Page 1 of 7

To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 4:12 am
by Ed Wood
This was going to be a multi part bunch of nonsense in the Fix the Team thread but apparently that thread doesn't want it so now it gets its own thread. The question on the floor is how much faith you have in advanced statistical analysis and to what extent you want it to be a part of how the Wizards operate after the presumed upcoming revamp of the organization. There will ideally be posts in the thread by myself and better posters who do real math to illustrate some of the precepts established by statistical analysis of basketball and the potential path down which that approach may lead.

For myself, I will shortly post far too much on the subject so my first foray will be brief: I think the team should incorporate statistical analysis to an extent that will frustrate me. There are aspects of what has come out of mathketball that I do not entirely enjoy (such as the value of players such as Danny Fortson) and that I might be tempted to ignore if I had any power to do so. I think that stuff should be shoved down my throat because I'm being stupid and I need to be saved from myself.

And now, you.

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 4:14 am
by Ed Wood
Here is the other post:

Ed Wood wrote: A Postmorten Basketballosis on the 04/05 Seattle Supersonics
or The Legend of Kevin Pittsnoggle


Recently (yesterday I think, the days do at times blend together) the Sports Reporters radio program ran a segment posing the question “what mid major success over the past decade, George Mason, VCU or Butler, is the best story?” Not a bad question, I’d probably say Butler because the argument for George Mason is dumb (they did it first, they showed everybody it was possible, like Butler wouldn’t have known it could win games otherwise) and the argument for VCU has some merit, having had the hardest and least likely path to the final four in a single season but that single season can’t trump comparable success in two consecutive seasons.

On the other hand it wasn’t really a question that interested me too terribly. Yes, I’m sure media types are enjoying Mid Major Mania but while I like seeing small schools win I’d rather talk about why they win than about how scrappy they are while they do. And that I did think about a little, and when I did it occurred to me that George Mason wasn’t really the first mid major to make a big splash in the tournament. Yeah there was Gonzaga and any number of teams from way back but never mind them, even this decade Mason wasn’t the first. The first was West Virginia. Okay, sure, West Virginia is a major conference team, but every other ingredient was there. That team had:

    A zillion white players, like really white. Like five different goofy white guy names, and the haircuts.

    So many threes, I jump in it.

    Not a soul on the team with any chance to be an NBA player, not even close.

    A coach who hates the media, hates dealing with the AAU generation of players, hates his mother.

    A system that was a lethal combination of confusing, effective and stupefying boring.


It’s all there, West Virginia was the paragon of a mid major success story, John Beilein just happened to do it at an ostensibly big conference school. But seriously **** all of that, if I wanted to talk about scrappy teams I’d be making a long rambling post about race code words in sports and that’s not what this is. This is about how a bunch of chucker and stiffs in the middle of nowhere managed to be a damn good basketball team, and how they did it with efficiency and specialization, and how most of the qualities that made that West Virginia team successful have been present in the Cinderellas since, particularly in Butler.

And that’s no accident; Butler has been built by Brad Stevens to take advantage of what you might call market inefficiencies in college basketball, if that term is appropriate. Stevens knows he isn’t going to have more talent than many of the teams Butler will have to beat in the tournament every year so he doesn’t win with talent. He wins by attempting to understand why winning happens at all, based upon the sensible notion that the better understood a process is the more likely it is to be successfully replicated.

Butler has it’s own Kevin Pittsnoggle, sort of, in Matt Howard. Five players on the team have taken over 80 threes on the year, and those players who play significant minutes and don’t take threes basically don’t take jumpers (Andrew Smith, Khyle Marshall) or don’t shoot at all (Ronald Nored). Everybody who does shoot basically shoots threes or takes shots right at the rim, with very little in-between and guys who do the later draw plenty of fouls (Both Howard and Smith shoot fewer than two field goals per free throw attempted).

The point of all this is that Butler succeeds because the entire program is designed around a very statistically efficient style of play, a style that was born in a computer as much as it was in a gym, and the precepts that have made Butler and West Virginia before it successful programs despite limited resources. It just so happens that the same ideas were being applied at the same time in the NBA by the Seattle Supersonics, with similarly surprising success.

Part two of this rambling piece of crap will cover the 2005 Sonics team and attempt to perform a postmortem on the team in order to understand the methods that led to its success, and look to the current NBA for franchises operating using the same ideas.

Part three will attempt to put together a plan based around these ideas for the Wizards going forward. You are a huge loser for reading all this. Neeeeeeeeeeeerrrrrrrrrrrrrrrrrrddddddddddddddd.


Look at this guy, quoting himself, how uncool.

