Great article breaking down Drummond and Oriahki efficiency | The Boneyard

Great article breaking down Drummond and Oriahki efficiency

Status
Not open for further replies.
Joined
Aug 26, 2011
Messages
6,135
Reaction Score
20,046
If you are one of the contributors, please keep it up! Those were a couple of interesting reads
 
Joined
Aug 26, 2011
Messages
6,018
Reaction Score
18,801
Interesting stuff, but indeed we'll eventually need a larger sample size before we can draw too many conclusions.
 

huskyharry

Hooyah
Joined
Aug 27, 2011
Messages
3,510
Reaction Score
3,904
Thank you for the link! Confirms that the team struggles when both AD & AO are both in together and JC has also limited their minutes together. Fortunately, depending upon the matchup, we have 3 really good players at the five (AO, AD & TO) and five really good players at the four (AO, AD, TO, RS and DD, DD against a smaller, quicker club). IMHO, I disagree with those who say that AO is more of a natural 4 at the college level. The strong parts of his game are strength & timing for shot-blocking. His weak points are his hands (ability to catch the ball in traffic) and ability to make an offensive move when away from the low block and passing. AD is very strong, jumps very quickly (and high) has very quick and strong hands and has a good instinct for passing but he struggles with fundamentals of positioning and how to utilize his strength without fouling (going over the back, reaching in, bringing the ball down in traffic). These deficiencies make both players more effective at center right now...definitely not a bad thing to have two really strong college centers, one of whom will likely show meteoric improvement as the year goes on (AD)
 
C

Chief00

Quite frankly, not a very impressive article. Let's wait until we play teams with real big men before we draw any conclusions. These 3 ball games are not for big men.
 

mets1090

Probably returning some video tapes...
Joined
Aug 25, 2011
Messages
3,617
Reaction Score
3,150
That is very cool - way better than my line graphs! I definitely want to see what that looks like after 10 games.

I wasn't going to bother following the link but if it can top your line graphs...
 

caw

Joined
Aug 26, 2011
Messages
6,973
Reaction Score
12,239
I appreciate the work.

However, something seems off, has UConn really only won 4 games by a combined 13 points?

Perhaps I am just not understanding the chart properly, but it says as a total in 160 minutes UConn has given up 27+28+105+113 points = 273 total points, but I see they only gave up 253 this season (57+66+60+70).

Secondly, it says UConn scored 7+54+108+117 = 286, but I see they scored 372 (70+57+78+80).

Like I said, I appreciate the effort, but something is off and by a substantial margin, or I am just too tired from work...
 
Joined
Aug 31, 2011
Messages
1,486
Reaction Score
6,680
Nicely done caw, I missed that the first time I looked but you're absolutely right. The numbers are way off.

I followed the play by play links in the blog and it appears the source data on substitutions might be the problem - guys are noted as coming in without ever having been noted as coming out, and vice versa.

In the Maine game alone I'm pretty sure the team scored well over 7 points with Oriakhi and Drummond on the floor together.

Still, have to applaud the effort. If this guy can get the bugs worked out I'll be a regular reader of the blog. I've always wondered why no UConn beat guy has tried to distinguish himself by doing this type of statistical analysis.
 

ctchamps

We are UConn!! 4>1 But 5>>>>1 is even better!
Joined
Aug 24, 2011
Messages
16,917
Reaction Score
41,377
I appreciate the work.

However, something seems off, has UConn really only won 4 games by a combined 13 points?

Perhaps I am just not understanding the chart properly, but it says as a total in 160 minutes UConn has given up 27+28+105+113 points = 273 total points, but I see they only gave up 253 this season (57+66+60+70).

Secondly, it says UConn scored 7+54+108+117 = 286, but I see they scored 372 (70+57+78+80).

Like I said, I appreciate the effort, but something is off and by a substantial margin, or I am just too tired from work...
Good catch to a good article. I think those were points when Uconn played 3 against five.
 
Joined
Sep 21, 2011
Messages
79
Reaction Score
54
Hey guys, this is my site. I'm still working on getting a few of the kinks out. Good call on the source data, that's been a problem in the past. I figured out a couple of the problems with this (one of my halftime scores was wrong in the Maine game), but I'll get it figured out. I was planning on doing a more full rollout of the site after the Atlantis tournament (because I'll have some extra time over the holidays to iron everything out).

