Tuesday, March 24th, 2009...12:00 am
Nazr Proof Stats
Hey guys, remember me?
Remember when Nazr Mohammed played for the Spurs? With each entry pass, it looked as if he was attempting to catch the ball between the business ends of two tennis rackets. If basketball is theater, his was a predictable 3 act play: executed pass, fumbled pass, change of possession. Despite all that he brought to the Spurs 4th championship run, his hands were a hard pill to swallow.
It had to grate at Tony Parker. Many of those fumbled balls were credited to TP as turnovers. How unfair, right? I wonder if Parker suspected Pop and Mohammed of conspiring against him in some sort of two man game monkeyshines. “Hey, Nazr, next time he hits you on the roll, just clumsily bat the ball out of bounds. He’ll start cursing in French. It’ll be great, you’ll see.” All joking aside, point guards are often penalized for mistakes on the receiving end of the pass. Raw assist to turnover numbers can misleadingly suggest a player is more careless than is actually the case.
Where am I going with this? I’ve read the comments of our readers with great interest whenever a discussion arises that includes recourse to new breed stats. I’m sympathize with commenters who think these stats sometimes miss the forest for the trees; that they unnecessarily muddy the waters of common sense with a fog of statistical esoterica. But, in general, I deeply appreciate the changes brought to us by the emerging stat geek army. They help us to see the game with greater clarity.
This brings us back to Nazr Mohammed. Earlier today, I was pleasantly surprised to “discover” (new to me) a stat called assist to bad pass turnover, tracked by the good people at 82games.com. It’s just what you’d guess. It weeds out turnovers that come off fumbled passes, dunderskulled moments of distraction, and the like. Traditional assist to turnover numbers rate Tony Paker at 2.62, or 32nd in the league. But when you adjust this to only include bad pass turnovers, Parker’s number jumps up to 7.0, second best in the league–only a tenth behind point guard wunderkind Chris Paul. I understand one might quibble with a straight comparison of these numbers. But you have to admit, having nuanced ways to evaluate assist to turnover numbers provides hoopheads with a more intelligible, detailed story.
This is an example of the positive power of the coming statistical revolution. But more importantly, it gives us a much greater appreciation for Tony Parker. His recent scoring outbursts are great, but let’s not forget he’s one of the very best at taking care of the basketball. The term “scoring point” is sometimes used derisively. This is unfair. It seems too narrow a conception of what a point guard can be to limit the position to “scoring” and “passing/true”. In some ways, Tony Parker demolishes such rigid categorization. Couple his 7.o assist to bad pass ratio together with his 21.4 ppg and what do you get? A point guard who is a careful, conscientious distributor and an elite, call-my-own-number scorer. Or, in other words, these numbers provide a window on one of the league’s elite point guards.
22 Comments
March 24th, 2009 at 4:57 am
Great analysis Tim. I know there’s been a lot of TP-bashing going on in the last couple of games, and this post puts Tony, for a change, in a much better light.
I hope these terrific stats can also help us track how Tony has matured throughout the years – if his assist-to-bad-pass turnover ratio has steadily improved each season just like his other stats. But still, just finding out that he’s second to CP3 in such a category makes me appreciate our little French PG even more.
March 24th, 2009 at 5:26 am
The issue with stats is always that it’s a team game, therefore they have to be analyzed in the context of the team. The other stat that makes me smile is that people compare 20 ppg on a team like the Spurs that rarely scores above 90 points a game, and 25 ppg on a team that routinely scores over 110…
Coming back to Parker, I was actually quite pleased by the reactions I read here and in other forums where there’s usually not a lot of love for him. The Spurs fans tend to have a balanced view and took the high road by pointing out that most of these games wouldn’t even have been close if not for Parker’s heroics. I wish he would have put these shots in, but as many others noticed, the real problem is that somebody else should be stepping up at some point.
Maybe the most representative stat recently was the one pointed out by Hollinger against Dallas: when TP was on the floor he outscored the whole Dallas team! It’s been a bit too much of that recently for the Spurs. Hopefully the offensive load will be better distributed during the playoffs – this is why we need Tim and Manu healthy more than anything.
March 24th, 2009 at 5:36 am
Why scrutinize Tony when we have such more deserving targets on the team? How about a new stat calculating the amount of times per game a defender half-heartedly waves his hands as an opponent dribbles past him?
I’m curious, what was FInley’s plus/minus in the Rockets game?
March 24th, 2009 at 5:52 am
Finley was +11 in the Rockets game.
The +/- is far more helpful when taken over a longer period of time.
The Spurs 5 man combos, for the season, are here. The single player accumulation is available here. Bonner leads the team at +222 and Finley is 6th at +134.
March 24th, 2009 at 6:21 am
As mentiod by Will as well as Graydon in a post last week about stats – they just don’t tell the story. I think what John Hollinger is trying to do at ESPN is interesting, but basketball is not baseball. Stats are a way of comparing players to each other when you don’t really watch other teams too often.
I like this new stat presented by 82games, because it does show Parker’s value. Of course, us Spurs fan didn’t need to see the stat to know how valuable he is to the team. Even those TP-bashers know the Spurs wouldn’t be the same team if we 26 of the other 29 start PGs in the league.
March 24th, 2009 at 6:25 am
Yes, some new stats are more helpful than others. As you point out, the beauty of this assist to bad pass turnover stat is that is a) simple and b) helpful.
March 24th, 2009 at 6:26 am
Well, the most recent slams on TP have been about being clutch. Doesn’t someone keep a clutch stat?, something about scoring in the last 5 minutes and then 3 minutes or some such? I recall that Manu was near the top. Where does TP fit? That might help complete the pic of the wee frenchman.
