Turn Your Z Around: The Ways of Wade
First there was the Z, then 2.0. A scientist cooked up the team version.
Now we bring you a re-imagining:
Soon, we will release the statistical baselines for skill qualification. Until then, an example:
Labels: dwyane wade, graphs, positional revolution, stats
posted by Ziller @ 4/13/2009 05:10:00 PM
35 comments
35 Comments:
What's happening to our hood?
I see something happening here and no you can't call me Mr. Jones, you hobnobbing sons of bitches.
Positional purity rests calmly on the shoulders of blind faith in a Canadian peach basket. There's something so Jung-like (that is to say scatological) about this all.
Don't stop the cock-rock emotif Ziller!
ceral
Hold on, I've gotta change my pants.
I always knew Dwayne Wade was secretly a center.
This kind of looks like the schemata of a sideways uterus, with the c and the pg as Fallopian tubes about to curl into ovaries.
Something something about a new birth?
I don't know what this means that I thought this.
I think that the guards deserve a skill about driving.
i've been updating my blackboard with z graphs in response to people who conspicuously scribble complicated math on theirs. i'm so excited about all this new blackboard dressing.
i'm sure you have this in mind but i hope you make dd rate-based, going against the letter but with the spirit of the category. also, while i do support quantifying every category i anticipate ignoring it in the case of handle.
jokingly i drew some phase diagram arrows/paths on an old z graph but it is slightly fitting now since this is actually in 2-d space (even if we are not dealing with individual points). the idea of stable and unstable distributions implied does seem reactionary, but maybe some distributions just aren't maintainable throughout a career.
joseph,
i thought that triple-double and double-doubles should be rate-based as well, but then i decided that since they're already measuring rebound, scoring, and assist rates, that wouldn't be completely fair, and what the triple-double and double-doubles are measuring is explosions of production. I think that +30 point games would be a good addition in the same spirit.
That's pretty awesome, though I don't think of triple-doubles as an essential skill of small-forwards. Besides LeBron, the guys with 2 or more triple-doubles this year are, as you can see here: Chris Paul, Jason Kidd, Rajon Rondo, Andre Miller, Kobe Bryant.
I would eliminate triple-double as a category in your scheme, as it correlates too strongly with assists and thus is not very useful.
I would say that instead of 30 point games it should be 40, the true measure of a guy really going off. Also i feel like we should something that is man on man defense to the already over crowded sf position (this says something about its re-evolution)
This isn't supposed to suggest that "SF" is all those things, but rather that it can refer to many things, some of which have it all, some which have very little.
It's cool-looking and interesting food for thought, but damn, this chart is even more problematic than the Z.
Handle is a small forward trait? More so than a point guard trait?
Are we all watching the same sport?
I can almost hear the gears grinding in your labors to construct a positional index that somehow tells a narrative story of the flow of basketball skills from one position to another. I think it's possible to do, but I think it's just a whole lot less complicated. How about a friggin' Venn Diagram, guys?
All the same, I really miss the subjective and less quantifiable criteria like "fearlessness", "drives to the hoop".
This version looks sort of like a space-squid, and I await the day when we can construct a pantheon of ghostly, skeletal extraterrestrial cephalopods. That day I will retreat from the world and start a religion, where Anthony Randolph's scattered brilliance guides humanity like a pseudopodal polar star.
I'm pretty much serious here. Awesome graphs, and I like the trend.
I like that you guys are pushing this forward.
Can you explain to me what the x-axis means though? Are you saying the more blue dots you have to the right, the more pure you are? But what about a positionally pure SF? All their dots would be to the left. Does that mean that there's no such thing as a pure SF. What about Paul Pierce?
Maybe I've misinterpreted this all.
in terms of the sf, I think what shoals said in combination with Dave illustrates the point i was trying to make. If we add lock down defense to SF, then we have a position where everything is so un-pure and everything is unrelated thus that the position is not a collection of skills required but possible skills where people exhibit brillaince by excelling in other fields. and paul pierce in himself proves this as the proto small forward is all about athleticism he in fact fills all the roles without being a blueprint.
I don't see any reason why these statistics, which are much more accurate, couldn't be placed into the more narrative and aesthetically pleasing Z-form.
Also the idea of placing it on a graph is marginalized by using purity on the x-axis and height on the y-axis. If you flip-flopped them then you could make it into a function, allowing for tons of mathematical possibilities (imagine the money GMorey would pay for standard deviations from prototypical player stats.)
random suggestion: drop "handle", replace with "freakish athleticism". for maximum entertainment value, quantify it with some sort of algorithm based on the otherwise meaningless numbers from the pre-draft combine.
