I have a confession to make: All-Defensive teams annoy me. The reason is two-fold - first, the statistics available for measuring defense are painfully inadequate. Sure, we can measure blocked shots, steals, and rebounds, but that's not all of defense. Not nearly. What about limiting shot attempts? Or playing a lot of defense without fouling? These things are a big part of defense, but we have no way to calculate it all. That leads to my second problem with All-Defensive teams - they're too subjective. Not only that, it seems year after year there are players on these teams that are pretty bad offensive players. It makes me wonder if college basketball has its own version of Nichols' Law. Now, some will argue that these guys must be good at defense because, after all, they play a lot of minutes for their teams in spite of their limited offensive abilities. Surely the coaches must be putting them in for a reason, the argument goes.
That's all well and good, but team defense would seem to fly in the face of this reasoning. Unlike individual defense, we have pretty good stats available for team defense. The biggest one of all is defensive efficiency, which is simply a measure of opponents' points scored divided by the number of opponent possessions. It's simple, intuitive, and relatively easy to calculate. What strikes me is that, year after year, the top defensive teams in the country are also pretty good offensive teams, by and large. Kansas, UCLA, Memphis...even Wisconsin was a top-30 offensive team last season. My guess is that you'll find a pretty strong correlation between offensive efficiency and defensive efficiency. So at least from a team defense perspective, it does not appear that the Chris Kramers and Chester Fraziers of the world are shutting opponents down.
Of course, that doesn't mean that we give up and just throw the best offensive performers on the All-Defensive team. Instead, we should be content to use the numbers that are available to us (Block%, Steal%, DReb%) in order to come up with our top performers. From a team defense perspective, good defenses have these things in spades. Kansas, for example, was the top defense in the country last season, ranked #5 in Block%, #16 in Steal%, and #23 in DReb%. The truth is, a high Block% doesn't just mean a lot of blocked shots - it means a low 2 point percentage. Now, be careful in how far you take this logic, but we do know that getting blocks, steals, and rebounds are really important. With that in mind, I give you the Big Ten All Defensive Team, and I promise that I didn't look at the offensive numbers for these guys.
Damian Johnson, F, MinnesotaConference Stats: 6.3 Block%, 4.8 Steal%, 12.7 DReb%
I really like Johnson's potential as a lock-down defender. But it's hard to call him an elite defender at this point because of that lackluster rebounding. The guy is 6-7, athletic (shown by his ability to block so many shots at 6-7), and frankly, not competing with any great rebounders on his own team. With leading rebounder Dan Coleman gone, Johnson's rebounding numbers should go up a tick this season, and I think he could end up being one of the best defenders in the country. He's my favorite for defensive player of the year.
Goran Suton, F, Michigan StateConference Stats: 2.5 Block%, 1.6 Steal%, 24.5 DReb%
The best returning rebounder in the conference, and it's not close. Most defensive lists sell the rebounding big men short, but that would be a mistake. Unlike a blocked shot (which voters pay much more attention to), a defensive rebound ends an offensive possession every time. Suton is also a very skilled offensive player, and will battle with incoming freshman BJ Mullens as the conference's best post player. My money is on Suton.
Chris Kramer, G, PurdueConference Stats: 1.9 Block%, 5.0 Steal%, 12.1 DReb%
Ok, so maybe Chris Kramer isn't just a Nichols' Law selection. But I think there is a big dropoff on this team after Johnson and Suton. Kramer does get a lot of steals, but so do guys like Trevon Hughes and Al Nolen. What separates Kramer is his willingness to rebound.
Robbie Hummel, F, PurdueConference Stats: 2.7 Block%, 2.6 Steal%, 14.8 DReb%
Hummel doesn't excel in any specific area, but he does a little bit of everything. It's likely that Hummel will so more time defending the post this season, and that might mean less opportunity for blocked shots (as Hummel will not tower over his opponent), and more rebounds instead.
JaJuan Johnson, C, PurdueConference Stats: 8.5 Block%, 1.9 Steal %, 11.3 DReb%
Johnson is the conference's best returning shotblocker, and he could be scary-good in that regard this season. Ekpe Udoh, who would be the conference's best shotblocker if he wasn't transferring to Baylor, had a block rate 11.4% last season as a sophomore. His freshman Block% was 8.9. Purdue fans hope Johnson is on a similar path.
