“Love looks not with the eyes, but with the mind,
And therefore is winged Cupid painted blind.”
– William Shakespeare A Midsummer Night’s Dream
One of the things , I consider an unique (and pleasant) feature in my life is that I’m constantly getting smart and interesting basketball questions from very smart people.
Questions, you see, are what drive me. It’s questions that lead me to spend 100 hours of my own time diving into 600 thousand game databases.
Interesting questions are even more precious. Interesting questions inspire me to build and create tools that deepen my understanding and increase my ability to answer future questions while blowing apart conventional wisdom.
The very best questions do all of the above and lead me to an interesting story.
Case in point, I got the following query the other day (I’m paraphrasing):
Had a research thought the other day. Is it possible that in the few games after a team loses a major scorer that performance of the team suffers as the team figures out a new shot allocation? But then after that period, players return back to what they were.
It seems like people perceive that losing a major scorer is a problem .And maybe in some cases it could be. But I also think this issue should be shortlived. ……
This is a primo, grade A1 question. I immediately built a tool that lets me look at just that using the tool library I’ve been building .
And I proceeded to look.
(Note: When I say Team Ratings in this context, I’ll be referring to ratings as defined by Dean Oliver in Basketball on Paper and explained here .)
Needless to say, as great a question as this is the response will be much longer than one measly post but I wanted to give you a trailer to wet your appetite.
I love trailers.
Let’s focus on the most recent famous cases of trading a major (and inefficient) high volume scorer:
- Iverson from Philly to Denver
- Melo to the Knicks
- Rudy Gay #1 (Memphis to Toronto)
- Rudy Gay #2 (Toronto to Sacramento)
- Josh Smith being cut
We’re going to focus on the impact to the team cutting the high usage, low efficiency scorer. Did the reality match the narrative? Did their old teams have trouble creating offense after divesting themselves of such notable shot “creators”?
Let’s start with the Answer first (Iverson):
Philly for Full Season in 2006:
Net Rating is -3.3, Offense Rating is 104.4
Pre Trade Iverson trade:
Net Rating is -6.3, Offense Rating is 103.3
Post Iverson trade:
Net Rating is -2.1, Offense Rating is 104.8
The Sixers got 1.2 points per 100 possessions better post trade and 1.5 better on offense per 100 poss. They did not really miss AI at all. In the short term it was even more pronounced.
The first 10 games after the trade:
Net Rating is -2.8, Offense Rating is 105.1
First 5 game it was:
Net Rating is -3.6, Offense Rating is 104.3
Here, I’m tempted to say that the benefit was immediate and actually tapers off as defenses adjust to not being able to lay off the volume chucker.
Let’s see Melo:
Full season Nuggets:
Net Rating is 5, Offense Rating is 112.9
Net Rating is 2.5, Offense Rating is 113
Net Rating is 10.6 Offense Rating is 112.6
First 10 games:
Net Rating is 13.8 Offense Rating is 113.9
Hmm second data point lines up with my assumption. Chuking the chucker has a positive benefit (better than Cheerios!) and is actually better in the short term. This is likely because defenses don’t know what the team’s offenses will look like without the chucker. It does taper off somewhat.
Rudy Gay Number 1 now:
Full season Grizzlies:
Net Rating is 4.7, Offense Rating is 105.9
Net Rating is 4.4, Offense Rating is 105.2
Net Rating is 5 Offense Rating is 106.7
First 10 games:
Net Rating is 3.5 Offense Rating is 105.9
Third point in agreement. This might indicate a trend.
Rudy Gay Number 2 now:
Full season Raptors:
Net Rating is 3.5, Offense Rating is 109.6
Net Rating is -.7, Offense Rating is 105
Net Rating is 4.8 Offense Rating is 111
First 10 games:
Net Rating is 4.4 Offense Rating is 107.4
Here, Rudy Gay was so bad that the immediate jump was there and it kept growing!
Full Season Pistons:
Net Rating is -2.1, Offense Rating is 104.6
Net Rating is -7.2, Offense Rating is 101.2
Net Rating is 6.1 Offense Rating is 110
What have we learned? I think it’s safe to say that there is no intrinsic value in a volume scorer independent of his efficiency. Yes, having a high volume, high efficiency scorer (Michael Jordan, Larry Bird, Dirk, Durant, Lebron) is a game changer because they draw an innordinate amount of defensive attention and create open looks for their teammates. However, the simple act of taking many, many shots is not what creates all those open looks.
Your opponent is smart enough to watch game film. He knows that he should let Josh Smith shoot as much as he wants. By giving those looks and shots to an inefficient spender of possessions you’re producing losses and not wins.
Write this down in stone folks: The simple act of trading or releasing the Rudy Gay/Josh Smith in your roster will make everything better.
Of course, the full answer will be even more interesting.