Statistical Analysis of Post-Trade Team: There is Hope!

[nate33]

My gut feeling is that the team has played much better since the Arenas trade (the Charlotte debacle notwithstanding). I know our record remains roughly the same, but the team doesn't look the same. I decided to run some numbers to find out. I used offensive and defensive rating, which is a measure of points scored and allowed per 100 possessions.

Prior to the trade, our team ORtg was 100.3, which ranked 24th. Our team DRtg was 109.3, which ranked 29th. Based on our point differential, we projected to have a win % of .232 which translates to 20 wins on the season.

Here is the good news:

Our team following the trade has an ORtg of 97.7, which ranks 29th. We've gotten a little worse. However our team DRtg is an incredible 98.6, which ranks 5th. We have improved from being the 29th best defense to the 5th best defense! Moreover, although we have won only 3 of 11 games since the trade, our wins have been big and our losses have been close. If you use point differential to determine our expected win/loss record, we project to have a .470 win %. That means, the post trade team is playing as good as a 39 win team!

Digging deeper, we can see how the defense has improved. Our pre-trade defense allowed opponents to shoot an eFG% of .519. Our post-trade defense allows an eFG% of .490. Our pre-trade defense forced 16.8 turnovers a game. Our post-trade defense forces 18.9 turnovers. And most notably, our pre-trade team grabbed 67.3% of the available defensive boards. Our post-trade team grabs 73.6% of the available defensive boards. That's a huge swing. We went from being the worst defensive rebounding team to an average rebounding team. (An average team grabs 73.8% of the available defensive boards.)

The losing is frustrating, but let's not lose sight of the goal. Our goal is to become a better team regardless of the win/loss record. Indeed, if we can play well but still lose close games, that's ideal from a draft positioning standpoint. The data suggests that Flip is getting through to these guys. The defense is playing much, much better. And defense is the foundation for success because you can play defense every night.

[nate33]

Here's another thing.

If Blatche played exactly as he has been playing, but only hit his jumpers (anything outside of 10 feet) at the same percentage he shot last year, the team would average another 2.2 points per game. Add those two points to our offensive efficiency and it improves to 99.8, which brings us up from 29th to 24th, brings our point differential to +1.2, and our expected win total to 45 wins.

So if one wants to be optimistic, one can say that this team, as currently constructed, is capable of playing 45 win basketball if only Blatche could return to his shooting prowess of last year. Nothing else is needed. We don't need to fire the coach. We don't need to get a bruising low post presence. We don't need to find a better swing man. The team is playing pretty good right now, except for Blatche's shooting. (Whether or not Blatche's 2009 season was just a fluke remains to be seen.)

[FAH1223]

Flip has returned to Yi. Its amazing how he doesn't even try on defense or try to grab a board.

This will make our defense crappy again.

Its gotten to a point where I hope he gets injured again

[MJG]

Does the point differential actually tell us something? I wouldn't say "our losses have been close, our wins have been big" so much as "our losses have been big, our wins have been huge." Our eight losses were by 9.5 points, our three wins were by 22.3 points. I know in a bubble, it's the net that matters, but there can't be any sort of precedent for that kind of gap, so I'm hesitant to take anything away from it just yet.

Regardless, the defense is clearly better (I can tell even with my limited watching), and that's good enough for me for now.

[hands11]

My gut feeling is that the team has played much better since the Arenas trade (the Charlotte debocle notwithstanding). I know our record remains roughly the same, but the team doesn't look the same. I decided to run some numbers to find out. I used offensive and defensive rating, which is a measure of points scored and allowed per 100 possessions.

Prior to the trade, our team ORtg was 100.3, which ranked 24th. Our team DRtg was 109.3, which ranked 29th. Based on our point differential, we projected to have a win % of .232 which translates to 20 wins on the season.

