Minnesota Timberwolves Assist Distribution
At Hickory-High I created a series of graphs showing how each team's assists per games are divided among distrbutors and shooters. This link will take you to the Timberwolves' graph. No surprise, Rubio and Ridnour are handing our plenty of helpers. Love and Beasley have been the main beneficiaries.
You can check out the rest of the league here.
I'd love to hear feedback. Easy to read? Interesting? Useful?
4 months ago
Ian Levy
68 comments
3 recs |
Comments
I like it.
Although it took me a moment to get it, I think it’s easy to read. I like the comprehensive feel it adds to the assist rates you can find at places like 82games.com.
Not sure if it would be too much of a hassle, but including something like the players’ eFG% after their names might be interesting…
Also, I think Wes Johnson needs a graph for “passes received that should have been assists…”
holy cow
your graphs are in team colors
by dontbesomean youngfella on Jan 31, 2012 1:20 PM CST reply actions
%
does that help
Authentic frontier Gibberish
by FunkDoobious on Jan 31, 2012 1:35 PM CST up reply actions
Yes...
I had assumed you meant it sarcastically, but the fact that the scale of the graphs can change, it was possible that you were looking at area and not at numbers. ;-)
by Krotz the Wall on Jan 31, 2012 1:42 PM CST up reply actions
It looks nice, but
It is kind of misleading. While the height of the spikes is determined by actual data, the area of the “spikes” is more closely related to the two players on either side (in this case by alphabetical order) around the circle. The area tricks your eyes.
It is a nice looking representation, but it is actually kind of difficult to quickly interpret real data from the graph.
It's also initially misleading...
Because the scale of the graph is scaled to the highest number of average assists for a player on the team. Thus, when you have a player averaging 8+ assists, like Rubio, the rest of the guys get squashed to a smaller area. Then you look at the Kings and they don’t have anyone averaging 5 and the rest of the graph seems more expansive. If the graphs themselves were on a standard scale, say 10, then they suss out in a more uniformly understandable way.
by Krotz the Wall on Jan 31, 2012 1:40 PM CST up reply actions
In general this is a must for graphic representations -- if we're looking across teams.
It’s like the stock market. Want to make it seem like the financial sky’s falling? Set the top and bottom of your graph to narrowly surround this week’s results, and every change seems huge. To make things look stable, set the bottom at zero and the top at 15,000.
If we’re just looking at a specific team, though? The point is mostly to contrast the “from” and “to” of assists on that roster. The spider-style areas might be kind of arbitrary, because of the order of the names involved, but this is a quick way to see how dramatic those swings between giving and receiving are…. Does it tell me something about Luke Ridnour’s role on the team, if I was another team’s fan looking in? I think so.
Maybe a doable tweak might be to order each team’s graph according to the roles of the players: starting five across five spots, with the PG always in the same spot and so on. Then eyeballing the graphs across teams would let you see rough differences in style of play. That main page with all the team links then gives you some conspicuous stuff to react to, maybe?
(The sort-of-analogous example I thought of was this two year old assist distribution by player line thing, which let you choose the player to concentrate on too. Yes, Jonny Flynn can’t feed the post, but he drives and finds wing shooters pretty well.)
"First, I'd like to blame the Lord for causing us to lose today."
Relative frequency?
This would put everyone on the same scale.
I also like the idea of rearranging the order based on position.
The biggest problem I have with the graph is that the only data that matters is where the point of the “spikes” meet the line going to the players name, and the area between the data is useless non-information.
by RandyCrouton on Jan 31, 2012 4:15 PM CST up reply actions
It's useless right now because the area doesn't tell us things about the teams.
Get the positions somehow adjusted, and you have a way to compare how teams play, roughly. Yeah, that’s our big new plan (for someone who does fun stuff like this to work on).
"First, I'd like to blame the Lord for causing us to lose today."
In general yes, certainly.
However in a stat that is fairly tightly confined in upper and lower parameters that do not differentiate or deviate much, then standardization can be another useful approach to the graph. In this case, all but 1 player in the league is averaging less than 10 assists. Take the the Kings graph again. Pan that on a 10 assist graph, and 10 assists per game is usually considered that elite level assist performance, you get an even more visually striking representation of just how little the Kings share the ball.
by Krotz the Wall on Feb 1, 2012 8:27 AM CST up reply actions
Yeah, I don't disagree at all.
I’m just saying, within a given team you can still take something from it.
I’d much rather see consistent use of positions and standard scale, to compare teams at a glance.
"First, I'd like to blame the Lord for causing us to lose today."
of course, where Randolph the Gray is concerned
It’s hard to squash “zero”
by dontbesomean youngfella on Feb 1, 2012 1:34 PM CST up reply actions
By season's end, this will be an interesting graph
but so far it just shows what’s pretty obvious to most Wolves fans: Rubio to Love for 2, and Rubio to Love for 3.
