ACC Players - Who should shoot more and who shouldn't?

[tbyers11]

Here is a link to a Dear Old UVa (http://www.dearolduva.com/basketball/acc-coaches-on-offense-who-s-good-who-isn-t-and-one-way-to-tell/) blog that uses 2 statistical measurements to determine whether ACC players are good or bad offensively. They plot Offensive Rating vs % Possessions Used. All data is from Ken Pom and here is a link to a brief description (http://kenpom.com/blog/index.php/weblog/stats_explained/) of his stats. Note: This data is from the entire season not just ACC play

http://www.dearolduva.com/files/media/image/WindowsLiveWriter/ACCCoachesonoffenseWhosgoodwhoisntandone_8436/acc_players_thumb_2.png

They set arbitrary but reasonable cutoffs for the axes at an ORating of 105 and %Poss at 20. Players in the High Usage Warranted quad are good offensively and have a lot of the offense run through them. Overused players use a lot of their teams possessions but aren't very efficient at it. Basically, you want to have your team's players fall either in the NE quad or the SW quad.

They undertook this analysis to see if their post player, Mike Scott was underused and make additional points about certain players that are labeled on the graph.

As for Duke, here is Ken Pom's data (http://kenpom.com/team.php?y=2009&team=Duke) on the Devils. Singler is the blue diamond that is almost obscured by the purple dot that is Tony Douglas. Gerald is used a little less but has a slighter higher ORating (right below the red triangle for Malcolm Delaney). Jon has an ORating of almost 120 and is bordering on being underused with just over 20% Usage.

Nolan is the only Devil in the overused category and barely so. Zoubek (and Greg and Lance) appear underused. While this isn't a perfect analysis, it reveals a lot about certain players and bears out certain trends about teams. For example, UNC is a very good offensive team because their good players use the ball on offense and their bad ones don't. GTech is not very good on offensive because they have a lot of high usage players with bad offensive efficiency.

I really like this stuff and thought it would be a good way for everyone to waste a few minutes on basketball this afternoon. Hat tip to Rush the Court (http://rushthecourt.net/)for bringing this to my attention.

[COYS]

Here is a link to a Dear Old UVa (http://www.dearolduva.com/basketball/acc-coaches-on-offense-who-s-good-who-isn-t-and-one-way-to-tell/) blog that uses 2 statistical measurements to determine whether ACC players are good or bad offensively. They plot Offensive Rating vs % Possessions Used. All data is from Ken Pom and here is a link to a brief description (http://kenpom.com/blog/index.php/weblog/stats_explained/) of his stats. Note: This data is from the entire season not just ACC play

http://www.dearolduva.com/files/media/image/WindowsLiveWriter/ACCCoachesonoffenseWhosgoodwhoisntandone_8436/acc_players_thumb_2.png

They set arbitrary but reasonable cutoffs for the axes at an ORating of 105 and %Poss at 20. Players in the High Usage Warranted quad are good offensively and have a lot of the offense run through them. Overused players use a lot of their teams possessions but aren't very efficient at it. Basically, you want to have your team's players fall either in the NE quad or the SW quad.

They undertook this analysis to see if their post player, Mike Scott was underused and make additional points about certain players that are labeled on the graph.

As for Duke, here is Ken Pom's data (http://kenpom.com/team.php?y=2009&team=Duke) on the Devils. Singler is the blue diamond that is almost obscured by the purple dot that is Tony Douglas. Gerald is used a little less but has a slighter higher ORating (right below the red triangle for Malcolm Delaney). Jon has an ORating of almost 120 and is bordering on being underused with just over 20% Usage.

Nolan is the only Devil in the overused category and barely so. Zoubek (and Greg and Lance) appear underused. While this isn't a perfect analysis, it reveals a lot about certain players and bears out certain trends about teams. For example, UNC is a very good offensive team because their good players use the ball on offense and their bad ones don't. GTech is not very good on offensive because they have a lot of high usage players with bad offensive efficiency.

I really like this stuff and thought it would be a good way for everyone to waste a few minutes on basketball this afternoon. Hat tip to Rush the Court (http://rushthecourt.net/)for bringing this to my attention.

I find these stats very interesting, too, and I think it shows why our offense has improved since we moved our most efficient guard (Jon Scheyer) to the point guard position where he handles the ball more frequently. Nolan is simply not an efficient player when he plays point guard. I really hope Nolan recovers from his concussion quickly because he's had a tough season in that he's been asked to take on a roll that just does not emphasize his strengths. I'd love for him to settle into a new roll as a scorer and defender off the bench. Of course, I want him to be completely healthy before he comes back first and foremost.

[Heelo]

Is there an interactive graph where I can see who all of the dots are?


Thanks.

[tbyers11]

Is there an interactive graph where I can see who all of the dots are?


Thanks.

If you go to the Ken Pom's website and click on the team page for any of the teams (Here's the Duke page (http://kenpom.com/team.php?y=2009&team=Duke)) and scroll down to the bottom of the page you will find the players stats. The 6th column ORtg is the y-axis on the graph and the 7th column %Poss is the x-axis on the graph.

That is what I did to figure out who is who. You can get to all the other ACC team pages by clicking on the team name in the schedule part of the Ken Pom team page. So outside of e-mailing the UVa blog guys and asking for their spreadsheet there isn't a truly interactive graph, but with a little clicking around on Ken Pom's site you can figure out who is who pretty quickly.

[loran16]

Problem is that the data is disturbed by the outliers of players who play sporadic minutes. For example, in the games Zoubek plays the worst and looks outmatched, he'll rarely play 10 minutes. In the games he dominates, he might hit 20 or so minutes. So on average, his Offensive rating will be inflated, and his minutes played will be deflated.

The same can be said of Elliot Williams who is just starting to play, and thus had poor stats. I suspect over the last 3 games those stats are much higher.

This sort of thing can probably be used on our team for Scheyer, Singler, and Henderson. They each play similarish minutes to the point where this isn't a problem. The data there suggests that Scheyer should be used more and so should Hendo, and that Singler should be #3 in terms of "possessions used". But given that scheyer is now playing point, i think this will certainly be evened out and more or less resolved.