Lies, Damned Lies and Statistics
by Steve Perrin on Jul 25, 2008 6:08 PM PDT
in News
Email
PrintFirst of all, I've always understood that Mark Twain coined the phrase that serves as the title of this post. Not so. Apparently, Twain popularized it's use in the United States, but even in his original usage in "Chapters from My Autobiography" he attributes the quote to Benjamin Disraeli. So there, I learned something today.
I love statistics as regards NBA basketball. There is much subjectivity in the game, and statistics provide an element of objectivity. But there is little question that they can be misused. For my part, I strive to understand the stats that I see, and even moreso the ones that I use. A statistic in a vacuum is a very dangerous thing. There needs to be some context.
What got me thinking about this was the discussion about Biedrins versus Camby in a FanPost today. Biedrins led the NBA in field goal percentage last season at almost 63%, while Camby shot a seemingly anemic 45%. True, comes the counter-argument, but Camby is a better shooter from the perimeter, and the reason Biedrins' percentage is so high is because he only shoots layups and dunks. However, comes the retort, if Camby is such a great shooter, why does he only make 38% of his jumpers, and shouldn't he take the hint and stop shooting so many - 57% of his shots are jumpers, after all.
This 57% jumpers, 38% on those jumpers stat was first thrown out there by Jeremy on Pickaxe and Roll as an indictment of Camby's offense. And let's face it, those are some ugly looking numbers. But let's put them in context.
Here is the same 82games data for five big men: Biedrins, Camby, Kaman, Tyson Chandler and Mr. Fundamental himself, Tim Duncan. (Sorry, I don't know if you can get these numbers side-by-side in 82games - I couldn't, so I'm just hopping between tabs to look at everything).
Are Camby's numbers ridiculously poor? Well, Camby was 37.7% on 57% of his attempts. Kaman was 40% on 38% of his attempts. Better on both counts, and certainly an encouraging trend for the newfound Kaman face up game. Biedrins and Chandler, first and second in the NBA in field goal percentage, take very, very few jumpers by comparison - Biedrins 11% and Chandler 16%. Their shooting percentages are atrocious on those jumpers (they barely make 1 of 4), so it's a good thing they don't take them. So it becomes pretty obvious how a couple of 'horrible' shooters lead the NBA in shooting percentage, especially when you see that fully one third of Chandlers buckets were dunks. ONE THIRD! He could only make one in three of all his other shots and still shoot 55%!
But what about multi-MVP, first team all pro, first ballot hall of famer, Duncan? He made 37.5% of his jumpers last season, and 55% of his shots were jumpers. He sucks! He actually shot worse than Camby, and took almost as many jumpers as a percentage of the overall.
Unfortunately, I don't know the exact definition of a 'Jumper' in this statistic. I know it's not a 'Dunk' and not a 'Tip' and not 'Close.' But where does 'Close' end and 'Jumper' begin? 5 feet? 10 feet? 12 feet, 4 and one half inches? I just don't know.
I do know that the 82games statistic seems to be out of sync with the NBA HotSpots feature. Go there and look up Tim Duncan and you'll see that they show 711 of his 1178 field goal attempts last season coming from within the closest zone they track, which is inside of the lane and well below the dotted line on their graphic - we're talking 5 feet and in. So it's hard to figure how 82games has him shooting 55% jumpers when the NBA.com HotSpot has him taking over 60% of his shots with 5 feet of the basket. Does it have to do with the form of the shot? Anything shot facing the basket with a jump shot motion, no matter how close, counts as a jumper? Hard to say.
So, all of this to say, I don't know what most of this means. It's easy to accuse players of 'settling' for low percentage jump shots, but there's a huge difference between a seven footer and a 22 footer, and just how many of those shots are being taken with 5 or fewer seconds left on the shot clock, when the guy has no choice? We just don't know.
Damned statistics.
19 comments
|
0 recs |
Do you like this story?
Comments
82games' "close" stat
I wish I could recall the source, but as I understand it, 82games regards “close” shots as ones that come from 6’00” on in.
At shooting the outside jumper?
He does.
But to say Duncan sucks would seem to invoke another classic Jax strawman.
Camby doesn’t suck either. The idea that he should be shooting 19 footers as a primary offensive weapon does.
Get me BD and 75 and I'm in
OK
Even I knew he was being sarcastic.
The Clippers! The (second) Best NBA Team in LA!
by Steve Perrin on Jul 27, 2008 9:28 PM PDT up reply actions
Sarcasm comes with a point
The true next level discussion would seem to be, is there a difference between Camby taking bad shots and Duncan taking bad shots, and if so, what is it? And does the value of either exceed Biedrins just not making bad shots.
Using that hot shots feature, Biedrins produces less than 2ppg less than Duncan around the basket. So the difference between the players is those attempts everywhere else. So Biedrins is like 2-12 not at the rim. The other possessions aren’t turnovers. Some other Warrior was using those possessions. Duncan was 109-321 from just the bracket out. 34%! Beyond that is just 145 total shots for the season, or less than 2 per game. Why isn’t Duncan concentrating on getting more dunks and layups and leaving the mid-range and out stuff to guys like Ginobili?
The answer to that question probably contains the truth about the ever elusive value to “shot creation”.