Thanks for the kind words!
 
Joined
Aug 31, 2011
Messages
1,486
Reaction Score
6,680
Hey guys, this is my site. I'm still working on getting a few of the kinks out. Good call on the source data, that's been a problem in the past. I figured out a couple of the problems with this (one of my halftime scores was wrong in the Maine game), but I'll get it figured out. I was planning on doing a more full rollout of the site after the Atlantis tournament (because I'll have some extra time over the holidays to iron everything out).

Thanks for the kind words!

Good on ya. A UConn-focused Luke Winn will be a great thing to have.
 
Joined
Sep 21, 2011
Messages
79
Reaction Score
54
Fixed. (Though it's not much of an article anymore haha)

caw, thanks for the catch. Sorry, everyone, that I had that link posted without making sure everything was correct. This whole thing is going to be somewhat of an experiment this year, but I think it will be fun and there should be some good stuff. Anyway, thanks everyone for taking a look, and, again, sorry I ended up posting something messed up like that. I can hear all my old math teachers telling me to check my work in my head. :oops:
 

jleves

Awesomeness
Joined
Aug 27, 2011
Messages
4,253
Reaction Score
15,065
If I read it correctly now, we should be subbing on the fly with AD and AO playing only defense and getting off the court for O.
 

caw

Joined
Aug 26, 2011
Messages
6,973
Reaction Score
12,239
Fixed. (Though it's not much of an article anymore haha)

caw, thanks for the catch. Sorry, everyone, that I had that link posted without making sure everything was correct. This whole thing is going to be somewhat of an experiment this year, but I think it will be fun and there should be some good stuff. Anyway, thanks everyone for taking a look, and, again, sorry I ended up posting something messed up like that. I can hear all my old math teachers telling me to check my work in my head. :oops:

Glad to help. Love the idea and will read it again.
 
Joined
Sep 21, 2011
Messages
79
Reaction Score
54
If I read it correctly now, we should be subbing on the fly with AD and AO playing only defense and getting off the court for O.

You're right. Based on those numbers, in a totally optimal world you'd sub both Drummond and Oriakhi in on defense and then take them out for Roscoe and Olander on offense. That would be assuming that the trends from the first four games are going to continue indefinitely (like if you were setting your lineup in a video game where players abilities were completely predetermined and unable to grow or change throughout the season).

I think the bigger point is that they need to adjust to playing together on offense and running the offense (everything away from the ball) together. I wouldn't use those numbers to say they can't coexist, but, rather, that playing together is something they're both adjusting to.
 

jleves

Awesomeness
Joined
Aug 27, 2011
Messages
4,253
Reaction Score
15,065
You're right. Based on those numbers, in a totally optimal world you'd sub both Drummond and Oriakhi in on defense and then take them out for Roscoe and Olander on offense. That would be assuming that the trends from the first four games are going to continue indefinitely (like if you were setting your lineup in a video game where players abilities were completely predetermined and unable to grow or change throughout the season).

I think the bigger point is that they need to adjust to playing together on offense and running the offense (everything away from the ball) together. I wouldn't use those numbers to say they can't coexist, but, rather, that playing together is something they're both adjusting to.
I was just trying to be funny - apparently I missed. If you want any help with such a project, I'd be willing to do stuff. I love stats and looking at things in different ways and presenting them in interesting views that give more significance than a standard box score.
 
Joined
Sep 21, 2011
Messages
79
Reaction Score
54
I was just trying to be funny - apparently I missed. If you want any help with such a project, I'd be willing to do stuff. I love stats and looking at things in different ways and presenting them in interesting views that give more significance than a standard box score.

The miss was on my end Hahaha. I will definitely keep you in the loop. I like those graphs you've been putting together
 
Joined
Aug 27, 2011
Messages
1,547
Reaction Score
693
I really would love to see the +/- with TO in the game. I think he's been playing fantastically.
 
Joined
Nov 7, 2011
Messages
906
Reaction Score
650
To the person who compiled the data and sorted it - well done. I find it interesting.