Your graph, however, does show how few other Wolves are dishing assists, which is probably why we’re near the bottom in that team category.
Yo ho ho and a FirstRow stream!
cool looking graph
I’d like to see a Pain Distribution chart, so the Pek and Ricky lines for Bonecrushing Picks Given/Received is opposite of the assist chart.
another idea
list players around the circle by position, rather than first name alphabetical. It’d be interesting to see if some teams are consistent by backcourt/wing/post players, or if the distribution is more individualized than that.
by BrazilYinzer on Jan 31, 2012 2:05 PM CST up reply actions
(Yep, I think I just seconded this above.)
That makes the shapes and areas of the graphs across teams carry more meaning. Unfortunately it’s not always easy to make calls like “Who is our starting small forward?” – not even for the Wolves’ coaching staff. ;-)
"First, I'd like to blame the Lord for causing us to lose today."
because no one else gets to touch the ball
If at first you don't succeed, try, try again.
If that doesn't work, cheat.
by TheEvilProfessor on Jan 31, 2012 3:22 PM CST up reply actions
Heh.
They can’t give what they don’t have!
"Oh my, oh my, oh my. Is that great basketball or what?" --Hubie Brown, Jan. 20, 2012
by Cynical Jason on Jan 31, 2012 4:08 PM CST up reply actions
It's an easy way to make yourself look more important to your team than you really are
Just ask Al Jefferson.
If at first you don't succeed, try, try again.
If that doesn't work, cheat.
by TheEvilProfessor on Feb 1, 2012 7:17 AM CST up reply actions
I'd like to see...
- buckets/pts that were not assisted (with or without free throws)
- player stats over league average for position (as hard as position is to define these days)
- normalization across team, like the previous version
- the coaches face in the center (with or without spider legs)
by Zev on Jan 31, 2012 2:48 PM CST reply actions 1 recs
Here's what's most amazing to me
Is that Rubio is the best at assisting others, but he gets the least help in return. A lot has/is made of his shooting percentages and whatnot, but look at Luke in comparison. Ricky clearly does the most on the team to generate his own shot, or put differently he relies the least on others to generate his own shot.
by Dr. Wolfenstein on Jan 31, 2012 4:35 PM CST reply actions
aye
rubio’s ball dominance is not going to result in so many assisted shots for him.
I guess
my thought is that Love, Ridnour, Pek, and all our other scorers – shoot, even ‘can create his own shot’ Beasley – are all benefitting offensively because someone, for some percentage of the time, is creating shots for them. Rubio, on the other hand, despite playing with Love or Ridnour a lot – that is, guys with some passing ability who could kick out of double teams – gets the least assistance with his shot of anybody on the team.
For comparison, Ty Lawson and Andre Miller receive more help.
Kyrie gets more help (but Ramon Sessions doesn’t).
Deron Williams gets more help.
Westbrook gets more help.
CP3 and Chauncey get more help.
Kyle Lowry gets more help.
Derrick Rose gets more help.
All of these guys receive more assists on their scoring chances than Ricky does, and I guess I just find that interesting because I’ve been pleasantly pleased with Ricky’s scoring, and especially his trips to the line. What I hadn’t realized was the degree to which he manufactures all of that himself, especially given that a lot of the above mentioned guys who he’s being compared to receive 0.5 – 1+ more assisted looks at the baskets (meaning easier looks) than Ricky does. Now maybe part of that is that his teammates don’t trust his shot, or maybe he quickly zips the ball away again. To me, tho, this little data reshapes the narrative and context of Rubio’s offensive game just a little bit.
by Dr. Wolfenstein on Jan 31, 2012 9:43 PM CST up reply actions
It is weird,
but it seems that the ball doesn’t move more than a few times, and Ricky’s primed to move the ball. I don’t remember a single time when a player got in trouble with a short clock and passed it out to Ricky where he had to take it or risk a shot clock violation.
Maybe that’s the key, though. Get Ricky involved in the returns and watch everything open up.
"Oh my, oh my, oh my. Is that great basketball or what?" --Hubie Brown, Jan. 20, 2012
by Cynical Jason on Jan 31, 2012 9:49 PM CST up reply actions
I would imagine that what you're seeing here in part due to the fact
that Rubio has been shooting rather poorly, getting a large percentage of his points on made free throws. There is no assist when you can drain the kick out. That being said, it doesn’t seem that Rubio gets the ball back once he’s put it into someone else’s hands. Which is also related to the fact that when Rubio gets it to you, you almost always are in a pretty good position to take a shot, quite frequently uncontested or in a high percentage area.
by Krotz the Wall on Feb 1, 2012 8:31 AM CST up reply actions
We could also simply conclude that Ricky's getting his shots when the defense doesn't respect him.
