33 Responses to “Losing Your (Inefficient) Leading Scorer Hurts Your Team”

1. Nathan Walker Says:
   May 23rd, 2011 at 11:35 am

   I wonder if David Berri thinks that the Bulls were better with Kerr than Jordan. EFFICIENCY Lulz.

2. David Says:
   May 23rd, 2011 at 11:38 am

   Hi Neil, excellent to see these tweaks added. 3 things caught my eye.

   The 1.2 value is statistically distinct from zero. 95% CI is 1.2 +/- 2 * 0.3852 using bootstrap.

   Not sure if this is meant to debunk the Berri stuff, which replaced the scorer with an average player. I realize that is a philosophical issue too, no need to revist. My question here is if there is anyway to do something similar (replace chucker with avg player) here using, say, WinShares? Another way to put this: is there some link between where a team is in your Diff column and the player that got the minutes when the scorer was out? You have HCA, SOS (I think) in this mix so this seems like a logical extension. Although I'll admit to not being sure if it would yield anything. But I'm curious if we aren't seeing some strong vs. weak bench thing.

   You defined lead scores relative to the league average. Would not position average be more useful? I'm just not sure a guard should be compared to a center in this regard. I guess I was surprised by some of the names there, like James in 2004 or Wade in 2008.

3. mystic Says:
   May 23rd, 2011 at 12:05 pm

   Great job, Neil. You just made my unfinished work about those "inefficient" scorers useless. I identified 57 cases since 2002-03 which I wanted to look into not only by using normal boxscore stats, but also by the +/- numbers provided by 82games.com. Adding the HCA term is really important here, I also added a term for SOS by using the average SRS values for the opponents to make the adjustment for different strong opponents during the stretch with and without the scorer. Overall I come to a similar conclusion as you (granted, I only collected the data for 36 cases so far). The interesting thing is that while the correlation coefficient to the expected point differential (using the Net+/- numbers for the players in games they played to calculate that) is at around 0.7, the teams are playing like 2pts better than expected, but they have still close to 1.5 pts worse point differential than the expected value with the scorer. When I add the more efficient scorers to the mix, the value goes up to 2.5 so far.

   So, the overall conclusion is the same. Even replacing the inefficient scorer isn't that easily achievable.

   David, we have to keep in mind that the scoring load is not replaced by the 7th or 8th man on that roster who gets more playing time due to the loss of that main scorer. The 2nd option will become the 1st option, the 3rd option the 2nd option and so on. In average the role of the main scorer will not be replaced by a well below average guy, but most times by a guy who should be an above average scorer/player anyway.

4. AHL Says:
   May 23rd, 2011 at 12:24 pm

   At the very least, this shows why Denver was so much better off without Melo, and Memphis without Rudy Gay.

5. ElGee Says:
   May 23rd, 2011 at 1:32 pm

   My comment turned into a supplementary post: http://www.backpicks.com/2011/05/23/more-on-low-efficiency-scorers-on-bad-offensive-teams/

   The short of it: I look at that data and see that
   (1) a lot of "inefficient" scorers played on bad offensive teams outside of the player in question
   (2) Those bad teams were helped by their "inefficient" scoring

6. mystic Says:
   May 23rd, 2011 at 1:55 pm

   Elgee, going by your data would you agree that inefficient scorers tend to shoot more when the team surrounding them is worse offensively? While the higher usage in the end would also drag down their scoring efficiency further more?

7. Jason J Says:
   May 23rd, 2011 at 2:02 pm

   It's interesting to see some of these names. I would not have expected Patrick Ewing to come up as an inefficient scorer so often. Also interesting to see which teams dropped off the face of the earth without his defensive presence and which ones survived.

   Two concerns come to mind on the results:

   1) Some of these guys were also their teams' best all-around players - like Ewing or McGrady or Wizards Jordan for example - so you lose the inefficient offense, but you also lose significant playmaking, rebounding, defense, and leadership (on court coordination of the team). I'm not sure if that overall 1.2 is really telling us about the importance of having inefficient high usage players or if it isn't just picking up on the fact that many of these guys bring more to the table than chucking (they'd almost have to).

