1. Welcome! Please take a few seconds to create your free account to post threads, make some friends, remove a few ads while surfing and much more. ClutchFans has been bringing fans together to talk Houston Sports since 1996. Join us!

What's Changed & What Hasn't? A Brief Statistical Look

Discussion in 'Houston Rockets: Game Action & Roster Moves' started by Tango, Nov 10, 2007.

Tags:
Page 3 of 4
  1. Seven

    Seven Member

    Joined:
    Sep 18, 2006
    Messages:
    2,251
    Likes Received:
    28
    Like him or not, there should be no doubt that JVG knows how to coach defense. If the players admit that they are using the same principles as last year, then how can you not give JVG credit?
     
    #41 Seven, Nov 10, 2007
  2. Carl Herrera

    Carl Herrera Member

    Joined:
    Feb 16, 2007
    Messages:
    45,153
    Likes Received:
    21,575
    1 Contested 2 pt jumper

    Gang Rebound

    These were both written on the wall of the practice facility during preseason last year. These were what Van Gundy wanted on D, and these were largely what the Rockets did and do right now.

    The explanations are:

    1. The Rockets don't want guys to get layups/dunks, or 3 pointers, these are the higher expected value shots. Instead, they want to give the opponent the expected value shot by mathematical probability: A contested 2 point jumper. They will (1) pack the lane to prevent penetration but (2) make the right rotations to chase 3 pt shooters off the line, but they'll concede the 2 point jumper, preferably with Shane Battier's hand in your face.

    This is still what then tend to do. Watch the tapes, or look at the numbers, you'll see.

    They also focus on eliminating offensive rebounds. They give up some, but statistically, they were and still are among the better teams at it.

    2. Related to that, the Rockets gang rebound. The Rockets want guys at the 1, 2, 3 positions to help the bigs out on the defensive boards boards.

    The current Rockets are more aggressive at having guys going off on breaks, but still, they keep up their defensive rebounding duties. The smalls pitch in on the boarrds.
     
    #42 Carl Herrera, Nov 10, 2007
    Last edited: Nov 10, 2007
  3. EGYPT

    EGYPT Member

    Joined:
    Nov 16, 2006
    Messages:
    1,308
    Likes Received:
    3
    the points are coming much easier this year, then bench is scoring in bunches vs. last year's bench
     
    #43 EGYPT, Nov 10, 2007
  4. ico4498

    ico4498 Member

    Joined:
    Feb 15, 1999
    Messages:
    3,764
    Likes Received:
    1,510
    great difference, great thread, thanks!
     
    #44 ico4498, Nov 10, 2007
  5. WildSweet&Cool

    WildSweet&Cool Member

    Joined:
    Jun 22, 2007
    Messages:
    1,768
    Likes Received:
    0
    One way in which the numbers may be deceiving: We've played 5 very-tough teams already this year, and last year we played only two.

    We're putting up the same numbers against harder opponents. That's awesome.
     
    #45 WildSweet&Cool, Nov 10, 2007
  6. haven

    haven Member

    Joined:
    Oct 22, 1999
    Messages:
    7,945
    Likes Received:
    14
    I love statistical analysis, but we're still at the point where individual opponents are going to influence the numbers too much. The Rockets have played the Spurs, Mavs, and Utah - 3 of the 7 best teams in the league. The Lakers are probably slightly-above-average.
     
    #46 haven, Nov 10, 2007
  7. Tango

    Tango Member

    Joined:
    Nov 7, 2002
    Messages:
    716
    Likes Received:
    12
    WildSweet&Cool: actually the comparison is apples and oranges from a # of games standpoint. I'm comparing the first 6 games to all 82 games last year. However, let me address that below :).

    For everyone who still thinks that 6 games in may be too early to assess, here is some more food for thought for everyone besides what I've posted on sample size and skewdness of the data already.

    For those who remember, I did the same thing last year and thought I would do it again this year for grins. Here's the thread I did similar to this one at 5 games into the season.

    http://bbs.clutchfans.net/showthread.php?t=120015

    Some folks had the similar objections about the number of games and having enough data to assess the team. Yes things can change and indeed they will but I'll let you chew on just a simple look back to looking at the stats 5 games in last year.

    (1) 2006 ORtg/DRtg 5 games in vs. 82 games
    5 games: ORtg 109.9 DRtg 104.3
    82 games: ORtg 108.3 DRtg 101.6

    (2) 2006 W/L Projection 5 games in vs. 82 games vs. Actual
    5 games: 55/27
    82 games: 55/27
    actual: 52/30

    I didn't do anything around other stats like rebounding etc. last year so I don't have anything to compare to. Some of the objections last year to the 5 game assessment were 1)the early Mavs game skewing the data and 2)not big enough of a sample size (sound familiar?).

    However as you can see the ORtg and DRtg actually wasn't that far off, nor were the W/L records using the pythagorean method for projection either. One of the reasons for this is around the sample size of the data. I mentioned the 250-800 minutes sample size. ORtg and DRtg is a per-possession stat. At an average of 90 possessions per game for the Rockets thus far that translates to about 540 possessions in 6 games that we are actually looking at.

    Anyway, yes it's early but before dismissing it so quickly you might want to at least consider some of the above! :)
     
    #47 Tango, Nov 10, 2007
    Last edited: Nov 10, 2007
  8. DaDakota

    DaDakota Balance wins
    Supporting Member

    Joined:
    Mar 14, 1999
    Messages:
    129,049
    Likes Received:
    39,522
    Lies, damned lies, and statistics.