They sag, he shoots it.
"First, I'd like to blame the Lord for causing us to lose today."
Thanks for the suggestions everyone
I didn’t give much thought to how the players are organized around the circle, but it’s true that the current setup leaves a lot of space devoid of useful information that could be transformed. I really like the idea of organizing by starting 5, it would give a nice look at how each team really organizes themselves.
I changed the scale by team for this one to give as much individual player information as possible. If you look at the Suns graph, where the scale is set by Steve Nash’s 10+ assists per game, it’s harder to hone in on the performance of some of the role players.
I’m sure I’ll update these graphs for later in the season and see what I can do to incorporate some of these ideas. I appreciate the feedback so much, its incredible how active and vocal the community here is. I’m always blown away by it. Thanks again!
There are really only two plays: Romeo and Juliet, and put the darn ball in the basket. ~Abe Lemons
Find me at IndyCornrows, Hickory-High and on Twitter
I calculated it using assist rate by location and makes per game by location from Hoopdata.
There are really only two plays: Romeo and Juliet, and put the darn ball in the basket. ~Abe Lemons
Find me at IndyCornrows, Hickory-High and on Twitter
Batum is day to day with that knee injury
I want him!!
by RaysOfHope on Jan 31, 2012 6:46 PM CST via mobile reply actions
Most amazing thing to me?
Anthony Randolph has 2 assists…. ALL SEASON.
I know that’s not his role, but come on… 2 assists and 73 shots… 1 assist every 36.5 shots..
Kevin Love, on the other hand has 33 assists and 378 shots… 1 assist every 11.5 shots..
I only compare the two because they’re both PF’s…
I also bring this up because I can not possibly be the only person that cringes whenever Randolph gets the ball when he’s got a defender on him, because he’s going to shoot it anyway.
Randolph is a great athlete with talent, but boy is he a big ball hog.
None shall pass!
"Oh my, oh my, oh my. Is that great basketball or what?" --Hubie Brown, Jan. 20, 2012
by Cynical Jason on Jan 31, 2012 7:00 PM CST up reply actions 1 recs
What are you going to do? Bleed on Me?!?
If at first you don't succeed, try, try again.
If that doesn't work, cheat.
by TheEvilProfessor on Feb 1, 2012 7:19 AM CST up reply actions
You've got no arms left!
(Heh. I was trying for Gandalf. I must be hardwired for Monty Python.)
"Oh my, oh my, oh my. Is that great basketball or what?" --Hubie Brown, Jan. 20, 2012
by Cynical Jason on Feb 1, 2012 7:32 AM CST up reply actions
I made a counting joke about 1, 2, 5 yesterday and no one got it
really depressing
I make jokes like that in the classroom all the time
and they always fall flat. Even stuff like, “Anyone? Anyone? Bueller?” These damned kids are too young.
"Oh my, oh my, oh my. Is that great basketball or what?" --Hubie Brown, Jan. 20, 2012
by Cynical Jason on Feb 1, 2012 7:49 AM CST up reply actions
the author's use of the word PRISON ...
gets no love from the kids…
Too hot to handle, too cold to hold
They're called the Ghostbusters and they're in control
If you want it, you can bleed on me
We all need someone.
I've been beaten down by this club for so long that it looks like up to me.
Good question...
I never even thought to put his numbers up…
Michael Beasley has 13 Assists and has taken 139 shots, good for 1 assist every 10.7 shots.
Last year, Beasley had 158 assists and 1246 shots: 1 assist every 7.8 shots.
Also, after looking at Randolph’s stats from last year he was much better, around 1 assist for every 10 shots. I would like to see what he can do if he passes the ball more, but recently it seems to be predetermined that if he gets the ball, he’s going to shoot it. And that really doesn’t help the team as much as he could, IMO
by mak07 on Jan 31, 2012 10:38 PM CST up reply actions 1 recs
Dammit, you made me go to Basketball-reference...
…to look up Al Jefferson.
Last year, 149 assists on 1319 shots = 8.85 blackhole quotient – which is actually well below his career average of 10.34.
I like this Blackhole Quotient idea
Just running this year’s stats through a google spread sheet and a quick formula function and we find that LeBron has 141 assists on 383 fga, for a 2.716 BQ. If we divide by his minutes and then multiply by 48 we get a .175 BQ48. Love has a 11.45/.664 rate.
The biggest black holes in the league by this raw measurement are, as follows:
1- Tristan Thompson, 59
2- Enes Kanter, 25.3
3- Chris Andersen, 25
4- Jason Smith, 23.16
5- DeAndre Jordan, 23
They’re pretty much all bigs near the top. 3 point specialists are up there too, with Anthony Morrow (14.41) and Daequan Cook (21.5) being in the top 20.
