Print and Go Back ESPN.com: TrueHoop [Print without images]

Thursday, May 15, 2008
When Chris Paul Plays Defense


There really is no good way, that I know of, to use publicly available numbers to tell how good a player is at playing defense.

You can see how many blocks a guy gets, how many rebounds, or how many steals.

Some teams have things like Pat Riley's special secret metrics.

But you really can't tell, by any traditional measures, how good a guy is at causing the other team to turn the ball over or miss.Chris Paul, Tony Parker

One of the only ways anyone is even attempting to divine such secrets is to assess how effectively a team gets stops when a certain player is on the floor, compared to when he is not.

Stephen Ilardi, one of the experts in the TrueHoop Stat Geek Smackdown, specializes in that kind of assessment. He writes:

CP3 is still not an MVP-caliber performer, due to his substantial negative impact on defense. Remember, during the 2007-2008 season, the Hornets gave up 6 more points per 100 possessions when he was on the court than when he was off, and this on-off differential on defense was by far the worst of any player on the team. Similar defensive liabilities, by the way, can be observed with other undersized point guards in the league (e.g., Jameer Nelson), so this finding should come as no great surprise to anyone. (And note that Paul and Parker have been completely unable to shut one another down in the current playoff series).

Bottom line: Chris Paul is a freakishly gifted, MVP-caliber player on offense, but one whose defensive liabilities render him merely very good, but not great, in his overall impact on the game.

The idea that Paul is not a great defender has been discussed by some very smart people on the APBRMetrics message board. Even in that insightful crowd, there is not anything close to a real understanding of what the numbers tell us.

There are a lot of different theories. Perhaps the numbers showing Paul is a bad defender are simply the result of some kind of mistake. Perhaps they are an artifact of small sample sizes. Perhaps they show that he has good defenders playing behind him.

One known problem with +/- is the notion that if you have really good people backing you up, it makes your numbers worse. And the Hornets' bench plays great D. So when Paul comes off, the team's defense might get a little better. But does that mean Paul is a bad defender? Not so fast.

Ilardi, however, addresses that concern:

Given this potential concern, it's crucial for us to rule out the possibility you raised by looking at the other Hornets starters to see if they have a similar negative rating on defense ... and what we find is that no other starter is close to Paul in terms of his negative on-off defensive rating.

For example, David West, who shares a VERY high number of on-court minutes with Paul, actually has a solid on-off defensive rating (opponents score 1.47 fewer points per 100 possessions when he's on court).

Thus, because Paul's on-off defensive value is by far the worst on his team, there's no way it can be a mere artifact of the high-caliber defense played by the Hornets' bench.

Hmm ...

I asked some other stat geeks what they thought about Paul's defense. Justin Kubatko of Basketball-Reference explains his basic assessment:

Assessing individual defense is hard, but I rate Paul as a very good defensive player.

This rating is primarily based on three things:

1) The Hornets were a very good defensive team this season, finishing 7th in the NBA in points allowed per 100 possessions. Given that Paul played the most minutes on the Hornets this season, he deserves some of the credit for that.

2) I estimate that Paul recorded a steal on 3.9% of the opponent's possessions while he was on the court. That was the best mark in the NBA this season by a mile. To put that in perspective, the runner-up, Ronnie Brewer (3.2%) ,was closer to tenth place than he was to first place.

3) Paul is an above average defensive rebounder for a guard, and his defensive rebounding is even more impressive when you consider his size. I don't know exactly where Paul ranks among NBA guards defensively, but based on the last three seasons I would definitely put him in the top ten in the NBA.

ESPN's John Hollinger had some insight:

He's great off the ball, decent but hardly spectacular on it. He's not nearly as vulnerable to post-ups as you might think given his size, as Dallas learned in Round 1.

Kevin Pelton (of the Sonics website, and many other publications) has a very thoughtful explanation:

Dean Oliver [author of Basketball on Paper and a Nuggets' statistical expert] has done a great job of emphasizing the idea that the numbers tell a story, and what good analysts do is use them to interpret and tell these stories. Sometimes that's an easy process; sometimes there's contradiction, and it's hard to come up with something that makes sense. Chris Paul's defense falls into the latter category.

When I was coming up with All-Defense teams, I threw a bunch of defensive numbers into a spreadsheet. Based on his box-score stats (rebounds, steals, blocks, fouls and team defense), Paul comes out as elite -- only Jason Kidd (on the strength of his rebounding) and Rajon Rondo rated higher amongst the players I looked at. However, his net defensive plus-minus was the worst of the group, and the PER posted by opposing PGs with Paul on the floor was second-worst.

