Yet another "How will the Rockets do?" Post.
As you may have heard, the 2010 Rockets don't look all that much like the 2009 Rockets. And that worries some of the fans. The basketball community at large seems to have mostly dismissed the Rockets' chances (though, as Kelly Dwyer put it in one of his chats, "They've surprised us too many times to completely write them off."), and a lot of us seem to expect a pretty crappy season.
Some, however, do not share in this view, and that includes the fellows over at Wages of Wins. WoW has a lot of detractors, of course, and I think that they fall into the same trap that anyone seeking some big rate statistic to capture everything in basketball falls into, but WP48 is an interesting stat (and, if nothing else, probably a lot more accurate than Hollinger's PER). For those who don't know, WP48 is a rate statistic derived from all boxscore statistics - Points, FG%, Rebounds, Blocks, etc. The formula is set so that a WP48 of 0.100 is the NBA average.
Somewhat ironically (actually, not ironic at all), given Morey's stated issues with boxscore stats, proponents of WoW seem to end up advocating basically what Morey advocates: there's a lot more than scoring in a basketball game, and certain players (guys like Iverson, Ben Gordon, Von Wafer, or any other "pure scorer") tend to be overvalued by NBA teams (evidenced by their contracts) and fans (evidenced by popularity). Correspondingly, there are many players (Guys like Chuck Hayes, Dennis Rodman, or even some star players like Tim Duncan or Chris Paul) who tend to be undervalued by the same groups. Essentially, the lessons that WP48 provides are:
1) "Scorers" are not as important as we tend to think.
2) Rebounding - particularly defensive rebounding - is critical to winning. A defensive stop is not acheived unless there's a turnover or a D-Reb, so great rebounders are much more important than most think.
3) Defense (and this is where WP48 tends to get its stats-inclined detractors, because it assigns almost all of the credit for a defensive stop to the rebounder) is undervalued.
That sounds a lot like what Morey says in every interview he has ever given, right? And, if you look at the WP48 scores for the Rockets, it's pretty much a WoW-dream, as Dr. Berri reminds us:
The Above Average Rockets
Then I looked at the following list of veteran players who should play for the Rockets this season.
Luis Scola: 2,448 min., 9.8 Wins Produced, 0.189 WP48
Shane Battier: 2,031 min., 6.2 Wins Produced, 0.147 WP48
Trevor Ariza: 1,998 min., 8.0 Wins Produced, 0.192 WP48
Aaron Brooks: 1,998 min., -0.5 Wins Produced, -0.012 WP48
Carl Landry: 1,467 min., 5.4 Wins Produced, 0.175 WP48
Chuck Hayes: 858 min., 2.5 Wins Produced, 0.139 WP48
Brent Barry: 857 min., 2.0 Wins Produced, 0.113 WP48
Kyle Lowry: 608 min., 2.3 Wins Produced, 0.182 WP48
An average NBA player posts a 0.100 WP48. Of the eight players listed above, seven were above average last season.
Excluding Brooks (whom I'll get into in a moment), all of these players area above-average. And these aren't one-time statistical increases:
With the exception of Aaron Brooks, these players were above average before 2008-09. This suggests that these players will be above average in 2009-10. If that happens, the Rockets have a good chance of being an above average team.
This shouldn't be too surprising. Morey clearly values "reboundy" (if undersized) power forwards (enough that I was fairly surprised that the Rockets didn't trade for DeJuan Blair on draft night), and the Rockets tended to outrebound their opponents last season. They play good defense, so they get more chances to rebound, and so the Daryl Morey Rockets (except ABZ) looks a lot like a WoW team.
So, what about Brooks? Is he really a negative contributor to the team?
Brooks is basically the type of player (like Gordon, Iverson, or Wafer) who is going to be overvalued by the NBA establishment. He scores, and that's about all he does. I think he passes better than his assist numbers suggest (and, in any case, Adelman clearly feels that he passes well enough to be given the starting spot), but he doesn't rebound at all (he grabs just 4.7% of rebounds when on the court), he's somewhat turnover-prone, and he has glaring defensive issues. I like Brooks, but those are his problems.
WP48 suggests that these problems are enough to overcome his scoring contributions. Maybe that changes a little if Brooks' improvements in the playoffs were permanent, but it's not going to turn him into the WP48 messiah.
Berri argues that the Rockets don't really have a problem because scoring is only part of the game. And that's true - it's only half of it. But the team has to score to win, and it seems to me that the Rockets are likely going to be able to score more points than their opponents with Brooks on the court. His defensive issues aside, the marginal payoff for more rebounding and defense is probably much less for the Rockets than for extra scoring. Again, assuming that he demonstrates an improvement next season, I think that Brooks simply fits better than most other guards. Consider this one of those "context issues" that rate stats have.
