Danny Granger was the subject of the following two posts in the month of March.
There is Not Much Difference Between Danny Granger and Kobe Bryant?
Bob Newhart, Danny Granger, and Group Therapy in Indiana
Of these two, I really like the one connecting the Bob Newhart show to Danny Granger. In fact, it’s such a good post (well, I liked it) that I should move on to another subject. But the sports media has decided that Granger is the Most Improved Player in the NBA in 2008-09. And after looking at the data, I think I have to comment on Granger again.
Table One: The Last Two Years of Danny Granger Basketball
Table One reports what Granger has done the past two seasons. Let’s start with the non-scoring aspects of his game. Granger did manage to improve with respect to blocked shots, assists, and personal fouls. But he got worse with respect to rebounds, steals, and turnovers. And the decline with respect to the net possession factors trumps the improvement in blocked shots, assists, and personal fouls.
What about scoring? Granger also increased his scoring totals. But his level of shooting efficiency really didn’t change. So Granger’s scoring totals only went up because Granger took more shots.
Now Granger is an above average scorer, so more shots from Granger will increase his Wins Produced and Win Score. The overall improvement, though, is rather small. As a small forward Granger posted a 0.148 WP48 [Wins Produced per 48 minutes] this year. Last year his WP48 at small forward was 0.124. Had Granger spent all his minutes at small forward (he also played power forward, which is not a good idea), this improvement would result in 1.2 additional wins for the Pacers. So yes, technically Granger’s additional shots did mean he was more productive. But the difference seems rather small.
Of course it’s possible that the sports media looked at every aspect of every NBA veteran and this is the biggest improvement they could find. I suspect, though, that something else is going on. I think many members of the sports media saw that Granger increased his scoring average from 19.6 to 25.8 points per game. This scoring leap was enough to impress these members of the sports media who often only look at scoring totals in evaluating players.
Now if I was really working hard at this blog tonight I would go look at every player who played in 2007-08 and 2008-09 and discover the one player who was the Most Improved. Unfortunately, I am less committed tonight and think I will just end the post here (if you need more, please go re-read the Bob Newhart post…. I really like that particular story).
- DJ
The WoW Journal Comments Policy
Our research on the NBA was summarized HERE.
The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score
Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:
Simple Models of Player Performance
What Wins Produced Says and What It Does Not Say
Introducing PAWSmin — and a Defense of Box Score Statistics
Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.
Really like this blog. Not impressed with Granger that much, either. Question: Is anyone calculating, and posting, season WP48 numbers for all players? I’m curious, but have neither the time nor the industry to do the work myself.
I should get numbers posted for each team throughout the summer.
Thank you for the response. I really enjoyed WoW, as well. Don’t have the statistics background to follow the nuances, but I thought the principles cogent. Best of all it was an entertaining and informative read, with what I considered fresh perspectives. Looking forward to the next one.
“Danny’s one of our hardest workers. He spends a lot of time on the court working on his skills and trying to improve them. That doesn’t go unnoticed. That’s why we signed him up last year. He’s a special player, and he gets better every year.”
Quote from Larry Bird sums it up really. Not sure if Granger deserved it statistically speaking, but I’m glad he won it nonetheless.
Does anyone know the Pacers’ record with/without Granger? It could be a nice way of illustrating the significance of him playing from a win/loss standpoint.
Phil the Pacers played 15 games this year without Granger in the line up and finished 9 – 6 in those games. However on a per 100 possessions basis the team was +1.1 points when he was on the court, and -5.6 points when he was off of it. The net points differential was +65 when he was playing, and minus 156 when he sat. This means the net difference in points was +221 for Granger this year. So despite the Pacers record whilst he was injured he clearly helped them overall. Also his Roland Rating of +8 was by far the highest on his team. PER tells the same story (ie he led his team) but Win shares gives Troy Murphy the edge (8.4 to 9.0) because of his superior defensive win score.
Well, it would have been embarassing to give the most improved award to Kevin Durant after giving him rookie of the year last year, as you previously noted here:
http://dberri.wordpress.com/2009/02/03/can-kevin-durant-be-the-mip/
I bet Roger Mason, Jr. improved a lot.
