Rockets Advanced Stats

There are a couple places on the web you can go to for some nice, advanced stats on the Rockets -- beyond what you'd find at ESPN.com or NBA.com.

82games.com is gaining in popularity of late, and it offers some comprehensive information regarding team and individual performance for the avid fan.

http://www.82games.com/0506/0506HOU.HTM

Knickerblogger's stat page is also up to date, and provides advanced "Hollinger" stats for each player. Here's the Houston page:
http://www.knickerblogger.net/stats/2006/jh_Rockets.htm

You can also see league leaders by various statistical categories.

One thing to consider is that its still early in the season, so statistics don't mean quite as much as toward the end of the season when there's more sample data to make the numbers really meaningful. Still, some things kind of jump out.

<font size=""4"">Guards</font>

Other than Jon Barry (who's saving grace is he's shot the ball very well) and Luther Head, our guards have been pretty abysmal -- in particular Derek Anderson and David Wesley. They both have 9.7 PER, which wouldn't qualify them to be a backup player.

Based on PER, our guards' performance can be ranked like this:

<font face=""Courier New"">Luther Head - 14.6 (above average backup, average starter)
Jon Barry - 13.0 (average backup)
Rafer Alston - 10.4 (below average backup)
Derek Anderson - 9.7 (just terrible)
David Wesley - 9.7 (just terrible)</font>

PER doesn't measure position defense, however. This is one area where JVG puts a lot of stock into, so maybe that would explain why he's given most minutes to the three guards who've played the worst, taking everything else into account.

The PER ratings correlate pretty well with the +/- for each player, as given by 82games.com.

<font face=""Courier New"">Luther Head : +8.9 per 48 min
Jon Barry : +5.1 per 48 min
Wesley : +1.6 per 48 min
Anderson : -8.2 per 48 min
Alston : -9.4 per 48 min</font>



<font size=""4"">Backup Center</font>

Mutombo has played poorly this year. His rebound rate is slightly down (though still leads the team by a good margin), but he's been a turnover machine and pretty much a non-factor on the offensive end. Last year, the Rockets actually played better when Yao went to the bench with foul trouble, because Mutombo was so effective. This year, we've really struggled (-11.1 per 48 min for the Rockets when Yao is off the court). This, however, is less to do with Mutombo's playing than the very ineffective combination of Howard and Swift. Neither are strong rebounders or provide much of a defensive presence. Which brings us to:

<font size=""4"">Power Forward - Who should start?</font>


Stromile Swift has outperformed Howard by a wide margin, statistically - 16.3 to 11.7 PER. This we need to be careful with, however. Howard has been a somewhat better rebounder, and is definitely a better passer, but scoring-wise has been terrible (11 pp40 minutes with a 45.7 TS% is terrible any way you slice it). Swift has been a solid scorer for us (which makes for the difference in PER), but is lacking in another areas. I do think we could expect Swift's scoring to go down significantly if he was a starter, and Howard's could go up as a backup.

Still, the +/- ratings gives ample reason to prefer Swift instead of Howard as a starter.

<font face=""Courier New"">Yao : +12.4
Bowen: +10.8
Swift: +3.8
McGrady: +3.7
Mutombo: -1.3
Howard: -19.9</font>

You read that last number correctly. Anyone who's watched the games knows that the trend has been for the Rockets to get off to sluggish starts in the 1st and 3rd quarter, and they come back when the more energetic bench players come in for 2nd and 4th quarter. Getting off to better starts to begin each half is crucial for us to turn around the season. I think there's a strong case to be made here for starting Swift over Howard.

 

I Love Bob Sura I Miss Bob Sura I Love Bob Sura I Miss Bob Sura

 

82games.com has published "Roland ratings" for every player in the league. The concept is simple. It takes the PER difference between the player and his opponent and it takes his +/- effect per 48 minutes and averages the two. Actually, it does give more importance to the PER difference by doubling its weight relative to +/-.

Roland Rating = (2*[PER diff] + [+/- diff]) / 3

PER differential captures the player's effectiveness while on the court (a per-possession stat, so it's not biased towards players who play a lot of minutes) as well as his individual defense (i.e., limiting the effectiveness of his counterpart).