Who takes the cake for the highest MN BQ? Would you believe it is a big who shoots a lot of 3s? Anthony Tolliver, come on down! (13.8)
On a per 48 basis, it goes as follows:
1- Tristan Thompson- 9.04
2- Chris Andersen, 5.08
3- Enes Kanter, 4.43
4- Daequan Cook, 3.53
5- Jason Smith, 3.15
6- Manther, 2.49
The NBA average is 6.668 and .736/48. I only used players that qualify for mp/game on BR.
Ricky is 1.09/.072.
He’s a magical unicorn.
by Stop-n-Pop on Feb 1, 2012 11:48 AM CST up reply actions 2 recs
What's the opposite of black hole?
A super nova? I’m no astronomer, but it sounds good to me. Ricky, the super nova.
I've been beaten down by this club for so long that it looks like up to me.
That actually works perfect with the analogy
Though if we start calling him “White Hole” I think some people are going to think we’re being racist. And gross.
"Pinch-bunters don't have a ton of value, even with the Twins"
by Steven Ellingson on Feb 1, 2012 3:35 PM CST up reply actions
Good stuff right there...
So so far this season, Anthony Randolph is the 2nd biggest black hole in the NBA? Behind only Tristan Thompson? That’s amazing…
Just think how good he’d be if he learned to pass the ball even at a Beasley-level… Anthony Randolph has 20-10 potential, but he’ll never be given the freedom to reach it if he doesn’t learn to pass. I compare him a little bit to LMA, the big difference is LMA plays within the system, and he’s become a borderline franchise player.
Anthony Randolph doesn’t play within the system, and he’s on his 3rd team (and he’s only 22)…
Yup - If Randolph gets the ball you know it is going up
No matter how far away from the basket he is.
"It's tough to make predictions, especially about the future." -- Yogi Berra
by Wile E Coyote on Feb 1, 2012 12:51 PM CST up reply actions
I think a more accurate measure
Would be FGA/touches, but I doubt that data is available.
Some guys are not skilled passers, or are not in an ideal ‘assist’ zone on the floor. So they may rightly pass the ball after touching it but don’t get an assist.
The real black holes are the sticky finger guys that, upon touching the ball, are absolutely going to either shoot it or turn it over.
"It's tough to make predictions, especially about the future." -- Yogi Berra
by Wile E Coyote on Feb 1, 2012 12:46 PM CST up reply actions
For some centers, you might argue a high BQ with low FGA is appropriate
As a coach, do you really want DeAndre doing a lot of passing and ball handling? I’d think the plan is (1) Guards – don’t throw DeAndre the ball unless he can dunk and (2) DeAndre – if you catch the ball and can dunk please do.
"It's tough to make predictions, especially about the future." -- Yogi Berra
by Wile E Coyote on Feb 1, 2012 12:48 PM CST up reply actions
There are data services that track that sort of thing...
….but the closest we could get for free on the intertronz is to calculate team possessions:
0.5 * ((Tm FGA + 0.4 * Tm FTA – 1.07 * (Tm ORB / (Tm ORB + Opp DRB)) * (Tm FGA – Tm FG) + Tm TOV) + (Opp FGA + 0.4 * Opp FTA – 1.07 * (Opp ORB / (Opp ORB + Tm DRB)) * (Opp FGA – Opp FG) + Opp TOV)).
And then take the usage rate of each player to make a per/poss statistic. You’d run into issues with that, as well.
by Stop-n-Pop on Feb 1, 2012 12:58 PM CST up reply actions 1 recs
My favourite point guard "mentality" chart

Top left: Pass-first-playmakers who create plays almost exclusively for others
Bottom left: Passive low-usage PGs who play alongside high-usage players (mainly SGs and SFs)
Top right: Guys who create for themselves & others at a high rate
Bottom right: Black holes
by in-the-game on Feb 1, 2012 3:52 PM CST up reply actions 3 recs
Westbrook?
If I'm pissing you off its probably sarcasm
by CoffeeJanitor on Feb 1, 2012 4:42 PM CST up reply actions
Maybe Love's "assists received" is lower
due to how indiscriminate Ricky is with his distribution. But for a guy whose fourth (FOURTH!) in the league in the scoring, AND a “big,” it’s amazing the amount of his scoring that is unassisted.
A big “F.U.” to the “he can’t create his own shot” crowd, I think.
by dontbesomean youngfella on Feb 1, 2012 1:38 PM CST reply actions
I think the bipolarity of the skewness of the range of latitude
is too coagulated to coalesce any real magnitude of dynamically static signifigance