    DD
     
    #48 DaDakota, Nov 10, 2007
  9. m_cable

    m_cable Member

    Joined:
    Dec 12, 2002
    Messages:
    9,455
    Likes Received:
    73
    Yes, but the difference is that we weren't working in a new offense last year. The projection was a lot more likely to hold since we were running the same scheme for a couple years.
     
    #49 m_cable, Nov 10, 2007
    Last edited: Nov 11, 2007
  10. onethreeeleven

    onethreeeleven Member

    Joined:
    Jul 24, 2007
    Messages:
    180
    Likes Received:
    0
    Way too small of a sample size.

    It has nothing to do with Adelman's offense not "gelling" or whatever, just that over six games, luck plays too large of a role for any kind of statistical analysis to be valid.
     
    #50 onethreeeleven, Nov 11, 2007
  11. onethreeeleven

    onethreeeleven Member

    Joined:
    Jul 24, 2007
    Messages:
    180
    Likes Received:
    0
    Even more egregious (in my mind) is that you're justifying this 6 game analysis with only one example from last year for comparison. That is, you're justifying a small sample size with a statistical analysis using ANOTHER small sample size (last year = 1 sample!).

    Sorry, its an impressive feat that 5 games last year predicted the season, but it hardly is a trend.
     
    #51 onethreeeleven, Nov 11, 2007
  12. Tango

    Tango Member

    Joined:
    Nov 7, 2002
    Messages:
    716
    Likes Received:
    12
    Luck is embedded in statistical analysis ;).

    And if you think sample size is too small, that's fine I have no problem with that. However give some evidence or proof as to why you think that sample size is too small. I've thrown mine out there. 6 games ~540 possessions. Is that too small of a sample size? Demonstrate how you think it is.

    Could it all be hogwash? Absolutely! But before you dismiss it so quickly one might just consider some of the rationale and evidence that might make you think otherwise.
     
    #52 Tango, Nov 11, 2007
  13. DwangBoy

    DwangBoy Member

    Joined:
    May 6, 2003
    Messages:
    2,136
    Likes Received:
    272
    Our offensive efficiency would go up by 150% with a better point guard.....
    if we had someone like mo williams or calderon... it'd be over.
     
    #53 DwangBoy, Nov 11, 2007
  14. hooroo

    hooroo Member

    Joined:
    Oct 16, 2003
    Messages:
    19,298
    Likes Received:
    1,917
    [​IMG]
     
    #54 hooroo, Nov 11, 2007
  15. DonkeyMagic

    DonkeyMagic Member
    Supporting Member

    Joined:
    May 22, 2006
    Messages:
    21,604
    Likes Received:
    3,487
    fire JVG...oh wait. nevermind :D
     
    #55 DonkeyMagic, Nov 11, 2007
  16. onethreeeleven

    onethreeeleven Member

    Joined:
    Jul 24, 2007
    Messages:
    180
    Likes Received:
    0
    Well, 540 possessions against only six teams doesn't constitute a fair sampling of all NBA teams, but its true what you say, that your numbers have merit, if only for speculative purposes.
     
    #56 onethreeeleven, Nov 11, 2007
  17. Carl Herrera

    Carl Herrera Member

    Joined:
    Feb 16, 2007
    Messages:
    45,153
    Likes Received:
    21,575
    Well, one example is this:

    http://www.basketball-reference.com/teams/ATL/2007_games.html

    I'd bet this team's stats in their first 6 games did not fall in line with that of their season.

    Or

    http://www.basketball-reference.com/teams/HOU/2005_games.html

    How about this teams first 21 games?

    Yeah, I know... the 04-05 Rockets had injury (Sura at the start of the year) and personnel change (subtracting Boki, Jim Jackson, Lue, adding Wesley, Barry, Mike James) accounting for much of the difference, but who is to say the Rockets won't lose someone to injury for a while or make a significant trade this year, too?

    I mean, what if we do deal our complaining star and aging African?

    :cool:
     
    #57 Carl Herrera, Nov 11, 2007
  18. GATER

    GATER Member

    Joined:
    Jun 25, 2000
    Messages:
    8,325
    Likes Received:
    78
    All the stats are very nice. Bottomline for me is in transition. This years' team is not afraid they'll get the hook or be chewed out for missing a quick shot in transition.

    My "stats"...Easier transition points + Yao not spending an entire game wrestling in the low block = A more productive Yao in the 4th quarter
     
    #58 GATER, Nov 11, 2007
  19. DaDakota

    DaDakota Balance wins
    Supporting Member

    Joined:
    Mar 14, 1999
    Messages:
    129,049
    Likes Received:
    39,522
    DING DING DING......

    We have a winnner !!!!!

    Not to mention the ability to attack the offensive glass.

    DD
     
    #59 DaDakota, Nov 11, 2007
  20. Angkor Wat

    Angkor Wat Member

    Joined:
    Jun 14, 2007
    Messages:
    13,150
    Likes Received:
    997
    WOW, look how our starting PG is all the way at the bottom. That was V-Span's spot last year.
     
    #60 Angkor Wat, Nov 11, 2007
(You must log in or sign up to post here.)
Page 3 of 4

Share This Page