- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
The Dwight Effect is BS
Posted on 10/30/14 at 12:53 pm
Posted on 10/30/14 at 12:53 pm
Boom keeps on posting about the Dwight Effect. It is a study done showing the effect that Dwight has on the game. But Boom makes it seem like it crowns him the best defender in the game. It being named after him doesn't make him an effective defender any more than somebody doing a study they title the "Hasheem Thabeet Effect" looking at the effect of dropping passes on a team's offensive production.
Firstly, a shot that you stopped from being taken is not necessarily as good as a shot blocked or altered. Part of the art of shot blocking is baiting somebody into taking shots that they should know better than to try. That often ends in a change of possession rather than just a different shot. Anyway, to the study......
The author is a bit of a clown in the introduction in that he starts running on about close shots being more Points Per Shot than 3s using data on shots including fast breaks. Obviously the numbers are going to be skewed by fast break uncontested layups. In any event, that doesn't really effect the numbers he used in the study because there he only looks at situations where a defender is within 5 feet of the rim.
Boom points to Dwight having the lowest percentage of shots attempted close range on him at 48.2%. Cousins has almost identical numbers(48.4%). But they also give up an incredibly high 3 point % and a high number of 3 point attempts. Dwight also gives up high mid range % while Cousins is solid here. Looking at the data, it seems that the percentage of shots taken close corresponds as heavily with bad 3 point defense as it does good interior defense.
When Dwight is within 5 feet of the bucket, the opposing team averages .482*.457*2(short range) + .32*.382*2(Mid) + .191*.436*3= .935 points per shot. Finding the Points per Shot of the 15 players who deter the most shots from being taken at the rim(Dwight Effect) plus Ibaka, who deterred the fewest shots of qualifying players, Dwight is 10th out of those 16.
The defenders Dwight is ahead of are Chris Kaman, Kevin Love, Tyler Zeller, Aaron Gray, a KG(I'm sure he would have been better 5 years ago), and Andrei Biedres. Some real stalwart defenders there. Leading the pack BY FAR in lowest Points Per Shot by the opponent are Serge Ibaka(.76 PPS) and Roy Hibbert(.776 PPS). About what you would expect from the 2 most dominated shot blockers(AD didn't qualify for this year and a half old study)
In conclusion:
1) The Dwight Effect exists but it does not correlate with good defense or helping the team unless you can somehow prove it causes turnovers at a higher rate(I would think the opposite-fewer turnovers for teams shooting 3s than going inside In any event, the study certainly does not do that.)
2) The player with the worst "Dwight Effect"(Ibaka) allows the fewest point per shot by the opponent when playing the role of "rim protector"
3) Hibbert and Ibaka(albeit he had a small sample size) are men among boys in anchoring a teams defense.
4) 3 point defense is very important in the NBA
5) Dwight is a better defender than Kevin Love and Chris Kaman
So next time somebody tries to claim Kevin Love is a better defender than Dwight Howard, Boom can reference the "Dwight Effect" and make them look stupid!
Study Here
Firstly, a shot that you stopped from being taken is not necessarily as good as a shot blocked or altered. Part of the art of shot blocking is baiting somebody into taking shots that they should know better than to try. That often ends in a change of possession rather than just a different shot. Anyway, to the study......
The author is a bit of a clown in the introduction in that he starts running on about close shots being more Points Per Shot than 3s using data on shots including fast breaks. Obviously the numbers are going to be skewed by fast break uncontested layups. In any event, that doesn't really effect the numbers he used in the study because there he only looks at situations where a defender is within 5 feet of the rim.
Boom points to Dwight having the lowest percentage of shots attempted close range on him at 48.2%. Cousins has almost identical numbers(48.4%). But they also give up an incredibly high 3 point % and a high number of 3 point attempts. Dwight also gives up high mid range % while Cousins is solid here. Looking at the data, it seems that the percentage of shots taken close corresponds as heavily with bad 3 point defense as it does good interior defense.
When Dwight is within 5 feet of the bucket, the opposing team averages .482*.457*2(short range) + .32*.382*2(Mid) + .191*.436*3= .935 points per shot. Finding the Points per Shot of the 15 players who deter the most shots from being taken at the rim(Dwight Effect) plus Ibaka, who deterred the fewest shots of qualifying players, Dwight is 10th out of those 16.
