good as they were precisely, and primarily because of Greg Oden.

The difference was not necessarily offensive (though, and I will get to this, I believe Oden was a better player to have on the floor in order to open up offensive options for other players than Durant was). In fact, I think it's safe to say that Oden's biggest impact is defensive. Keep in mind that Oden only played 59% of the total minutes that Ohio State played this year (missing 7 games and being a big man limits minutes), and he still logged 105 out of 218 blocks for the team. He blocked 12.7% of all opponents' 2 point shots while on the floor, and I have no idea what kind of atmospheric number he altered. The paint was virtually impossible to score from while Oden was on the floor. Furthermore, he grabbed well over 23% of the possible defensive rebounds (although, to be fair to Durant, he grabbed about the same percentage).

Now about the offensive side, obviously Durant scored more and scored in many more ways than Oden did, but Oden scored more reliably. I look here to floor percentage (essentially a stat used to estimate how often a player scores when he tries to score) and find that Oden's is an impressive 61.8% and Durant's is 55.1%. Furthermore, since Oden played in the paint and was such a dominating physical presence compared to the other college big men, he was able to net 15% of all possible offensive rebounds: a tremendous number in itself. To summarize, Oden increased his team's chances to score by dramatically increasing the team's ability to get offensive rebounds, and was a constant threat because he scored nearly 2/3rds of the times he tried. One other stat not to be overlooked is Oden's FT rate vis-a-vis Durant's. While Oden shot about 64 free throws for every 100 FGA, Durant shot about 39.6, but against different competition. I think the implication here is that Oden will almost always get an opposing big man in foul trouble and open up the middle for easy, predictable (but much less exciting) baskets while Durant will mostly be scoring with less reliable midrange jumpers and 3s (Oden shot 61.6% from the field while Durant shot 53.6% - eFG% of course).

There has long been a sort of wisdom in basketball that a great big man will bring you championships, but for some reason people have gotten away from this. My best guess is that the most hyped center of all time (Ewing) brought home zero rings to the Knicks in his long hall of fame career. Of course, that was primarily due to a freak named Jordan standing in his way, because the Knicks had championship caliber teams with Ewing in at least 2 years (and probably more). Kareem, Russell, Wilt, Walton, Hakeem, Shaq, Duncan - all these guys anchored (multiple) championship teams offensively and (most importantly) defensively, and within 10 years I imagine Oden will be added to that list.

So I ask again, Durant?? Really??

Sunday, June 3, 2007

Cavs v. Spurs

I like how Cleveland was able to really attack Detroit in all the right ways for 6 games. I don't like the insinuation that Detroit just played poorly throughout the series. Detroit was coming off a series where they beat a very good Chicago team pretty easily, so there's no reason to think they wouldn't be playing well. Furthermore, teams don't just randomly play poorly for 6 games straight - Cleveland forced them to play poorly. The Cavs may be ugly to watch, but they play good defense, rebound well, and are able to score well enough to win.

Now a comment about San Antonio:

Here's the thing (off topic a sec), I think San Antonio (as much as I dislike them for their style of defense and for all the Euro-soccer like whining and drama) is not an ugly team to watch play basketball. They aren't a fast team to watch play, but neither was Georgetown this season (and I think the Hoyas were great to watch - spectacular on offense). San Antonio has the best big man in the NBA right now, one of the most explosive playmakers, and an off-guard who can score from all over the court. The reason people assume that the Spurs are bad to watch is that they control the ball and keep a somewhat slow tempo. But their offense is right at the top of the league. Let's sit back and enjoy good basketball, because the Spurs are the best team in the league.

The Spurs will beat the Cavs and win another championship, but the Cavs will compete. I think 6 games is appropriate. But that being said, Duncan is not a young man and as Duncan deteriorates, so do the San Antonio Spurs. Meanwhile, LeBron pretty much single handedly took his team to the Finals, and he's 22. With Durant and Oden both going to the West, we are probably going to be looking at the Cavs in the NBA Finals a lot over the next decade or so.