Here is the good news:

Our team following the trade has an ORtg of 97.7, which ranks 29th. We've gotten a little worse. However our team DRtg is an incredible 98.6, which ranks 5th. We have improved from being the 29th best defense to the 5th best defense! Moreover, although we have won only 3 of 11 games since the trade, our wins have been big and our losses have been close. If you use point differential to determine our expected win/loss record, we project to have a .470 win %. That means, the post trade team is playing as good as a 39 win team!

Digging deeper, we can see how the defense has improved. Our pre-trade defense allowed opponents to shoot an eFG% of .519. Our post-trade defense allows an eFG% of .490. Our pre-trade defense forced 16.8 turnovers a game. Our post-trade defense forces 18.9 turnovers. And most notably, our pre-trade team grabbed 67.3% of the available defensive boards. Our post-trade team grabs 73.6% of the available defensive boards. That's a huge swing. We went from being the worst defensive rebounding team to an average rebounding team. (An average team grabs 73.8% of the available defensive boards.)

The losing is frustrating, but let's not lose sight of the goal. Our goal is to become a better team regardless of the win/loss record. Indeed, if we can play well but still lose close games, that's ideal from a draft positioning standpoint. The data suggests that Flip is getting through to these guys. The defense is playing much, much better. And defense is the foundation for success because you can play defense every night.

Thanks for gathering these stats. This is what I have been posting about. Easy to take things as opinion when there are no hard stats to support the position. Not that facts will get in the way of what some believe, but I will help others.

[nate33]

Does the point differential actually tell us something? I wouldn't say "our losses have been close, our wins have been big" so much as "our losses have been big, our wins have been huge." Our eight losses were by 9.5 points, our three wins were by 22.3 points. I know in a bubble, it's the net that matters, but there can't be any sort of precedent for that kind of gap, so I'm hesitant to take anything away from it just yet.

Regardless, the defense is clearly better (I can tell even with my limited watching), and that's good enough for me for now.

Statistically, point differential is a better predictor of future performance than win-loss record. A 19-22 team with a +2.0 point differential is more likely to have a winning record over the second half of the season than a 22-19 team with a -2.0 point differential.

If there's a big fat hole in my statistical analysis, it's not my use of point differential, it's my use of a relatively small sample size.

[hands11]

Does the point differential actually tell us something? I wouldn't say "our losses have been close, our wins have been big" so much as "our losses have been big, our wins have been huge." Our eight losses were by 9.5 points, our three wins were by 22.3 points. I know in a bubble, it's the net that matters, but there can't be any sort of precedent for that kind of gap, so I'm hesitant to take anything away from it just yet.

Regardless, the defense is clearly better (I can tell even with my limited watching), and that's good enough for me for now.

I would be more encouraged by the winning spread. To me it shows they can do it and the problems are more mental then physical.

[20MexicanosIn1Van]

How is the projected win percentage extrapolated from point differential?

[tontoz]

Statistically, point differential is a better predictor of future performance than win-loss record. A 19-22 team with a +2.0 point differential is more likely to have a winning record over the second half of the season than a 22-19 team with a -2.0 point differential.

Very true. Statistically there is reason for optimism but watching the offense sputter it is hard to stay very optimistic.


As far as the rebounding goes Shard has to be the big difference maker post-trade. His rebounding has been a big surprise. Hopefully it will rub off on the other bigs.

[JWizmentality]

Flip has as much hope as that deuce I just dropped in my toilet.

[Induveca]

0-18 on the road. 9-26 overall.

It goes well beyond Blatche, this team is mentally weak, unprepared and poorly coached. The 0-18 in particular proves that beyond doubt.

[Dat2U]

Stats lie. This is a great example of it. We can manipulate certain stats to say anything but nothing is better than actually watching the product on the court to get a real opinion.

This team is brutal to watch. Passable at home and historically bad on the road. That's a perfect example of a bad basketball team. There is hope, but it comes in the form of John Wall and little else. Maybe Javale McGee too depending on what day of the week it is.