March 24th, 2009 at 6:31 am
Agreed, Tim. I might just fall in love with this assist to bad pass T/O stat… I’ll have to give it a little time to grow on me. It’s nothing like the complex PER in which you have to read a 200 page book to figure out what exactly goes into it. In the end you just give up and figure, “If Lebron is at the top, then it’s probably a good stat.”
March 24th, 2009 at 6:37 am
82games keeps clutch stats here.
Parker’s numbers during clutch time are well below his regular season averages. During clutch minutes he shoots .400 from the field, 77% from the line, and has a +/- of -11.
The point of my post was not to defend Parker against this criticism, but rather to highlight that in terms of taking care of the basketball, he is nearly peerless.
March 24th, 2009 at 6:42 am
I really like the “shots created” stat as a judge for point guards, and I am pretty sure that Tony Parker is either at the top or near the top of that all of the time.
March 24th, 2009 at 6:48 am
I’ve not seen that stat. But I’ve lobbied elsewhere for a points created stat. I’d love to see both shots created and points created in a re-invented box score. Do you know where I could find an up to date shots created stat?
March 24th, 2009 at 7:18 am
Talking abt stats, do you know if anybody keeps track of assisted free-throws, i.e. the player receiving the ball is fouled to prevent an easy basket but then makes the FTs?
March 24th, 2009 at 7:23 am
Navin,
No, I don’t. But you may be interested to read my article “Bruce Bowen and the Evolution of the Box Score” from last year.
To respond to your comment, I think it would be helpful to tally assisted free throws. I’m of the persuasion that the NBA and its media partners need to re-invent the box score. But that’s getting beyond the scope of my Parker post.
March 24th, 2009 at 7:44 am
I hadn’t seen that Bowen article before and it’s interesting, but doesn’t that sort of thing get kind of subjective after a while? The assist already is pretty subjective, as evidenced by the fact that the league will occasionally overrule one. Coming up with stats like a Gretzky allows even more personal preference of the statistician, in my opinion.
The answer here is, if you want to compare players to one another, you simply have to watch more basketball.
March 24th, 2009 at 8:59 am
Interesting stat especially when you consider that 82games must have pored over hours of game tape to figure all of this out. I have one major concern though. If counting only Tony’s “bad passes” makes his ranking jump from 32nd to 2nd in the L, then what does this say about our players catching the ball? How many “Nazr’s” do we have and who are they? Is there a corresponding stat to figure this part out?
March 24th, 2009 at 9:00 am
Bryan,
I don’t agree. I think stats are generally objective. There is some subjectivity, but it’s marginal. In general, stats are reliable. Otherwise, why track even the most basic of numbers, such as rebounds or assists.
One thing to remember: most team scouts already track most of the stats were are talking about, and use them to evaluate free agents and the like.
March 24th, 2009 at 10:09 am
I definitely agree about stats being generally objective, Tim. My point was just that if we start getting carried away with stats that we keep, a lot more subjectivity starts being included.
I do realize that most teams keep their own internal stats to evaluate players. That article that you guys posted a while back from the NY Times about Shane Battier was a really interesting read. Conveniently, Simmons had an article today on ESPN that was a good read about the statistical revolution in the NBA.
All in all, I’m a mathematics guy… I’m an engineer. I’m just not so sure you can properly rate guys with respect to each other using statistics in the sport of basketball. Different guys will always bring different things to the table, which I think makes comparing players very difficult.
March 24th, 2009 at 10:28 am
Bryan,
Fair enough. And it’s true that “bad pass” is much more subjective than, say, a deflection.
Now, I’m off to read a Bill Simmons article…
March 24th, 2009 at 1:48 pm
While assist to bad pass ratio seems to undo the statistical damage a guy like Nazr can inflict on a guard, (if I am understanding the stat correctly) it also lets the Marburys of the world off the hook for getting the ball stolen while they are on the dribble. I think it is a useful stat when considering a player’s passing ability/care, but incomplete as a measure of how well they protect the ball in general. I’m not sure how much that would help or hurt Parker’s rank specifically, but it seems that the clean picks more often lead to a breakaway dunk at the other end and are therefore more detrimental than most passing mistakes.
I come most recently from an Ultimate (Frisbee) background and I can say that Ultimate is lightyears behind basketball when it comes to meaningful stats. However, we’ve always had the good sense to separate throwing turnovers and receiving turnovers. Basically, if you get your hand on a pass without having to lay out for it… that one is on you. It doesn’t seem to make much sense that basketball has ever placed the blame so often on the passer… but this would not be the first nonsensical tradition in the NBA.
March 24th, 2009 at 7:25 pm
Excellent discussion. Another stat tangent, 82games introduced a new stat, a “Player versus Quality of Opponent” comparison versus good teams, average teams, and below average teams. The Spurs top players were pretty steady no matter opponent, except Bonner. He upped his scoring about 2.5 points versus good teams.
All this to state the obvious, we really need Bonner to continue to kick it up a notch against good teams, or he’s a real liability on the floor.
Thanks for all the help, Tim.
March 24th, 2009 at 11:07 pm
It seems like this stat would really reward guys who tend to overdribble the ball as well (like Parker and Chris Paul) rather than risk tough pass. It sounds like travels, steals off the dribble, etc would not be included in this comparison.
I like the Episode IV suggestion from above. Give blame where blame is due. If a pass hits you in the hands and goes out of bounds, that’s a turnover against you, not the guy who passed it to you.
March 25th, 2009 at 9:22 am
To change sports, but this reminds of passes the bounces off of a wide-receiver’s hands and into a the defence’s creating an intereception that is credited to the quarterback. That always drives me crazy. Somehow stats have to reflect that.
Leave a Reply