The defenders Dwight is ahead of are Chris Kaman, Kevin Love, Tyler Zeller, Aaron Gray, a KG(I'm sure he would have been better 5 years ago), and Andrei Biedres. Some real stalwart defenders there. Leading the pack BY FAR in lowest Points Per Shot by the opponent are Serge Ibaka(.76 PPS) and Roy Hibbert(.776 PPS). About what you would expect from the 2 most dominated shot blockers(AD didn't qualify for this year and a half old study)
In conclusion:
1) The Dwight Effect exists but it does not correlate with good defense or helping the team unless you can somehow prove it causes turnovers at a higher rate(I would think the opposite-fewer turnovers for teams shooting 3s than going inside In any event, the study certainly does not do that.)
2) The player with the worst "Dwight Effect"(Ibaka) allows the fewest point per shot by the opponent when playing the role of "rim protector"
3) Hibbert and Ibaka(albeit he had a small sample size) are men among boys in anchoring a teams defense.
4) 3 point defense is very important in the NBA
5) Dwight is a better defender than Kevin Love and Chris Kaman
So next time somebody tries to claim Kevin Love is a better defender than Dwight Howard, Boom can reference the "Dwight Effect" and make them look stupid!
Study Here
5
Posted on 10/30/14 at 1:00 pm to TheSexecutioner
quote:
2) The player with the worst "Dwight Effect"(Ibaka) allows the fewest point per shot by the opponent when playing the role of "rim protector"
If the calculations are true, thats all that matters, no other point you wanted to add in his as much weight as this.
A single solid defender that drastically reduces his opponents PPS is more important than any interior defender that forces a defender to take an outside shot.
Why? It becomes useless against players like LaMarcus Aldridge and Kevin Love who take majority of their shots from mid-range, rendering an interior defender useless. It does matter that they force slashers that are bad mid range jumpers though, so I do agree it does bear weight.
As opposed to a defender like Serge Ibaka who can defend out to the 3pt line, he lowers the opponent's offensive output based on his defense everywhere.
I'll take the guys that can lower a team's PPS rather than a guy forcing players to take lower % shots. Two different kinds of defense, but the PPS is the one that matters in the end because thats how much you get scored on.
Posted on 10/30/14 at 1:01 pm to TheSexecutioner
As for case study 2(opponent field goal % when a player was within 5 feet of the shooter), I didn't really address that because Dwight Howard is 15th there. Not terrible, but behind monster defenders like Larry Sanders, Hibert, MicGee, Duncan.
Also note, I think Dwight is a good defender. But what makes him good is his incredible rebounding paired with pretty good(but not quite elite) shot defense. Not the "Dwight Effect." So Boom really needs to stop saying that like it proves anything.
Also note, I think Dwight is a good defender. But what makes him good is his incredible rebounding paired with pretty good(but not quite elite) shot defense. Not the "Dwight Effect." So Boom really needs to stop saying that like it proves anything.
Posted on 10/30/14 at 1:04 pm to htran90
quote:
If the calculations are true, thats all that matters, no other point you wanted to add in his as much weight as this.
Yeah, agreed. But read my whole post. Pretty much 90% of it is based on that but with much bigger sample size than one player with a relatively low sample.
But yes, this is 100% based on us taken it as given that PPS are the determinant of success in a study absent of Turnover and Rebound figures. I don't see how anybody could argue that.
Posted on 10/30/14 at 1:07 pm to TheSexecutioner
Although everything you typed was spot on, im sorry you had to spend the energy to type all of that up. Most of the Pelican talk posters realize boom is a joke.
Posted on 10/30/14 at 1:10 pm to TheSexecutioner
Oh no, its a good write-up, I wasn't dissing it because I read it. I just think that one line bears every weight you needed to say.
The extra stuff will be used against you and that line will be ignored by Boom
The extra stuff will be used against you and that line will be ignored by Boom
Posted on 10/30/14 at 1:10 pm to TheSexecutioner
Nice discussion, just like other statistics the calculations used to formulate the Dwight effect need to be combined with other forms of measurement to get the whole picture. In Dwights case 3x defensive player of the year and 4x all nba first team defender say more about Dwight's effectiveness as a defender than even the Dwight effect paper could.
Posted on 10/30/14 at 1:14 pm to wallowinit
quote:
Although everything you typed was spot on, im sorry you had to spend the energy to type all of that up. Most of the Pelican talk posters realize boom is a joke.
Oh I know. But I am interested about the results, not just fact-checking Boom. Obviously, I did it more to illustrate how much Boom is misusing The Dwight Effect trump card, but there are still some interesting facts to be seen from it. Eager to see how AD would fit in after a couple of years.
Posted on 10/30/14 at 1:15 pm to Boomshockalocka
quote:
Nice discussion, just like other statistics the calculations used to formulate the Dwight effect need to be combined with other forms of measurement to get the whole picture. In Dwights case 3x defensive player of the year and 4x all nba first team defender say more about Dwight's effectiveness as a defender than even the Dwight effect paper could.