Watching Paul play, his quick hands are clearly a huge strength, but he only seems average as a one-on-one defender. I was really impressed the other night with the way he dealt with a much bigger player in Manu Ginobili when paired with Jannero Pargo in the backcourt, and his strength works to his advantage.

Overall, I picked Paul as an honorable mention and would probably call him one of the top 5-8 defenders at the point (ranking him higher in terms of defensive value because he plays so many minutes compared to specialists like Chris Duhon) ... but that's still a very tentative ranking, not a confident one.

So, I guess you can put all that together and conclude ... what can you conclude exactly?

Don't you just find yourself wanting more infomation?

Video! I fired up Synergy and watched dozens and dozens of recent Paul defensive possessions.

It's fascinating, really.

In recent games, mostly he's guarding Tony Parker. And clearly, the Hornets are playing with a strategy that is very determined to defeat the points in the paint that drive San Antonio's offense. Most of the time, all five Hornets have at least one foot in the paint, making it tough for Tim Duncan to operate on the block, and frustrating the drives of Manu Ginobili and Tony Parker.

So, at any typical moment, there is Paul, a foot in the paint, cheating well towards wherever the ball is, even if it means leaving many yards between himself and his man.

In fact, I get the feeling from watching this that the Hornets really want Tony Parker to shoot threes, which he's not so good at this year.

But here's the thing: When the Spurs do swing the ball to Parker, and it's Paul's turn to close the gap, his lack of height and length become so apparent. Here's wher
e I think about Ilardi's first point, about how Jameer Nelson has similar statistics. In this situation, it's way better to be Morris Peterson or Julian Wright, you know?

Paul really never bothers Parker's shots. Even when he can close the massive gap, he's sometimes still so far from the ball that he doesn't even bother to put his arms up.

If Parker gets hot from downtown, someone else is going to have to be the one to try to stop him.

In fact, knowing how competitive Paul is, it's no wonder he has trained himself to become the steals master. They give him a way to be competitive at that end of the floor.

Paul doesn't just give Parker a lot of room. He also gives that half-acre to Manu Ginobili, Michael Finley, and Bruce Bowen.

He just doesn't get out there to bother shots, even against great shooters.

The Spurs have obliged by missing a ton of wide open looks in many of their games. But the open three-pointers are there, and they are there in large part because of a combination of a clog-the-paint defensive philosophy, and Paul's lack of size.

Here he surely hurts his team.

He's also not exactly deadly in getting around screens. Whereas Jannero Pargo often manages to stay with his man even chasing him around Robert Horry or Tim Duncan, Paul typically goes well under -- giving up a mid-range jumper that Parker can hit -- or gets stuck a step or two behind, leaving the nightmare of an unchecked Parker freelancing on his way to the hoop.

On another play Paul seemed to have a miscommunication with Morris Peterson, and Paul ended up leaving Bruce Bowen WIDE open, I mean, as open as NBA players ever get, in the corner for his trademark three. (Bowen missed.)

The worst play of his that I saw was a miserable bit of getting stuck on a pick chasing Manu Ginobili. Ginobili curled around a pick for a wide-open layup in the half-court. That's the exact shot the Hornets are clearly most interested in stopping, and Paul let himself get trapped rather easily, even though at that moment he was chasing Ginobili away from the hoop.

All that said, Paul is not at all a lazy defender, and he's certainly not stupid. He is dedicated to what he can do well: Help to pinch off drives, and mess up the offense with the threat, or the practice, of stealing the ball.

He's such a good thief, that he sometimes comes up with steals even when no steal opportunity seems to be there.

Do you remember a play where Tony Parker fell down curling around a pick at the left elbow? Paul ended up with the ball, and after watching it a dozen times, I still can't tell what happened. Did Paul poke the ball free? Did he knock Parker down? Did Parker just lose the ball and then fall all on his own? Who knows, but those things happen around Paul, and it helps the Hornets.

That's what Chris Paul can offer a defense: Non-stop effort, smarts, and playing within the scheme. But he's not able to do many of the things that bigger guards can do, which, I suspect, is why the numbers show the Hornets have not been as effective on defense when he is on the floor.

Now, in the final analysis, does this matter? Only in the sense that he's a relatively new product in the NBA, and it is worthwhile to get to know the strengths and weaknesses of his game. Also, if he ever gets way better at defense, he should be appropriately celebrated for those effort.

But in the meantime, he's such a good overall player, that of course it's not like you'd ever bench him.

(Photo: Ronald Martinez/Getty Images)