+/- differential captures the impact of the player's presence on the court for the team. So, if the team plays well while he's on the court relative to when he's not, he'll have a good +/- rating, and vice-versa.

Here's some info on the Roland Ratings, along with league leaders early in the season:
http://www.82games.com/0506/rolandratings0506.htm

And, here's the Rockets page:
http://www.82games.com/0506/0506HOUH.HTM

Not surprisingly, Yao and McGrady lead the pack. Yao is greatly helped by the ineffectiveness of his backup (Mutombo hasn't played well so far, and the Howard/Swift combination has been terrible for the Rockets so far). He's 14th in the league in Roland Rating.

According to the rating, Howard has been the Rocket's least effective player. I think it's shortchanging his contribution to the team for a couple reasons, however. Howard plays most of his minutes with Yao on the court, so he doesn't get as many scoring opportunities as he would if he was coming off the bench. As a result, early in the season there hasn't been a large enough sample to accurately guage his ability on the offensive end. Also, while Howard isn't a shut-down type of defensive player, he does play solid team defense (certainly compared to Swift) and this aspect of his game isn't covered too well by the Rating. While Howard's +/- is the worst on the team by far, I'm inclined to think that it isn't so much his fault as it is the lackluster play of the starting guards. If I can find some adjusted +/- numbers that take into account who's on the court with the player, that would be more helpful.

 

I actually like Howard's contribution...he's a great offensive rebounder (averages 2 per game) and is solid and setting screens and knowing where to be on the court...Wesley has shown a lot of promise the last two games and has done a nice job of running the team...if he keeps his solid play up, it'll be interesting to see whether JVG will stick with him or Alston as the starting point...

In my eyes, our most effective players are Head, Howard, and Wesley (as of late)...Stro and DA have been inconsistent; JB looks lost and doesnt know when to shoot...Yao posts nice solid numbers but he wastes so many offensive possessions for us due to silly turnovers like traveling or offensive fouls...each possession for this Rockets team is priceless, and Yao has cost us too many...

this team just has to get smarter with their play...

 

I took the latest team stats provided by ESPN.com along with possessions per game for each NBA team to get a sense of how we rank in various categories.

These rankings are as of 11/23/2005. 1/30 is best, 30/30 is worst:

<font face=""Courier New"">Offensive Efficiency (pts/100 poss): 28/30
Defensive Efficiency (" allowed/"): 18/30
FG%: 29/30
Opponent FG%: 2/30
3FG%: 5/30
Opponent 3FG%: 9/30
FT%: 7/30
Off. Reb %: 25/30
Def. Reb %: 18/30
Tot Reb %: 15/30
Pace (possessions/game): 30/30
Point differential (per 100 possessions): 25/30</font>

As expected, it's pretty ugly. These don't necessarily reflect our true rankings with respect to the whole league, given our particularly difficult schedule, but you can see how we've performed relative to everyone else.

We're the third worst offensive team in the league. While last year we were a solid top 5 defensive team, this year so far we're only 18th in limiting opponents points per possession.

Our shooting has been pretty bad, but we're suprisingly decent at 3 pointers and free throws, relative to everyone else. We've really picked up our 3-point shooting over the last few games.

While we're a below average offensive rebounding and defensive rebounding ball club, strangely the numbers show we're 15th in total rebounds. Maybe "team rebounds" accounts for this seeming anamoly.

We're the slowest team in the league, along with Orlando (both 85 possessions a game, compared to Phoenix which outpaces everyone else at 97 possessions per game).

Finally, our point different (adjusted for pace) puts us at 25 out of 30. Basically, that means if you wanted to rank all the team's performance so far, not accounting for difficulty in schedule, we're number 25 out of 30. Pretty sucky.

 

Great breakdown of such stats durvasa.

The Rockets are a team known to run down the shot clock on almost every possession, but I think that the fact we're last in possessions per game is largely because of the lack of McGrady. Without him, we no longer have a guy who will create for himself, instead we're seeing many more picks set, many swings around the perimeter to find the best entry point for the lob to Yao, and also a lot more dribbling around waiting for plays to develop. Is it our fault? I'm sure it is but I'm also sure that we'll pick it up when we get McGrady back.

