Tracking Individual Defense

Sunday evening, I began a new basketball project: tracking individual defensive stats for the Toronto Raptors. It began in the loosest of senses when I attended the Raps/Pacers game on the 15th, spilled over to the stat boards, continued on with the Thunder game and then at last took legitimate form as I watched the Raptors outlast the Bulls in an overtime game this past Sunday.

Borrowing liberally from a study begun by Schuyler Davis; some of his work is published over at basketball-statistics.com. The basic method is, at least for now, strikingly similar to the one Mr. Davis used. The details of his method can be found at Basketball-Statistics.com, here. Essentially, I looked at a host of defensive events discretely and recorded when a player either failed notably or played especially good defense. Average events were not recorded, nor would one person be reasonably capable of recording all of those anyway, bearing in mind that there are basically five defensive events happening on every possession. It is worth mentioning that though I performed a trial chart against the Thunder, this was my first full-game chart. There are assuredly many errors, and some subjective judgments that need to be refined and clarified for the next time that I chart a game. It is also true that I am working on finding a more objective way to quantify the results of the game, but in the meantime, the collection of this raw data is an important step to better understanding individual and team defense in a more effective manner than we do now, especially given how clumsy most defensive metrics are.

In any case, I will continue to refine the process as I go along, but for now, the results were still interesting. Bearing in mind that they represent a single-game (and thus very small) sample, the numbers did at least match with the general consensus of Raptors fans: Jose Calderon is not a good defender. Beyond that, the Raptors appeared to play decent defense and the Bulls hit a lot of contested shots (Ben Gordon in particular).

In the first half, Toronto sent Chris Bosh over to help every time Rose got a screen up at the top of the circle and that kept him in check, but once Ben Gordon started putting pressure on the Raptors (he scored only 9 of his 37 points in the first half and only 18 after three quarters), Bosh had other concerns to worry about and Rose was free to attack Calderon in isolation, a matchup decidedly in Chicago’s favor. Rose himself had only 7 of his 23 points after three quarters and erupted on Toronto thereafter. The guard play of the Bulls was nearly too much for Toronto to handle and Chicago’s offense was quite potent, outscoring the Raptors by 13 in the fourth quarter, only to let things slip away in OT because of Rose’s missed layup and the general ineptitude of Tyrus Thomas, who bricked a 16-footer and a 19-footer, his only contributions to the OT period besides a defensive rebound.

Returning to the defensive stats, I was looking at a series of specific events. In particular, I tracked isolation defensive coverage, pick-and-rolls, help rotations/recoveries, contested defensive rebounds and transition defense. In the case of all but the latter, I tracked a positive and negative category. With transition defense, since the opportunities for positive events were so prolific, I only tracked negative events and there were very few, mostly because a player got fouled with no call and fell down (or flopped and fell down) and the Raps were short-handed as a result.

The biggest problem, I found, was choosing whether to record pick-and-roll coverages by the help defender in the pick-and-roll category or the help rotation category. I unconsciously recorded several such events in different columns, so that is definitely something I have to fix the next time around by making a clear choice one way or the other.

The second biggest problem was determining how much was coaching strategy and how much was the players themselves. Before the game, Coach Triano did talk a little about how he planned to play Derrick Rose and you could see it in the first half before Ben Gordon disabused them of the notion that focusing on Rose and not him was the best plan. Once both got going, well, the Raptors couldn’t really do anything because they have but two players who are decent perimeter defenders against guys under 6’5 and those are Parker and Roko. Both did a reasonable job, but it’s difficult to contain that kind of athleticism without a shot-blocker, so we got to see the upside of running two small, athletic guards against a defensive line such as Toronto fielded. The Raptors seemed to design their possessions to that Marion tried to guard Rose, but that was a mistake, and one not much better than if Calderon had been left on Rose. Bosh found himself guarding Gordon in isolation sets repeatedly in the second half and as you can see, that did not work out well at all, though on the possession where Gordon tied the game at the regulation buzzer, Bosh had indeed played him well straight up; Gordon just hit a long J in his face.

So, the basic assumption that Calderon is a poor individual defender was borne out. The demise of Anthony Parker’s defense seems a little presumptuous after watching that game and the Raptors as a team didn’t look bad on many possessions. It was rather stunning to see that in such a high-scoring game, the player doing the most damage was doing it in spite of the defense as opposed to on account of its absence. The Raptors dominated the glass, holding advantages at either end of the floor but the Bulls hit a lot of long jumpers until basically the fourth quarter and overtime, at which point the combined scoring threat of Gordon and Rose was too much for the defense to handle and the lead, as much as 17 points in the fourth quarter, wilted with incredible alacrity.

This proved to be an interesting exercise. As the project progresses, it’ll be interesting to establish a baseline performance level for each of the Raptors players and to be able to compare their performance against specific kinds of offense and see where the Raptors strengths and weaknesses lie in a more particular sense than “athletic guards with jumpers and power-post players,” as is essentially the case for all but the most elite defensive squads. After that, it’ll become interesting to see how effectively this data can be applied to player assessment models and perhaps some kind of “alternate box score” rating metric, perhaps.

Until next time.

SSOS

So it's time for a new post. Actually, it's past time for a new post, but we'll ignore that.

Today's topic is the Phoenix Suns.

There is a great deal to say about what has passed before, so I'm going to ignore all of that for now. Instead, I'm going to focus on a couple of more immediate things, mostly the Orlando game and what's coming up in Miami.