The real question is if he's a better option than Lowry. That's difficult to answer, given Lowry's superior passing abilities. Berri would state that Lowry should clearly be the starting PG, but it's a more complex question than he thinks.
This has been something I've thought about for the past two months, and it's suggested in the comments section: The '09 Rockets (if McGrady either doesn't return to form or doesn't contribute until late) are something of a Grand Statistical Experiment. We have a collection of non-scorers and "non-creators," guys who Dr. Berri and Morey seem to value highly (or, rather, more highly than most). If the Rockets, as currently constructed, can play well over the first half of the season (until the trade deadline hits and the stupid/poor teams start selling off their players, I guess), it will have cemented Morey's genius and made guys like Dr. Berri look very smart. If not... well, I guess most of us weren't expecting much, anyways.
So read the article, even if you have your doubts. Regardless of the merits of the statistic, I think the Rockets really are going to do much better than most think.
No cursing in title. No pirated material, such as links to online game streams. Thanks.
1 recs |
20 comments
Comments
Good points
I do think that while statistical efficiency is important to consider when building a roster, it should be complimented by a scorer-type. Like you said, we have a bunch of non-creators this year. If you pair the guys we have with a healthy Yao or McGrady, I love it.
We’ll see how far unsexy efficiency can take us this year.
The Dream Shake ...on Twitter.
"I think girls are probably just better shooters." - Steve Novak
by Tom Martin on Aug 14, 2009 5:19 PM CDT reply actions 0 recs
Perhaps the injury difficulties the team has had with TMac and Yao might have played into the decision to pass on DeJuan Blair?
by DribbleHooper on Aug 14, 2009 5:55 PM CDT reply actions 0 recs
maybe
Part of it was the injury thing. I’m sure part of it was the fact that the Rockets, even with Yao out, have plenty of talent in the frontcourt. They needed wing players on draft night, and that’s what they got.
Your friendly neighborhood Dream Shake mod.
by Only_A_Lad on Aug 14, 2009 6:01 PM CDT up reply actions 0 recs
Wait
If the Rockets do exceed expectations and if WP48 becomes more telling than PER (I hate it too), would it become more accepted by the basketball community?
What I’m getting at is: instead of being the no-stats all-stars, would the Rockets in fact have a group of well-regarded players whom other teams will covet? This could cause big-time problems for the Rockets in the future.
Finally, wouldn’t it almost force all teams to adopt a WP48-type statistical analysis? Obviously, Morey’s ability to evaluate isn’t solely based on statistics, but I think it would cause him to be less effective as a GM due to other teams adjusting.
I know Bill Simmons doesn’t see the stats era of the NBA coming anytime soon, but a successful 2009-10 Rockets team could bring that about; but after thinking about it, I don’t know if that’s what I want.
How many Biletnikoffs does he have? NOT TWO!
by ak2themax on Aug 14, 2009 10:02 PM CDT reply actions 0 recs
it takes ppl alot to see whats staring them in the face.
morey has been making great trades, considered the top gm by many – among the top 3 by pretty much every1.
yet ppl still dont really belive in his stats based gming. the nba isnt really adjusting
by AlDe2356 on Aug 14, 2009 10:56 PM CDT up reply actions 0 recs
The NBA is adjusting in some ways.
First, there’s the simple fact that there are other statistically-inclined GMs out there (Presti, Pelton).
Beyond that, we’re starting to see some of the things Morey values get valued by other teams. The best example, I think, would be Brooks. I remember reading a few articles after the Lakers series talking about how many GMs wanted a quick little PG (though the league had been moving in that direction, anyways). If nothing else, Brooks’ success at the starting position might have made some more executives and coaches more willing to accept undersized, speedy guards (Lawson, Flynn).
And plenty of teams have statistical divisions and look at advanced stats work, but none of them have made as large of an investment in statistical analysis as the Rockets have.
Your friendly neighborhood Dream Shake mod.
by Only_A_Lad on Aug 14, 2009 11:10 PM CDT up reply actions 0 recs
pelton is by far the most overrated gm in the nba
by Ben. on Aug 17, 2009 5:37 PM CDT up reply actions 9 recs
heh
whoops. Meant Pritchard.
