This is our old blog. It hasn't been active since 2011. Please see the link above for our current blog or click the logo above to see all of the great data and content on this site.

PI Finds: Kidd Tidbits

Posted by Neil Paine on November 1, 2010

BBR reader Jim Keller used our Play Index Game Finder to do some interesting research on Jason Kidd's performance vs. Charlotte last Wednesday, so I thought I'd post his findings. Take it away, Jim (what follows in blockquotes is his work):

Kidd talked about the game here:

Plus/Minus traditionally loves it some Jason Kidd, and it's not hard to see why. Even at his advanced age, Kidd does just about everything you could ask of a guard at both ends (except stay with smaller, quicker PGs)... He distributes as well as anybody not named "Nash", "Paul", "Williams" or "Rondo"; he continues to rebound the ball at a high level; he's suddenly a deadly outside shooter; he can defend SGs or big PGs effectively; and he generally makes the Mavs a lot better when he's on the floor. People scoffed when Wayne Winston's +/- system said Kidd was one of the highest-impact players in the game, but I'm not so sure that was such a crazy idea after all.

12 Responses to “PI Finds: Kidd Tidbits”

  1. Gil Meriken Says:

    I've posed this question at Wages of Wins as well: If, by the end of this season, Lebron James' WinShare per 48 minutes drops even 10% from his previous mark around 0.300, would that imply that not only did his productivity drop, but also that he got worse simply by switching teams? Would it be too much to say that because of this drop, he somehow had less skill as a basketball player or was a worse player than he was last season? If not, then why can we do this in comparing players' (from different teams) WinShares per 48 minutes, when there would be an example here of an individual whose Winshare was different merely because his role changed? If he ends up with a .300 WS/48 min, then my point is moot, but I'm just wondering what the implications would be IF his WS/48 ends around .250 for the season.

  2. Neil Paine Says:

    Well, to talk about this we first need to make the distinction between value (backwards-looking) and ability (forward-looking). Read this baseball article:

    http://gosu02.tripod.com/id11.html

    It's definitely possible that LeBron's ability stay the same but his value/production decline -- i.e., he could be less valuable in this new situation due to diminishing returns. In other words, the redundancy of what he brings to the table vs. what Wade brings to the table could make his value in Miami less than it would have been on another team.

    Now, a metric with an ideal usage-efficiency tradeoff would theoretically account for this. The lighter usage burden we assumed LeBron would take on (which, btw, hasn't actually happened so far) would coincide with an appropriate increase in efficiency, and his WS/48 would stay constant, just at a different level of usage and offensive rating. Does Win Shares have that ideal tradeoff, though? I'm not sure. And as Jason J. noted a while ago, the best model may not even be the two-axis usage v. efficiency plot, but rather a system that also accounts for "difficult shots" as a percentage of possessions used. There's still plenty of headway that needs to be made in that area to refine the process.

    Also, while I have a feeling you're talking about offense only, remember that Win Shares depend heavily on team defensive rating to determine an individual's defensive ability and/or value. LeBron could have an identical offensive year to last season and still see his WS/48 decline if Miami is worse defensively than Cleveland was. Would this imply that LeBron has been worse defensively? In a way, yes, but again, perhaps it's saying more about his value -- philosophically speaking, are two equal defensive performances, one on a bad defensive team, one on a good defensive team, equally valuable?

    What exactly does this have to do with Jason Kidd's performance last Wednesday, though?

  3. Gil Meriken Says:

    2. What does this had to do with Jason Kidd's performance? Everything.

    Just kidding. Nothing. Totally off topic, and thanks for the response.

  4. Neil Paine Says:

    OK, just making sure there wasn't an obvious Jason Kidd angle I was missing. :)

    In fact, oddly enough, Kidd's WS/48 after going to Dallas has been basically what you would have expected from his last few years in New Jersey, despite playing a pretty different role in the offense.

  5. Gil Meriken Says:

    That is an interesting case of switching teams and keeping similar WS/48. But I wouldn't say that Kidd's role is all that different in Dallas, unless you mean the magnitude of his role - obviously he is not the main star, but his skill set as point guard that he utilized in New Jersey is the same one which the Mavs rely on. He also came into a situation where he was not sharing time with another player with a similar skill set/role, as the Mavs traded away Devin Harris. In contrast, Lebron will be playing with Wade, creating some duplication of skill set/role. Whether this matters or not to overall production is my big question, and that's why I will be following this season very closely to see the implications of what a different WS/48 for Lebron or Wade would mean. If they end up close (within 5-10%?) to their historical WS/48, this would all be a moot point, and would seem to indicate that WS/48 does work to measure some intrinsic value of a player, something I'm not yet convinced of, but maybe you already agree with that. As you wrote, it's a measure of a value to a specific team, so that a player could be less valued, but still be the same player he always was.