It was clear that Steve Nash needed to rest more; Shaq looked a lot better when Nash wasn't messing around and turning the ball over; an A:T ratio of nearly 1:1 is terrible for Nash, especially because the Suns can't afford those turnovers on account of their poor defense. In any case, Nash had a decidedly unusual game, particularly in the second half; his ankle was still bothering him, it must have been, because he was so tentative. There was one play where he drove, had a point-blank layup and elected to pass through traffic out to Jason Richardson, who fortunately hit the three.

The Suns lost this game mainly because they did a terrible job of guarding Rashard Lewis and Hedo Turkoglu, both of whom torched the team (Rashard had 11 points in the fourth quarter alone and had a great night on the glass, too). One might also point a finger at how Rafer was getting into the paint, drawing fouls and hitting the three-point shooters very well and criticize Nash, but it was his first game back from a mobility injury, so it would be fruitless.

But why did I bring up this game? It's done and past, right? Nash was busting chemistry because he came back too early and the game was winnable but for a few miscues and a really rough 3rd quarter. There were a few notable aspects of the game that were problematic for the Suns.

The first of these was finishing on the break; this wasn't a critical failure on their part but Grant Hill missed a couple of what should have been gimme-layups coming from the right side (not his favorite side) and the Suns really didn't have anyone to work with on the break that could finish. As a corollary to this, Nash didn't have his favorite play available to him, the high sidescreen.

When Amare's in the game (or in Dallas, Dirk), Nash loves to catch a screen on the wing, and work from there. It's a very effective play, but more importantly, while it creates a lot of options (ball-handler drives, pass to the roll, pass to the pop, pass to the wing, pass to a cutter), it's also a very fast-developing play. And it's his bread-and-butter. This is what has made Nash so effective. He could run it with Amare, he could run it with Diaw...

He can't run it all that often with Shaq, nor can he do all that much with it if he's running it with whichever of Barnes or Hill happens to be running small at the 4. Amundson isn't super-effective in this play either.

The other major areas where the Suns were noticeably lacking are rebounding and help defense. Now, in this particular game, the Suns were only -1 on the glass, largely because Shaq grabbed 6 offensive rebounds, but the point is they couldn't put down an advantage in the rebounding column. In April 2007, The Painted Area published an interesting blog post that noted that only two teams had ever won a title with a negative rebounding margin (the Rockets title squads of the mid-90s) and in the last 20 years, every title squad had at least a rebounding margin of +1.0. Only two other teams besides those Rockets squads even made the Finals with a negative rebounding margin.

Right now, the Suns are at about 0.99, so they've improved their rebounding from back in the D'Antoni days, thanks to Shaq's presence. But now they're missing what they had then, which was roughly acceptable perimeter defense on account of Marion's versatility. Kevin Pelton added to this in a comment to the follow-up post on The Painted Area in January of this year:

"The average champion over your original timespan (1987-2006) is 4.2 percent better than league average on the offensive glass and 1.7 percent better on the defensive glass."

The Suns, for the record, are currently 24th in the league in total team offensive rebounds. They're tops in the league in team eFG% and they're 5th in FTM/FGA; their offense is generally pretty good, but they're a little below league average in offensive rebounding, which doesn't square with Pelton's comment about NBA champions in that specific timeframe. They're also 22nd in defensive rebounding percentage.

So, getting to my long-winded point, while the Suns have improved on the glass, they're still not doing a good enough job.

Beyond this, they also don't have a lot of help defense; Shaq's doing a pretty decent job down low preventing post scorers from having career nights every time they face the Suns, but he was never a dominant shot-blocker but for a year or two in his career, just a very good one.

Enter the Suns' newest addition; according to various sources, Stromile Swift has signed with the Suns and will play out the rest of the season with them... and by all accounts I've heard, is likely to become the starting PF for the team, which means he'll probably be seeing 25+ mpg for Phoenix within a games. If these reports are true, he might even be in uniform for the Miami game.

Swift is many things; worth his draft selection in the 2000 draft is not one of them but what he has always brought is strong offensive rebounding, shot-blocking and athleticism. He's 6'9 or 6'10, and has a wingspan of about 7'3. He's pretty strong and does a decent job of single coverage in the post. He's an outstanding transition finisher and even in the halfcourt, he moves pretty well without the ball to get dunks from teammates. He's got a decent face-up game with a competent left hand, and some years he's got an acceptable mid-range jumper. Usually, he's also not all that bad at the line, too.

So in one move, the Suns have signed a guy who should really help with a lot of the problems we've just been discussing. It falls to Stro to perform, but you expect that he will because it's a limited role but one in which he can succeed, and it's a contract year.

What does this do for Phoenix?

Well, it makes them a much more significant threat to make the postseason. Even if Stro only matches the production he's managed over his career (about 9/5 and a block, or about 15/8 and 2+ blocks PER36), then you're talking about a guy who is still going to shore up their frontcourt rotation. He's a guy who can give Shaq a rest, who can work well with Nash...

This is probably the best situation into which Swift has ever walked. He's as close to a really poor-man's Amare as you can get, which is exactly what Nash has been missing in Amare's absence this year. This isn't an epic steal like Gasol to the Lakers, but it could prove to be an important transaction as the stretch run begins. Swift has been an efficient player when he has shot over his career. He's not an All-Star in hiding, nor is he going to really even approach replacing what Amare brought to the team offensively, but he'll give Nash that extra option and he'll be another efficient tool for a team that's already shooting a hair over 50% FG as a team, which is dangerous when you consider the defensive and rebounding advantages that he will also bring to the team.

If he's given 25 mpg, I'd expect something like 10 and 6 out of Stro on good percentages, maybe even efficiency greater than usual on account of Nash and Phoenix's offensive system. If he can do that while blocking shots, playing good D and hitting the glass hard, then he's going to do a lot for a team that's looking for a booster shot so they can push for that last playoff spot and take it from Dallas.