 

Dean Oliver, a stats guru who currently works with the Supersonics and author of "Basketball on Paper", writes that there are four significant factors for winning basketball, in order of priority:

(1) Shooting % from the field - (fgm + .5*3pm) / fga
(2) Turnovers per possession - (turnovers) / possession
(3) Offensive rebounding % - (off. reb) / (off. reb + opponent's def. reb)
(4) Getting to the foul line - (free throws made) / fga

Winning basketball means doing better than your opponent in these areas.

I took the raw data for each team's last 10 games (which you can find at www.dougstats.com) and determined how the Rockets faired according to these factors. The only team who did better in each category than their opponent over their past 10 games is the Cleveland Caveliers. Not surprisingly, they're also leading the league in points/possession differential -- the best rating for team success. As a comparison, the Cavs are scoring 10.75 points per 100 possessions more than their opponents in theri past 10 games, while the Rockets are scoring 6.25 points per 100 possessions less than their opponents over their last 10 games (26th in the league).

This is how the Rockets rate according to these factors. The ranking in parenthesis is our rankings for the second half of last season when we were one of the top teams in the league, for comparison.

<font face=""Courier New"">Shooting % differential: 21/30 (3/30)
Turnovers/possession differential: 28/30 (12/30)
Offensive rebounding % diffential: 17/30 (5/30)
Getting/making free throws differential: 6/30 (8/30)</font>

Oliver estimates relative weights of 10 for (1), 5-6 for (2), 4-5 for (3), and 2-3 for (4). So, relatively speaking, (1) is twice as important than (2), slight more than twice as important than (3), and 3-5 times as important as (4).

Clearly, our major problem areas in the last 10 games were shooting % differential and turnover differential. Here's a closer look at these areas, where ranking 1/30 is best and 30/30 is worst. Again, the second rankings is for the second half of last season, for comparison.
<font face=""Courier New"">
our effective fg%: 45.35%, 24/30 (49.74%, 8/30)
opponent's effective fg%: 47.10%, 9/30 (46.17%, 3/10)
turnover %: 16.83%, 23/30 (13.11%, 5/30)
opponent's turnover %: 12.99%, 29/30 (13.72%, 21/30)</font>

We've been shooting very poorly relative to the rest of the league, though we're decent at limiting the opponent's shooting. Not a surprise. We've been a high turnover team, and at the same time have been second worst over the last 10 games at forcing turnovers.

We'll never be a team that forces a lot of turnovers, but we need to minimize the differential by taking better care of the ball. Most importantly, we need to find ways to improve our shooting %. We've been ok from 3-point range (10th in the league over last 10 games). The big problem for us has been shooting inside the arc. We're 27th in shooting % on 2-point field goals. The best way to fix this is pushing the ball in transition to get shots near the basket before the defense sets, getting penetration from our guards, and getting Yao, Howard, and Swift to shoot better from inside.

 

We have let opponents shoot almost 60 more shots than us this season...meaning we're allowing way too many second-chance opportunities...

 

Nice stats. The offense is definetly loosing us games. We havent even hit 100points yet. The reason for that is JVG.

 

Swift will probably get a chance to prove himself in the upcoming games. He's having his worst rebounding year of his career. In particular, his defensive rebounding has been horrible. Let's compare defensive rebounding % by our key front-court players:

<font face=""Courier New"">Yao: 19.0% (8.5% off)
Mutombo: 21.1% (10.1% off)
Howard: 14.6% (7.7% off)
Swift: 12.7% (7.7% off)
McGrady: 16.2% (3.0% off)
Bowen: 8.4% (6.3% off)</font>

Bowen is usually busy chasing his defensive assignment, and he makes up for his lack of rebounding on defense by doing a good job on the offensive boards (for a perimeter player).

But look at Swift's defensive rebounding rate. While Swift is at least able to match Howard on the offensive glass, his defensive board work is really weak. And it's not just that he's playing along-side better rebounders than Howard (and it's not very clear that's the case). There's a significant decline in team defensive rebounding with Swift on the court versus off the court -- 67.8% versus 73.0%.