  6. DSMok1 Says:

    ASPM has Kidd a consistent top-flight performer. Here is his career:

    Year Team Player Age Games Min ASPM VORP
    1995 DAL Jason Kidd 21 79 2668 0.66 2.45
    1996 DAL Jason Kidd 22 81 3034 2.27 4.01
    1997 PHO Jason Kidd 23 33 1173 3.57 1.94
    1997 DAL Jason Kidd 23 22 791 0.73 0.77
    1998 PHO Jason Kidd 24 82 3118 2.18 4.05
    1999 PHO Jason Kidd 25 50 2060 6.05 7.74
    2000 PHO Jason Kidd 26 67 2616 4.01 4.63
    2001 PHO Jason Kidd 27 77 3065 4.43 5.75
    2002 NJN Jason Kidd 28 82 3056 4.61 5.86
    2003 NJN Jason Kidd 29 80 2989 6.63 7.29
    2004 NJN Jason Kidd 30 67 2450 4.85 4.88
    2005 NJN Jason Kidd 31 66 2435 4.45 4.56
    2006 NJN Jason Kidd 32 80 2975 4.71 5.80
    2007 NJN Jason Kidd 33 80 2933 3.94 5.12
    2008 NJN Jason Kidd 34 51 1895 1.54 2.17
    2008 DAL Jason Kidd 34 29 1011 4.42 1.90
    2009 DAL Jason Kidd 35 81 2886 3.23 4.54
    2010 DAL Jason Kidd 36 80 2881 3.08 4.41

    He has been consistently very good. His total VORP puts him 15th since 1977, right there with Kobe, Drexler, and Payton.

    Perhaps the most underrated PG, I think.

  7. Anon Says:

    The usage-efficiency model is definetely simple, but it just...WORKS. Not all truths have to be of the complicated kind and usage-efficiency is certainly an observable phenomenon that is an important part of how basketball works. It isn't always a nice and neat linear relationship though, which is why the +/- regressions is also useful.

    As for LeBron, as Neil noted he actually hasn't taken a step back in the offense so far this season (as some of his always loyal critics have quipped, "Is this the Miami Heat or the Cleveland Cavaliers offense?"). Part of this is picking up the slack for Wade as he was working his way back into playing shape after his hamstring injury, and also he's still figuring out his role in the offense. Notice the spike in turnover rate since he joined the Heat - ORtg doesn't like high turnovers and LeBron is sitting sub-100 as a result. But I think once he gets more accustomed to his teammates (and they become more accustomed to him; except for Wade, Bosh, and Big Z who seems to always be ready for a LeBron pass none of these guys have played with a 6'8 point guard) he'll get back to his usual LeBron-esque performances.

  8. DSMok1 Says:

    Here is a full chart of Jason Kidd's advanced statistics and his ASPM, over the years. I don't think I can imbed it here, so here is a link to it: Jason Kidd Chart

  9. Anon Says:

    Thanks for the info DSMok1. Great stuff.

    If you look at J-Kidd's SPM and WS/48 they seem to match up for the most part: he always gives you above-average to occasionally great seasons, and also gives you consistency and longevity. He actually exemplies the usage-efficiency tradeoff with his career - he goes from being "The Man" in Jersey to complementing Dirk on offense, so what does he do? He continues to set up his teammates but he also finds his sweet spots on the floor. Dirk constanty draws double teams on the high-post/extended wing, and J-Kidd gets open looks on the perimeter and become accustomed to knocking down the jump shot, especially from long-range. Higher ORtgs as a result.

  10. Gil Meriken Says:

    I'm finding the Lebron switch of teams and roles to be all sorts of interesting: today, Lebron finished with zero rebounds for the first time since his rookie season. It's one game, obviously, but you have to concede this isn't just a random occurrence, it's simply a consequence of his role on this new team. So the troll in me wants to state "Lebron is now a worse rebounder than last year!", but I know this just means "Lebron is a less productive rebounder because of the dynamics of his team". Curious to see what all of your takes are on this fascinating experiment of a big superstar changing teams and also changing his role (not in magnitude, but in terms of what the team needs from him). We've seen previous examples like Jason Kidd, and other stars, but for the most part they were expected to do pretty much what they were doing on their old teams.

  11. Neil Paine Says:

    Not saying I disagree with the idea that his production is changing because of his new situation, but I'm also not sure you can say it isn't just a random occurrence. He had 1 reb in a game as recently as Feb. '09:

    http://www.basketball-reference.com/boxscores/200902260HOU.html

    He had a 2-reb game less than a calendar year ago:

    http://www.basketball-reference.com/boxscores/200911180WAS.html

    Sometimes outlier games like these can happen without indicating any underlying skill change. Granted, his Reb % is down from his 2008-10 levels, but that's a 5-game sample -- I think you could probably find a 5-game stretch over the past 3 years where he rebounded just as far below his usual rates -- and even with the decline he's still at his 2006/07 rates.

  12. Gil Meriken Says:

    It's very early in the season, but I'm already setting 'em up! If Lebron ends up at his previous levels, I'll have to skulk away into the recesses of cyberspace.