Your friendly neighborhood Dream Shake mod.
by Only_A_Lad on Aug 17, 2009 8:41 PM CDT up reply actions 0 recs
Well, if they make the playoffs,
there’s a few angles that will probably be played in the media:
- As Berri notes, a lot of credit will go to Brooks (he thinks undeservedly). Some of it will probably go to Ariza, too. So “conventional” analysts will have a clear explanation for what happened: “Brooks is a great PG and Ariza is a star” or some such.
- Again, as Berri says, Morey will probably win Exec of the Year. He’ll be recognized as awesome, and more people will be open to the idea of statistical analysis in basketball.
- Those who predicted it will be seen as geniuses, too.
Will it rob the Rockets of their edge, the way it did Oakland in baseball? Probably not, simply because virtually all of the Rockets’ stats are kept very secret, and we have very little idea of what they look at (Morey has said they look at +/-, and that they keep track of their own boxscore stats, but that’s about all we know). Maybe more Computer Science grads get hired by teams, but that’s not going to change much.
For guys like Berri (and all the other stats analysts on the web), I think they’ll get more respect (somewhat ironically, I guess, since Morey hasn’t seemed too impressed with some of the WoW stuff in interviews, and he’s probably not just assembling a team by looking at WP48 or any other rate stat). Maybe ESPN starts carrying WP48 alongside Hollinger’s PER or something. Maybe Yahoo does it. I can only hope, because while I think WP48 is an analytic dead-end, it at least is better designed than PER.
But, no, if the Rockets make the playoffs, it won’t change much in the NBA scene. Some will say it’s a fluke, others will just look at increased scoring from a few players (and, just as they do with so many players, ignore increases in playing time or other factors explaining improvements in scoring numbers), and a lot will say the Rockets just have “hustle” and “heart” or whatever. Nobody is going to go out and start paying Chuck Hayes the $10 million or so he was worth in 2007 (that’s according to Berri’s calculation, anyways). NBA people (and the vast majority of fans) just won’t believe that a guy like Hayes is worth just as much as many “stars.” Damn few really believed the Battier thing, anyways.
Your friendly neighborhood Dream Shake mod.
by Only_A_Lad on Aug 14, 2009 11:30 PM CDT up reply actions 0 recs
I know this is silly but...
I do player edits for a video game, NBA 2K9 (and generally, I’m considered the best). I use a lot of the general advanced statistics you can find on basketball-reference.com, read lots of scouting reports, and watch what tape i can get my hands on.
Based on my edits, the Rockets without Yao and McGrady are actually the #8 west team.
This isn’t exactly ground-breaking, since this is essentially a 9 team conference. Conventional wisdom should tell us that a team with Luis Scola, Trevor Ariza, and Shane Battier will be vastly better defensively and more efficient offensively than say, the Grizzlies led by Zach Randolph, Rudy Gay, and O.J. Mayo.
The Rockets have the lowest rated offense, but they also have the 9th best defense. And it’s defense that wins in this league. A bad defense will make a bad offense look great. It doesn’t work the other way around.
by Real 2K Insider on Aug 15, 2009 1:54 AM CDT reply actions 0 recs
In case you guys are interested
The bulk of my work can be found on thereal2kinsider.blogspot.com. I’ll probably make a Rockets writeup tonight. Here’s the overall ratings of the roster.
91 Yao (injured)
87 McGrady (injured)
84 Scola (avg starting PF: 83)
80 Landry
80 Ariza (avg starting SF: 83)
79 Battier (avg starting SG: 83)
79 Brooks (avg starting PG 84)
79 Lowry
75 Barry
74 Andersen (avg starting C: 81)
73 Hayes
72 Budinger
72 Taylor
71 Cook
70 Dorsey
The talent is a little thin right now, but between Landry and Lowry they at least have reasonable depth. The Suns are a five player team, and two of those guys are Nash (35) and Hill (37). They just gave up Shaq for nothing after a season where they missed the playoffs, which is pretty much the same as losing Yao for the season.
by Real 2K Insider on Aug 15, 2009 2:06 AM CDT reply actions 0 recs
love that game
Do you have an updated roster available for download? Id like to simulate a season and see where the Rockets land.
by jaydottcomm on Aug 15, 2009 10:27 AM CDT reply actions 0 recs
Roster
I’m working on one, it’s about 75% complete
by Real 2K Insider on Aug 16, 2009 2:50 AM CDT reply actions 0 recs
How does this differ from Win Shares?
The numbers are close, but not quite the same. (Unless the WP is counting playoffs.)
Anyway, the one issue I have with these is that they aren’t minute/injury adjusted. If you look at minute adjusted WS, Landry becomes one of, if not the, most underrated players in the NBA. Lowry is also underrated, although his sample size of minutes is also pretty small.