Why would it be embarassing to give Durant most improved player if he was the most improved player regardless of what awards he recieved as a rookie. Usually the rookie of the year is not nearly as good as the elite players in the game. I am not saying Durant was the best rookie last year. He clearly was not and clearly should not have won the award. I am also not saying he is one of the best in the game this year. He clearly is not. I am just saying that the gap between elite rookie and NBA superstar is every bit as big as the gap between Denny Granger in 08 and Granger in 09, even from the perspective of the “typical” NBA fan. Therefore, going from Rookie of the year to Most Improved should not be unthinkable. If the “typical” fan felt Labron James was only as good as the best rookie last year, he almost certainly would have won both MVP and Most Improved player this year.
Michael,
Your comment got me thinking, so I went to 82 games and did some looking into those numbers. Of the Pacers top 20 combinations of on court players Granger was only absent from 1(a unit that played a total of 36 minutes). Isn’t it hard to draw any useful conclusions unless we compare that unit to its closest counterparts?
I was looking over New Orleans to units from last year because on/off court numbers said New Orleans was a much better defensive unit with Chris Paul off the court. However, once I looked over those numbers I realized that they were heavily skewed by one specialized unit which was able to compile a .92/100 defensive rating without paul on the court that help me put those numbers in perspective.
If one player is going to be so integral to a team’s plan I think it makes it tougher to try and gauge their worth with on/off court numbes because they are most likely going to suffer from some severe biases.
I was listening to Colin Cowherd this morning. He said ” I call 97-07 the years where short guys shot too much”. Then he made a direct reference to Allen Iverson’s teams all getting better once he left or worse once he got there. His conclusion was that because of Michael Jordan the league marketed shorter guys who shot a lot and the game suffered because of it. I felt as if he reading cliff notes from this site. So, not all media guys are bad.
Yeah, without any number-crunching, I vote for Kevin Durant as MIP, especially after coach PJC was fired, and KD got (properly) moved from SG to SF… his immediate and drastic improvement was eye-opening..
So, not all media guys are bad.
Just the one with votes are :P
Do any of you watch PTI or Around the Horn much?
If so, does it scare any of you that a lot of those guys get to vote for all these awards?
I mean, there’s no way that they analyze the data using any metric, let alone a metric like this one that a lot of us like. Heck, with all the TV and writing they do, I’m guessing most of them don’t have time to watch many games either.
To me, that makes a lot of these decisions borderline fraudulent.
And this is a great case in point. The only significant ‘improvement’ with respect to Granger’s statistics is scoring average. But ‘improvement’ is questionable at best when we consider that his overall efficiency (on offense, anyway) didn’t improve. He just shot more.
It seems to me that we shouldn’t expect anything different, given the voting system we have in place. If the people who vote don’t have (or maybe don’t spend) the time to actually watch a high volume of games and then analyze the data, should we really expect anything different?
brgulker,
If I was running the league I would absolutely allow the media to shape perception. Otherwise, how would my league get publicity. That being said, that doesn’t mean I have to take their thoughts with anything other than a grain of salt.
^Except for when you agree with them lol :-D
Also can you link me to the page which has Granger in 19 of the top 20 because looking at the top 20 five man floor units I can see him in 15 of them and the five he wasn’t part of combined for 348 minutes! Don’t forget that Granger was injured for 15 games this season!
Michael,
I screwed up. He was in 19 of 20 of the previous seasons 5 man units. However, I hope that doesn’t take away from my point about using these small sample sizes for on/off court. I have no idea what the threshold for minutes should be but I am definitely still uncomfortable with 348 minutes worth of actually useful lineups though that is substantially more than the 36 minutes I was originally referencing.
And regarding agreeing with the media, sometimes people can use big words correctly without knowing what they mean. Similarly just because I think they got the first team all nba team mostly right doesn’t mean I agree with the method they used to arrive at that conclusion
Pingback: The Inconsistent Troy Murphy and the Consistent Pacers « The Wages of Wins Journal