Indu, not only is this team mentally weak but it displays an incredibly low b-ball IQ. Were not very skilled or smart. Maybe we have our fair share of athletes but the decision making is shockingly bad, especially considering how good our coaching staff is supposed to be.

[DallasShalDune]

Thanks for posting that, nate. It'll make watching this team easier knowing there is some improvement. McGee and Blatche have been defending better post trade. So has nearly everybody else. It helps that we have two vets on the healthy roster now (Howard and Lewis) who can play D. I hope Howard stays fine because he's better than Thorton or any other option at SF other than Lewis.

[nate33]

Stats lie. This is a great example of it. We can manipulate certain stats to say anything but nothing is better than actually watching the product on the court to get a real opinion.

This team is brutal to watch. Passable at home and historically bad on the road. That's a perfect example of a bad basketball team. There is hope, but it comes in the form of John Wall and little else. Maybe Javale McGee too depending on what day of the week it is.

Indu, not only is this team mentally weak but it displays an incredibly low b-ball IQ. Were not very skilled or smart. Maybe we have our fair share of athletes but the decision making is shockingly bad, especially considering how good our coaching staff is supposed to be.

There is no "manipulation" of the stats. The stats are what they are.

I agree that the team is mentally weak and brutal to watch. I agree that we're not very skilled or smart. But nevertheless, the post-trade team defends. Defense may not be fun to watch, but it's effective. It's easy to lose sight of this amidst the horrible offensive execution and the constant losing.

The point is, there's a viable foundation being built. Despite all the ugliness, mental weakness and bad offense, this team is in a lot of games and plays good enough defense to win them. There is a light at the end of the tunnel, and it's not an oncoming train. If the team can just develop enough experience to execute down the stretch, they could be good. And in our point guard dominant offense, I think it's very likely that we will learn how to execute down the stretch as John Wall matures.

[Hoopalotta]

There is no "manipulation" of the stats. The stats are what they are.

It doesn't get any more meat and potatoes than than DRTG and ORTG. Those are "four factors" numbers.

[Donkey McDonkerton]

Awesome, I will be celebrating in the streets! We can be a 45 win team.....wooo hoo!

[AceDegenerate]

Honestly, I've seen nate do this for years now. This is just sad.

I don't care what stats you want to pull out of thin air, the team is 9-26 and 0-18 on the road. Until those numbers change you are only patting yourself on the back for being a fan of this team by trying to justify the record with stats like this.

Using an average point differential to determine how many wins the team can gather? I mean, c'mon, who really thinks this has any bearing in reality?

[AceDegenerate]

Moreover, although we have won only 3 of 11 games since the trade, our wins have been big and our losses have been close. If you use point differential to determine our expected win/loss record, we project to have a .470 win %. That means, the post trade team is playing as good as a 39 win team!

It's rationalizations like this that make your whole use of stats suspect. You cannot simply say our wins have been big and our losses have been close. This is not the NHL no matter how bad Ted would like it to be. A win counts as 1 win and a loss counts as 1 loss. Any use of statistics to show otherwise should be construed as manipulation for the purpose of skewing an argument your way.

[Hoopalotta]

Using an average point differential to determine how many wins the team can gather? I mean, c'mon, who really thinks this has any bearing in reality?

John Hollinger.

[Hoopalotta]

I will say though that I'm not expecting the wins to start rolling in; not by any means. Our offense is TERRIBLE and we totally lack a reliable first option, so that would be tough even in a vacuum.

Also, we've got 3 "easy games" (yeah, I know), but then it's straight murderers row for 21 straight games (I posted the list in the fire Flip thread). Realistically, that's probably going to be a 5-16 stretch, so that could be 16-43, easily. 16-43 might even be optimistic. Thereafter, the schedule cleans up quite a bit with some cup cakes, so there might be some winning depending on a great many variables.

So I'm definitely not going on record saying I'm expecting much as far as actual wins and losses.