I don't think this fanbase would be happy until you rolled out some neon-colored Rorschach blots and added fortune-teller speak to signify something hazy, yet prophetic.
Keep pushing that envelope further to the edge, people.
These would be make great cards.
The axes basically mean that the more extreme a player's height, the more their positionality will likely be constrained. That is, a 7'2" center is likely not going to do shit below power forward, and a 5'10" guard is not going to do shit above SG skills. If this doesn't happen, then there are major problems with the system. To me, the lines are most useful for this, but it is strange to place the independent variable on the y-axis. Are you guys economics majors or something?
This is also why the curves are the shape that they are, but I don't understand why they aren't smooth.
I miss the red dots that confused everyone.
Come back confusing red dots! Come back!
You guys will have to figure out a way to combine the pleasing simplicity of the Z with the larger informative ability of the weird uterus/fish mouth.
Two words: boomerang spectrum (octopus uterus).
Guys, the Z was slick and all, but try and explain it to me as an actual graph. When we tried to get at what we thought the "z" showed, it quickly stopped being a "z". And believe you me, we tried. Someday, we'll show you the 3-D sketch I did on TuxPaint that included a axis plotting "likelihood of stat to be linked to stats before and after it." That, with height and purity, was kind of a Z. Sort of. But impossible to comprehend, and that aspect was really arbitrary, even if "determination" vs. "determination of positional purity" sounded cool.
http://www.zshare.net/info.html?58644262-cdde895516c694a80c8822607d94843a
A musical response to this graph
It's irrelevant, but could anyone please create one for Gerald Wallace?
Ok, so reading Marc Stein's ESPN chat today, he said John Wall could possibly be eligible for the draft THIS YEAR. Anybody know anything else about this? Supposedly, depending on your reading of the CBA, 5th year seniors who turn 19 before December 31 can be eligible.
Griffin-Rubio-Wall. Sounds like fun to me.
I couldn't get the file to open. What is it?
Also, yes, if Wall were eligible this year, that might cause a mini-daylight savings time or something.
Stare at the graph long enough and it becomes a ribbon of DNA in the interphase (uh???) phase of meiosis. I want to see this baby.
this is beautiful but do taller guys receive more free throw attempts and shorter players earn more triple doubles?
the zshare is 22 (23) tracks of (somehwat) freedarko inpspired sample-collages made by yours truly. I hope the link works, I'll double check it.
I look forward to wearing a tshirt with this diagram on it one day soon.
the word verification is dershing: how kanye west may have referred to his own fashion sense with his jaw wired shut.
This graph > the Z, but I agree with whoever said to flip the axises; which would lead to limitless functions.
So, would Lamar Odom's and Lebron's graphs be similar, but Odom's would be looked on through "What if?" glasses and Lebron's would be the impending sign of his Apocalypse?
This comment has been removed by the author.
At first glance I thought this graph was just aesthetically pleasing without being particularly descriptive but it really makes sense when you think about it in terms of where skills are on the X axis rather than their proximity to a position floating on the graph. Triple doubles are far to the left not because SF's accumulate them but because they are a sign of having many skills and thus not being positionally pure. Whereas blocks and mgmt are skills relatively isolated to big guys or small guys (except for freaks like Wade, of course).
One thing that disappoints me is that Wade's blocks skill should be the big nonsense bubble. Pretty much any guard that gets over a block a game, it's absolute nonsense and the graph should be able to reflect that. For instance Rondo is a great defensive guard and should have a good defensive +/- but he doesn't rack up blocks. Maybe defensive +/- is too far to the upper right of the graph?
I think that Wade's BK bubble should be HUGE since his 1.4/game are one of the best all time for anyone under 6'6". (if i remember correctly wade has 4/5 of the top block #s for anyone under 6'6")
i know this doesn't work in your flow chart (it isn't 2 skills removed) but maybe the chart or the structure of the charting needs to be redone?
dooflow is onto to something...how cool would it be if basketball cards just had these graphs on the back instead of the regular stats? do basketball cards still exist? Either way, it would be a sick FD-Fleer collabo
I think what i finally understand now is that this graph, intentionally or not, is not about the positional purity of players, but of skills. Thanks Ritchie.
I've been wondering about basketball cards myself recently. Like are thay still around, do people still give a shit about them? Those seriously used to be crack for me when I was like 10-11 years old.
Post a Comment
<< Home