What's the takeaway from this team? Well, if the rest of Purdue can put together adequate defense, the Boilermakers should be very difficult to score on.
5 comments:
Thanks for the stats explanation. One of my all-time favorite stats is the Birdie--named after Larry Bird, of course. You simply add up all the positives for a player, points, rebounds, assists, steals, and blocks, then subtract all of the negatives, missed field goals, missed free throws, fouls, turnovers, technicals. That number is the raw Birdie. After you've done that, divide by the number of minutes played for a percentage and you have a pretty good idea of the player's value--adjusting for position. For example, a center or power forward should be around 70 to 75% while a point guard is doing well with a 60%. Do this a few times with a full NBA box score and discover how truly over-rated Allen Iverson really is. He's the only guy around who can score thirty points and have a Birdie of 5 and a percentage of 8%.
Fellas, love the site so far, but one quick question. Did someone from Ohio State torment you in the past? You seem to have a bias against my Bucks. No Buckeyes on the first team or all-D team? Preposterous. I am thinking you are a couple of northern Indiana kids. Draft Express rates David Lighty and Evan Turner as numbers 2 and 3 for the NBA on the strength of their defensive prowess. Lighty can gaurd any position from 1-4 on the floor. Those 2 wings will cause fits for the rest of the BT this year. Goran over BJ? I seriously doubt anyone outside of a 20 mile radius from EL would agree with you on that trade. But keep writing, cause I love reading and Go Bucks!
Derm,
I can assure you, that we have nothing against your Buckeyes.
DraftExpress is a "scouting" website. Numbers mean less to them than they do to us. For us though, the decisions have to look good "on paper." With Lighty, I see a guy with
DReb%: 7.4
Stl%: 2.1
Block%: 0.6
Really, nothing special there, and the rebounding is actually pretty weak, although he had 2 formidable rebounders to work against on his own team last year.
Turner was a bit better, at
DReb: 13.3
Stl: 2.4
Block: 2.3
Those are numbers that are impressive across the board, but who would we put him ahead of? If we had a second team, he'd be on it, but I can't rank him ahead of anyone we listed here.
As for Suton, well, seniors are pretty valuable players. I have no doubt that Mullens will be a top talent next season, but I don't see any reason to believe he'll be closer to Oden than Koufos.
Using the only available individual defensive stats is woefully inadequate, as you point out in your initial article. Whenever you use only that info, you come up with really silly-looking conclusions.
One of the big differences is stylistic defensive strategies. Right now Painter and Tubby play more for the steal than anybody else. Wisconsin and MSU play more of a containment style defense which emphasizes good position, forcing tough shots and defensive rebounding.
Some guys get lots of steals but they take lots of chances and get lots of whiffs lunging for the ball and missing it. That usually results in a foul or a 5-on-4 situation. Brody Boyd from Iowa a few years back comes to mind. Same thing with shot blockers. Some guys leave their feet on every head fake and ball fake and maybe rack up lofty numbers on blocked shots but due so at the expense of giving up lots of easy shots, offensive rebounds and lots of fouls.
So shame on you for noting the problems with using just those three defensive stats and then going on to using it for your all defensive team.
Anonymous -
Should we cast the defensive statistics aside then? Rely on one set of eyes? The nice thing about numbers is that they carry some inherent authority that subjective opinions lack.
And we didn't strictly worship at the altar of steals and blocks. Suton isn't especially gifted at either, but makes this team because of his rebounding. Furthermore, it was Kramer's rebounding that put him on this team and kept a guy like Hughes or Al Nolen off.
Carrying your point further, does that make Trevon Hughes a bad defender? After all, if he's in a containment defense, yet grabbing all these steals (3.9 Steal%), is he gambling for his steals, and playing outside the team strategy? If not, how do we know that?
Finally, did all those MSU players forget they weren't supposed to be blocking shots? Because, as a team, their rate was awfully impressive (12.2).
Post a Comment