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 11:20 am
by pancakes3
bloody brilliant. and to answer your question? enormously. the wiz should be tripping over itself to have a stat department. i'm not sure they have one though. back around 2005 i sent HR an email asking if they would direct me to their stat intern dept directed me to their general intern page where they had marketing and operations internships. i guess they could just not offer the stat internship but... i dunno. just my thin little slice of life experience.

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 11:21 am
by verbal8
Very entertaining. I think an improving and winning team will make it easy to overlook the stigma of the stat nerds methods it will probably take to get there.

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 12:04 pm
by nate33
I recall Ted Leonisis being asked about advanced statistics and he said he would definitely use them. Of course, Flip's offense is predicated on shooting lots of long 2-pointers so there's reason for concern.

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 12:26 pm
by doclinkin
I like MIke Gansey. Also turtles.

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 1:07 pm
by dobrojim
it would only make sense to do so, so I would say to a considerable extent.

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 1:21 pm
by Zonkerbl
Well, to take advantage of the three point shot you have to have people on your team who can make them.

But yeah. Three point shooting is undervalued, as is defense. The three pointer because basketball coaches are idiots who can't do math -- shooting 40% from the three is like shooting 60% from the field. 60%!!!!! Are you kidding me? So someone who can shoot 40% from the three is the statistical equivalent of Shaquille O'Neal (well, not really, because they don't draw fouls and get the opposition into foul trouble. But yeah.) And defense is undervalued because it doesn't sell tickets or shoes. So the most valuable players in the nba, mathbasketballwise, are good three point shooters who can play defense.

And it just so happens that we have a player like that named Nick Young. He shoots 38.8% from three and plays good man to man defense. Aren't we lucky?

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 1:53 pm
by nate33
Zonkerbl wrote:The three pointer because basketball coaches are idiots who can't do math -- shooting 40% from the three is like shooting 60% from the field. 60%!!!!! Are you kidding me?

Sounds like an argument for drafting Jon Diebler from Ohio State. He shoots 50% from three. That's equal to 75%!

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 2:25 pm
by Zonkerbl
meh, that's the college three though.

But you know what I mean, when evaluating NBA talent, if you place relatively more weight on three point shooting and defense you will get more NBA talent per dollar than other teams. Doesn't necessarily mean you should draft Shawn Respert #6.

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 2:27 pm
by GhostsOfGil
mike lee just tweeted
"John Wall needs 103 assists in last 9 games to become 4th rookie in NBA history to average 9 per game. Is it possible?"

thats about 11.4 per game...

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 2:34 pm
by nate33
Zonkerbl wrote:meh, that's the college three though.

But you know what I mean, when evaluating NBA talent, if you place relatively more weight on three point shooting and defense you will get more NBA talent per dollar than other teams. Doesn't necessarily mean you should draft Shawn Respert #6.

Sorry my post came out as condescending. I didn't mean it to be. What I'm saying is that maybe the Wizards should honestly consider drafting a guy like Diebler in the 2nd round. He unquestionably has pro 3-point range (though presumably not at 50%). He is one of the best 3-point shooters in the history of college basketball, and he's 6-7 so he can hopefully get away with guarding other "3&D" SF's (where his defensive responsibilities would be more limited). If he can shoot, say, 44% from 3-point range, that's like having a guy shooting 66% from 2.

He's a terrible defender though. He could only play in certain matchups - like if we're going up against somebody like Shane Battier or Aaron Afflalo who basically stand at the 3-point line and don't do anything.

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 2:37 pm
by nate33
zaRdsAndZeRos wrote:mike lee just tweeted
"John Wall needs 103 assists in last 9 games to become 4th rookie in NBA history to average 9 per game. Is it possible?"

thats about 11.4 per game...

Not unless Young comes back. Wall just can't rack up assists with this team of woeful shooters. At the very least, we need to see a little more of Cartier Martin and a little less of Jordan Crawford.

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 2:51 pm
by Zonkerbl
nate33 wrote:
Zonkerbl wrote:meh, that's the college three though.

But you know what I mean, when evaluating NBA talent, if you place relatively more weight on three point shooting and defense you will get more NBA talent per dollar than other teams. Doesn't necessarily mean you should draft Shawn Respert #6.

Sorry my post came out as condescending. I didn't mean it to be. What I'm saying is that maybe the Wizards should honestly consider drafting a guy like Diebler in the 2nd round. He unquestionably has pro 3-point range (though presumably not at 50%). He is one of the best 3-point shooters in the history of college basketball, and he's 6-7 so he can hopefully get away with guarding other "3&D" SF's (where his defensive responsibilities would be more limited). If he can shoot, say, 44% from 3-point range, that's like having a guy shooting 66% from 2.