But that’s the thing about sarcasm, right? Its like the Corey Maggette of discussion. It brings one strong thing to the table, but takes so much away.
Get me BD and 75 and I'm in
At least it's not the Q Ross of discussion
There are a number of problems with Biedrins, some of which I’ve mentioned, such as his poor FT%. He also is exceedingly foul-prone. For that reason, he only played 27 mins / game last year. As I’ve said, Biedrins isn’t bad, but he’s no Camby. In addition, he plays for a very flexible coach, which helps his game.
There are so many things wrong with your analysis on the Duncan / Camby / Biedrins comparision I wouldn’t know where to begin. Or have any interest in beginning.
I suggest we move along.
Or have any interest in beginning. Then why did you comment?
Move along any time you like. My comment was directed to CS and was based in a sort of long running thread about the value of shot creation and how it should be represented in advanced statistics. So I’m not sure why you even responded.
There is lots of basketball truth in what value, if any, those 4 ppg Duncan scores from the field beyond like 5 feet. Combine with the 2ppg from the rim itself, we are talking about 6ppg from the field between Biedrins and Duncan.
It shocks me that any basketball fan would find no interest in discerning the value of 6 moderately inefficient points per game. When the difference in points per 100 possessions between the best team and the worst team is only 13 points, those 6 points should be very interesting. I mean the Spurs were smack in the middle of the pack last season in p/100. They were also very low in FTA/FGA, so banging inside more would be better in that regard. Would being more Biedrins like make the Spurs better? Maybe so, maybe not. It should be interesting to think about though.
Get me BD and 75 and I'm in
Sorry to hear that a blog post shocked you. Hopefully you aren't hurt too bad.
One of the reasons why I think we should move along is because it doesn’t appear that you understand how the offenses run by the various teams, and the overall play of the players involved, enter into the equation. Instead, you extrapolate things like “values” solely from naked statistics. In that regard, I think I’ll take Coach Pop’s evaluation of Timmy’s role in his offense over yours, with all due respect.
And while I understand that you were responding to CS, his post was about my post, and your Corey Maggette reference was obviously an attempted dig at me, and as such your assertion in that regard was unnecessary, vapid, and silly. For some reason, you continue to be obsessed with me. I’m not interested in a running battle of insults with you, and I would imagine that most of the bloggers here would share my lack of enthusiasm in that.
If, however, you’d like to settle your obsessions on the basketball court, let me know.
Until that day, however, I suggest that you move along.
The Maggette Reference
The Maggette reference was actually directed at me – John R and I have had this conversation many times before. The question of number of jump shots and shooting percentage on jump shots reveals a similar trend with Maggette.
At some level, when a player ‘settles’ for a jump shot, it’s just that. They settled for a jump shot. Maybe they had to. Maybe it’s what the defense gave them. Maybe there were 2 seconds on the shot clock and they had the ball in their hands standing 20 feet away from the basket. It’s easy to suggest that teams ‘should’ work to get a better shot or that Duncan ‘should’ only shoot from within 6 feet, but the defenses know these percentages too (and even if they don’t, it’s pretty clearly the goal to push the offense further away from the basket). NBA players shoot between 95-100% on dunks. If they only ever took dunks, there would be very few misses – and a lot of 24 second clock violations. So it’s a continuum from taking a really great shot like a dunk where your percentage is in the high 90’s, to taking a bad shot, like a jump shot where you’re percentage is in the low 30’s. But bad shots happen.
82games has percentages for jump shots and percentages for when in the shot clock shots happen. What we really need is jump shots early in the shot clock. A Tim Duncan 18 footer with 12 seconds on the shot clock would be a bad shot.
The Clippers! The (second) Best NBA Team in LA!
by Steve Perrin on Jul 28, 2008 10:56 PM PDT up reply actions
I don't see the connection
The Maggette reference was a general rip on Corey, a comparison of Corey to sarcasm, not a reference to jump shots, in my opinion.
Why not
email Eric and ask him why his number is over 50% where other sites have it at less than 40%? I have always found Eric to be very responsive and instructive.
Good suggestion
The Clippers! The (second) Best NBA Team in LA!
by Steve Perrin on Jul 25, 2008 11:05 PM PDT up reply actions
More important is the role in the offense
Camby shooting jumpers (even if he misses a lot) makes room near the basket for A.I. and Melo (And JR smith), whereas Biedrins is left unguarded when more then 6 ft from the basket. It’s not like Camby can’t finish inside eather.
On skill set alone, I’ll take Camby over Biedrins any day of the week (durability is another matter).
One thing is forgotten
In the NBA, you don’t have to actually be able to hit a shot out there to draw your man away. If you want to clear the lane, you put him weak side and clear out of the lower defensive box. If the defense doesn’t want to go with him, their options are dictated by what sort of zones are allowed by rule. And the rules are designed to give the offense the advantage.
You can place your center anywhere you want. That doesn’t mean its a good idea for him to actually take a shot from there.
Get me BD and 75 and I'm in
i'd take beans
on defense. camby is definitely better on o right now. But he is slowing down on d, and does not really stop players going to the hole, unless he blocks them, which is still a lot, but he camps down low. Biedrins is more active on d, from what i can see.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Certain photos copyright © 2012 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