I think he's still confused by his role on defense, and that's throwing off his timing on the defensive boards. Actually, 82games.com shows that while he's playing PF, he's rebounding at only 8.2 boards per 48 minutes, but when he's playing center his rebounding rate increases dramatically to 13.9 boards per 48 minutes. Unfortunately, his offense goes south when he's playing center and the opposing centers usually go off as well.

So, what can we expect from Swift in upcoming games? With Yao and Mutombo injured, he'll get a lot of playing time at center. Based on the numbers so far this season, his rebounding will probably go up. But as a team, I expect we'll struggle on the boards and stopping points in the paint. Fortunately, we aren't facing any great teams any time soon so maybe we can weather the storm.

 

Well, Swift will definitely come in handing when defending Bosh against Toronto Wednesday night. He is the only Rockets power forward with the quickness to come close to stopping Bosh. As for his rebounding, I find that yes, he does tend to disappear sometimes, but the problem is existent in Howard as well.

 

Well, it's been about a month without Yao, and 82games.com has posted some +/- stats for every NBA player for the last 30 days. I thought it would be interesting to take a look at which players the Rockets have played well as a team with, and which ones we've played poorly with. Not surprisingly, virtually everyone with significant minutes has not had much success.

http://www.82games.com/0506/0506HOUR.HTM

It shows some interesting things. The team has played relatively well with Alston, Norris, and Lucas at PG. Without McGrady and Yao, it's important to have a PG that can run an offense semi-effectively. The Wesley/Head combination should probably go.

We've really struggled with Swift and Bowen. Our defense has been horrible when Swift has played, and with Bowen in there our offense somehow gets worse.

JVG, take a hint. In 14 games, the Rockets have been outscored 13 times with Bowen in the game. The dude has the lowest PER in the entire league. The experiment is over. Whatever benefit might be had from him on the defensive end are nullified by his ridiculously inert offense.

Swift had a couple nice offensive games before his "pink eye," but he's basically a Ryan Bowen on the defensive end. What happened to the shot blocking, or at least some intimidation? I really expected him to step up when Yao went down and provide some positive contributions. He has rebounded a little better, but he isn't close to meeting expectations.

After a very impressive start to the season, Head has slid back considerably as he's had to take on a greater share of the offensive load. I do think Wesley has played better as of late, making an effort to penetrate more since McGrady went down.

I've been down on Howard for most of the season, but I do have to acknowledge that he's stepped up considerably the past few weeks. Unfortunately, at his best he's still only slightly above average (consider the number of shots and minutes he's taken to get those 20-point games) and we aren't going to win many games with him as the focal point of our offense. But he's playing hard, and that's all you can ask.

 

What's with the Rockets and these slow starts? This seems to have been a problem ever since Van Gundy got here.

We have yet to score 30 points in any first quarter this season. In 16 home games, we've only scored at least 25 points in the first quarter two times, and have scored less than 20 points nine times.

In the 16 home games, we've only averaged 18.8 points in the first quarter. In 21 road games, we've done a little better -- 20.9 points.

<font face=""Courier New""><font size="3">Slow starts</font>

Home (first quarter)
# games: 16
< 20 pts: 9 (13, 13, 13, 14, 16, 18, 18, 18, 19)
20-24 pts: 5 (20, 20, 22, 22, 23)
>24 pts: 2 (26, 26)
average: 18.8 pts

Road (first quarter)
# games: 21
< 20 pts: 7 (12, 12, 14, 15, 17, 19, 19)
20-24 pts: 10 (20, 20, 20, 20, 21, 23, 24, 24, 24, 24)
>24 pts: 4 (27, 27, 27, 29)
average: 20.9 pts</font>

 

The Rockets spend too much energy trying to establish the tempo and control the game defensively in the first quarter, that's my reasoning behind our slow starts. Also, TMac doesn't really like to shoot a lot in the first quarter, and Yao has rarely started first quarters well this season either. It's not much of a surprise that we're not scoring much this year, but we were expected to at the start of the season. It was predicted that we would be more of a running team with Swift, Anderson and Alston, but the fact that all three, and TMac, haven't been healthy spoiled that.