Regarding sample size - I have no issue with the n number. It reflects, with a fairly low statistical significance level, the interactions of AO and AD on the court.

I note that comparing the +/- is really difficult, as they, as far as I can tell, are not normalized.
In this case, you'd want to know +/- per minute. Or per game.

If you do that math, here's what you get:

AD/AO 40' Diff
Together - +27
Just AD - +4
Just AO - +14
Neither - +39

Sure, it's only 4 games, but that last number is just eye popping.
The numbers suggest that we're better with both of them out there or both of them on the bench. Weird.

If somebody makes the point - "yeah, but the other teams played small . . . ," my response would be, "so why didn't AO and AD take better advantage of that on the offensive end?"

It will fascinating to see these numbers in another 5 or so games.
 

jleves

Awesomeness
Joined
Aug 27, 2011
Messages
4,253
Reaction Score
15,065
The miss was on my end Hahaha. I will definitely keep you in the loop. I like those graphs you've been putting together
like your data, they don't mean much right now and they need some work in terms of format, but like you, I'll keep working on it. I'm trying to post something now, but the upload image size makes it unreadable.
 
Joined
Aug 26, 2011
Messages
9,336
Reaction Score
23,496
To the person who compiled the data and sorted it - well done. I find it interesting.

Regarding sample size - I have no issue with the n number. It reflects, with a fairly low statistical significance level, the interactions of AO and AD on the court.

I note that comparing the +/- is really difficult, as they, as far as I can tell, are not normalized.
In this case, you'd want to know +/- per minute. Or per game.

If you do that math, here's what you get:

AD/AO 40' Diff
Together - +27
Just AD - +4
Just AO - +14
Neither - +39

Sure, it's only 4 games, but that last number is just eye popping.
The numbers suggest that we're better with both of them out there or both of them on the bench. Weird.

If somebody makes the point - "yeah, but the other teams played small . . . ," my response would be, "so why didn't AO and AD take better advantage of that on the offensive end?"

It will fascinating to see these numbers in another 5 or so games.

Thanks for posting the numbers I was just going to. Posters in this thread have mentioned the struggles of AD and AO on the court at the same time, but really the numbers do not back that up. While the sample size is very small, a combintion of AD and AO out there at once has made up our best defensive lineup by a long shot, while a combination containing neither AD and AO has been by far our best offensive unit. I continue to be interested by Calhoun's resistence to play both at the same time. He seems to think it would be better for their development for them to not play together, but I would assume they are going to be spending a lot of time on the court together when the tougher part of the schedule rolls around. You can't really say AO and AD have failed to co-exist because they really haven't had the chance to. Best case scenario, this AO/AD dynamic will play itself out like LeBron and D-Wade last year, where they struggled to play together early in the season only to turn into a very dangerous tandem later on.
 
Joined
Aug 26, 2011
Messages
6,369
Reaction Score
17,384
Regardless of stats, just by watching the games, it seems like the offense flows better with one on the floor and Olander at the 4. Obviously this will change once AO and Drummond learn to play together but right now i agree with Calhoun's thinking that its very hard to keep a guy like Olander off the floor.
 

ctchamps

We are UConn!! 4>1 But 5>>>>1 is even better!
Joined
Aug 24, 2011
Messages
16,917
Reaction Score
41,377
Hey guys, this is my site. I'm still working on getting a few of the kinks out. Good call on the source data, that's been a problem in the past. I figured out a couple of the problems with this (one of my halftime scores was wrong in the Maine game), but I'll get it figured out. I was planning on doing a more full rollout of the site after the Atlantis tournament (because I'll have some extra time over the holidays to iron everything out).

Thanks for the kind words!
Great site. Thanks for taking the time to put it together. I find a lot of recent threads, including this one, a welcome change in discussions regarding the men's team. It's a lot preferable imo, to the attack mode we often observe. Keep up the good work.
 
Status
Not open for further replies.

Online statistics

Members online
388
Guests online
3,019
Total visitors
3,407

Forum statistics

Threads
155,758
Messages
4,030,550
Members
9,864
Latest member
leepaul


Top Bottom