Exactly. Like I said Dwight is a great defender. His rebounding is what makes him elite IMO. But using Dwight Effect as a trump card is not correct when we discuss something like blocked shots. He is not a better shot blocker or shot deterrer than Ibaka/Hibbert/etc, and you have no evidence to claim he's better than Davis, which is how you seem to always use it.
Posted on 10/30/14 at 1:17 pm to Boomshockalocka
Lol Kobe has 9 all-NBA first team defender. That award doesn't mean dick.
And why do we care about something Dwight did 7 years ago?
And why do we care about something Dwight did 7 years ago?
Posted on 10/30/14 at 1:17 pm to TheSexecutioner
quote:Yes it is very interesting to see these types of studies, they are good for fans to kinda get past the traditional stats as the end all be all.
Oh I know. But I am interested about the results, not just fact-checking Boom. Obviously, I did it more to illustrate how much Boom is misusing The Dwight Effect trump card, but there are still some interesting facts to be seen from it. Eager to see how AD would fit in after a couple of years.
Posted on 10/30/14 at 1:26 pm to Boomshockalocka
quote:
In Dwights case 3x defensive player of the year and 4x all nba first team defender say more about Dwight's effectiveness as a defender than even the Dwight effect paper could.
We should congratulate Magic fans
Posted on 10/30/14 at 1:28 pm to TheSexecutioner
Defensive win shares
Defensive rating
Defensive real plus/minus
Defensive rebounding %
Opponent FG%
Opponent point per shot
All favor D12.
But but Davis blocks more shots, right?
Defensive rating
Defensive real plus/minus
Defensive rebounding %
Opponent FG%
Opponent point per shot
All favor D12.
But but Davis blocks more shots, right?
This post was edited on 10/30/14 at 1:34 pm
Posted on 10/30/14 at 1:30 pm to Boomshockalocka
quote:
Yes it is very interesting to see these types of studies, they are good for fans to kinda get past the traditional stats as the end all be all.
I think what the author failed to consider(or maybe not since he never really points to Dwight Effect being effective) is that although as a team you don't want opponents shooting close to the rim(including fast breaks, unprotected rims, rims protected by backup scrubs), you might want them shooting close when you have an elite rim protector there. When Dwight/Ibaka/Hibbert/AD/etc are at the rim, allowing a team to chuck up a 3 as opposed to challenging those monsters is an "out" that helps the team.
What I gather from this study is that 3 point D is incredibly important to dis-incentivize a team from avoiding the big. In some cases(Ibaka/probably AD), the big man can help himself by playing an active role in helping in team 3 defense. In other cases(probably Hibbert), they allow a great defensive surrounding group. Dwight probably has a combination of never playing with elite defensive guards and not being exceptionally effective against perimeter play himself.
Posted on 10/30/14 at 1:33 pm to Boomshockalocka
Dude. Davis is 21. Dwight is on the other side of his prime.
Posted on 10/30/14 at 1:36 pm to Boomshockalocka
quote:
Defensive win shares
Defensive rating
Defensive rebounding %
Opponent FG%
Opponent point per shot
All favor D12.
But but Davis blocks more shots, right?
From when? AD's past year vs Dwight's last year? Even if thats true, AD clearly improved as the year went on. Using the average of a player constantly sloping upwards is not useful. Yeah, I understand you want a bigger sample size, but you really just can't do that with AD. The slope is too much to use an average. He's improving too fast and everybody should know that.
In any event, I've seen you run on about the Dwight Effect countless times. Now that that was exposed as not indicating quality, you have chosen other metrics. Makes me suspect you are cherry picking stats to fit your desired conclusion.
Posted on 10/30/14 at 1:52 pm to TheSexecutioner
He thinks we, PT, are overrating the Pelicans and AD.
In reality, its most sports head pushing AD as #3 behind LJ and KD.
In reality, its most sports head pushing AD as #3 behind LJ and KD.
Posted on 10/30/14 at 1:57 pm to TheSexecutioner
You guys need to stop giving Boom what he wants on this board. You're playing right into his trap if you have 4 page arguments with the dude in every thread on here.
Posted on 10/30/14 at 2:01 pm to TotesMcGotes
Yea so that means in a couple years Davis will be better, unfortunately for you that isn't the discussion.
Posted on 10/30/14 at 2:02 pm to danman6336
Hey as long as people don't bash Dwight / Harden / Rockets on here you'll be hard pressed to find me posting here. I'm gonna set the record straight when I see people posting misinformation about my guys. 
Page 1 of 5
Popular
Back to top
Follow TigerDroppings for LSU Football News