Anywho, I’ve been screwing with the WS formula because of the lack of minute adjusting. Also, WS overrates role players on good teams and underrates good players on bad teams. (Regardless of what WS says, Landry is not better than Bosh.)
If anyone has an educated guess at the minutes breakdown for the Rockets I can give an educated guess at the number of wins you get :-)
Poster for next year? I'm thinking My Little Pony.
by Zaig on Aug 17, 2009 11:55 AM CDT reply actions 0 recs
WP48 weights the data differently.
The biggest differences are:
1) Rebounding is weighted much more heavily by WP48.
2) Volume-shooters are penalized a lot by WP48.
3) WS incorporates D-Rating, which figures defense out in a different way (I haven’t read the book that includes WS, so I only know vaguely how it works).
WS likes players who score more than WP48; WP48 likes players who rebound much more than WS. A good example of this is WoW’s opinion on Troy Murphy vs. the more “conventional” view.
Your friendly neighborhood Dream Shake mod.
by Only_A_Lad on Aug 17, 2009 8:48 PM CDT up reply actions 0 recs
Interesting, not sure about the Troy Murphy example though
He’s an efficient scorer and a crazy good rebounder. Shouldn’t both of the systems love him? (Win share had him at 9.0, above Granger with almost identical minutes, so I know it does.)
Murphy’s major downside is his low volume.
2.8/5.6 2pfg a game (50%)
2.2/4.9 3pfg a game (45%)
Those are some nice numbers, but he’s only taking 10.5 shots in 34 mpg. If he can up that to 13 shots and keep his FG%, he should be an All-Star. (That much more improvement for a guy his age would be unusual, but it could happen.)
Poster for next year? I'm thinking My Little Pony.
by Zaig on Aug 18, 2009 12:03 PM CDT up reply actions 0 recs
yeah, both like him
but WP48 had him as – by far – the best PF in the league last year. 17.0 wins produced, good for the 7th best WP in the league. And he was a full 7 wins ahead of the next-best PF (Garnett).
Your friendly neighborhood Dream Shake mod.
by Only_A_Lad on Aug 18, 2009 12:37 PM CDT up reply actions 0 recs
Interesting
Although I’m not sure if I like Przybilla (1900 minutes) producing as many wins as Yao Ming (2500) minutes. Especially when Yao is clearly more important to his team. I think it probably overinflates a good shooting percentage on low shots. (Billa shot crazy good, but when you take so few shots, it really doesn’t help a ton. Give me a 55% shooting center who can take 15 a game over a 65% guy who shoots 4-5 a games.)
I’ll have to look into WP more when I get some free time though.
Poster for next year? I'm thinking My Little Pony.
by Zaig on Aug 18, 2009 3:30 PM CDT up reply actions 0 recs
Brooks
By adjusted +/- he was a real drag. Without Yao and McGrady making things easier for him it should be even worse but it hardly can get worse.
He’ll put up a good number of 20+ point games but I think he’ll put up fewer victories than hoped for.
by StatRaven on Aug 18, 2009 2:29 AM CDT reply actions 0 recs
Low shooting percents
TS% of .521 and eFG% of .476. Dropping 20+ means nothing if you do it inefficiently. It’ll make the TV people love you, but it won’t win you games.
Yao was 618, 549
Scola 573, 531
Landry 634, 575
Artest 512, 475 (lawl)
Wafer 541, 510
One game from last year that makes a good example. Beating Portland on Feb 24. Everybody in Portland complained about the defense of our PGs because Brooks was allowed to drop 20 on us. The problem is that he did this on 8/20 shooting, with only 2 assists and 2 FTs attempted. That’s bad. That means in 21 offensive trips (20 shots 1 trip to line), he got you guys 20 points. Or less than 1 point per possession, which is BAD.
The real reason you guys won the game was because the defense held Roy/LMA to 17/42 shooting, or 40.4% (the same that Brooks shot.)
So after the game all the TV disussed Brooks 20 points and Ron Ron’s 21 (on 5/13 shooting), COMPLETELY ignoring why the Rockets actually won. Sure, Roy and LMA had 24/21, but they did it inefficiently, thus the lose.
(The real stuf of that game was Scola. 5/9 with 4 offensive rebounds. That’s some efficiency right there.)
Poster for next year? I'm thinking My Little Pony.
by Zaig on Aug 18, 2009 11:57 AM CDT up reply actions 0 recs

by 