He's a terrible defender though. He could only play in certain matchups - like if we're going up against somebody like Shane Battier or Aaron Afflalo who basically stand at the 3-point line and don't do anything.


Oh, I thought you were kidding. Yeah, I'm done drafting project bigs with second round picks. Why not take a flyer on a three point shooter instead?

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 3:09 pm
by tontoz
Zonkerbl wrote:
nate33 wrote:
Zonkerbl wrote:meh, that's the college three though.

But you know what I mean, when evaluating NBA talent, if you place relatively more weight on three point shooting and defense you will get more NBA talent per dollar than other teams. Doesn't necessarily mean you should draft Shawn Respert #6.

Sorry my post came out as condescending. I didn't mean it to be. What I'm saying is that maybe the Wizards should honestly consider drafting a guy like Diebler in the 2nd round. He unquestionably has pro 3-point range (though presumably not at 50%). He is one of the best 3-point shooters in the history of college basketball, and he's 6-7 so he can hopefully get away with guarding other "3&D" SF's (where his defensive responsibilities would be more limited). If he can shoot, say, 44% from 3-point range, that's like having a guy shooting 66% from 2.

He's a terrible defender though. He could only play in certain matchups - like if we're going up against somebody like Shane Battier or Aaron Afflalo who basically stand at the 3-point line and don't do anything.




I Oh, I thought you were kidding. Yeah, I'm done drafting project bigs with second round picks. Why not take a flyer on a three point shooter instead?



I also think shooting should be the major focus of this draft. There are already enough players on this team who cant shoot.

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 3:21 pm
by dobrojim
How heavily should we weight DWilliams shooting 60% in college?

low sample size but not that low.

But I'm guessing we take him if given the opportunity
unless something dramatic happens btw now and draft day.

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 3:45 pm
by GhostsOfGil
tontoz, add a healthy shart and a resigned ny thats some pretty good spacing. from the 2 and 3 spots, thats a rough career average of 39% and 40% from 3, respectively. not to many teams have that kinda shooting in their starting lineups. if ny is resigned i think we need to focus on getting a reliable presence in the post rather than another shooter.

infact only 3 other teams have a tandem that rivals that kind of long distance shooting

denver: affalo 40% and gallinari 37%
orlando: jrich 37% and turky 38%
and sac town: garcia 37% and thornton 38%

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 3:59 pm
by tontoz
zaRdsAndZeRos wrote:tontoz, add a healthy shart and a resigned ny thats some pretty good spacing. from the 2 and 3 spots, thats a rough career average of 39% and 40% from 3, respectively. not to many teams have that kinda shooting in their starting lineups. if ny is resigned i think we need to focus on getting a reliable presence in the post rather than another shooter.



Shard is just a rental. I think resigning Nick is a must. Some people here seem content to let him walk which i find hard to believe.

If Kanter is available i would certainly take him. But with the Atlanta pick and the 2nd rounder i would be looking at shooters. I doubt there will be many quality post players late in a weak draft.

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 4:01 pm
by nate33
dobrojim wrote:How heavily should we weight DWilliams shooting 60% in college?

low sample size but not that low.

But I'm guessing we take him if given the opportunity
unless something dramatic happens btw now and draft day.

Yup. I think he's number 1 on our draft board. This team is in desperate need of a skilled big man so Blatche can be properly disciplined with ass-splinters when necessary. Right now, Flip just can't take Blatche out of the game without our offense sputtering.

Re: To What Extent Should Statistics be a Part of Your Wizards?

Posted: Wed Mar 30, 2011 4:07 pm
by GhostsOfGil
tontoz wrote:
zaRdsAndZeRos wrote:tontoz, add a healthy shart and a resigned ny thats some pretty good spacing. from the 2 and 3 spots, thats a rough career average of 39% and 40% from 3, respectively. not to many teams have that kinda shooting in their starting lineups. if ny is resigned i think we need to focus on getting a reliable presence in the post rather than another shooter.



Shard is just a rental. I think resigning Nick is a must. Some people here seem content to let him walk which i find hard to believe.

If Kanter is available i would certainly take him. But with the Atlanta pick and the 2nd rounder i would be looking at shooters. I doubt there will be many quality post players late in a weak draft.


idk about shard being a rental. if i had to guess, shard will be with the team until he can be moved as an expiring. im not sure if his trade value will be any higher than as a max expiring contact. plus he helps space the floor and provides a veteran presence. lets not forget how much better we played while he was healthy too. im not advocating him staying, im just sayin