   2) The usefulness of the findings might benefit some if the possessions played without the high usage player had a 200+ possession minimum or something in place to account for randomness. If a player misses one or two games worth of possessions, is the team's production over that span really telling us anything of note? Look at 2008 Rudy Gay. He missed one game that the Griz happened to play against the hapless Sonics before Durant figured out how to be Durant in this league. As a team they shot 13-16 from 3, and Bobby Jackson had a phenomenal game. Probably not a repeatable success story.

   http://www.basketball-reference.com/boxscores/200801180MEM.html

8. DSMok1 Says:
   May 23rd, 2011 at 2:16 pm

   Wonderful research. Keep it up!

9. Neil Paine Says:
   May 23rd, 2011 at 2:23 pm

   #7 - 2) is addressed by weighting the differences by possessions without the scorer (so a case like Gay's makes basically no impact on the overall average). 1) is an interesting point. I wonder how we could measure all-around play to eliminate situations like you mentioned, since efficiency is so intertwined with a metric like win shares or even PER. Perhaps look at non-scoring SPM?

10. Jason J Says:
    May 23rd, 2011 at 2:35 pm

    #9 - Gotcha.

    How about factoring in versatility index?

11. DSMok1 Says:
    May 23rd, 2011 at 2:38 pm

    I just put up a post with a chart visualizing this data. The slopes of the lines are interesting:

    http://godismyjudgeok.com/DStats/2011/nba-stats/chart-with-or-without-inefficient-scorers/

12. ElGee Says:
    May 23rd, 2011 at 2:40 pm

    #6 - Mystic, I would absolutely agree.

    If the team can't apply on pressure on a defense, the "best" option is left with harder attempts, and those attempts are sill better than whatever the other guys could muster up without him. The usage is going up, driving efficiency is going down. Even though they're "inefficient" a lot of times they are helping the offense a good deal. (And they might not be inefficient if they were on a decent offensive club to being with.)

13. DSMok1 Says:
    May 23rd, 2011 at 3:02 pm

    Okay, I revised and added to the last chart:

    http://godismyjudgeok.com/DStats/2011/nba-stats/chart-with-or-without-inefficient-scorers/

    Basically, If you're on a good offense, losing you hurts, period. If you're on a bad offense, it may not hurt.

    I think that's contrary to what Neil said...

14. Greyberger Says:
    May 23rd, 2011 at 5:40 pm

    Boo versatility index. An idea who's time has come, and went.

15. Michael E Sullivan Says:
    May 23rd, 2011 at 5:47 pm

    I'm not sure I buy your interpretation DSMok.

    Looking at your graph it seems the main thing pulling your offensive line down is that there are a few data points where the offense without the player was terrible, but with the player actually fairly good. This suggests it hurts a lot more. Basically to get at what you are asking, the appropriate question is what kind of spread is there between the best and worst with player data points for a give without player rating. The spread is much bigger on the low end than on the high end.

    The teams that were very good without the high scorer, generally did better with, but not a lot better with and they rarely did a lot worse with -- suggesting that NBA coaches and GMs are not stupid, and do not often let players take a ton of possessions when it would make their team significantly worse. This would be expected.

    So generally inefficient scorers are either on poor teams that don't have a lot of better options, or their high usage lets other players be more picky about when to take shots, and thus they are more efficient with than without the scorer around, meaning that these guys could be somewhat more valuable than raw efficiency suggests. Though, only somewhat, I notice that most of the guys with positive differentials brought some other stuff to the table.

16. Matt, Colombia Says:
    May 23rd, 2011 at 5:57 pm

    I think this shows that WoW simply does not understand usage curves.

17. Bob M. Says:
    May 23rd, 2011 at 10:00 pm

    This confirms that Jermaine O'neal was an offensive black hole for the Pacers in the mid 2000s. Good research.

18. Anon Says:
    May 23rd, 2011 at 10:25 pm

    "Boo versatility index. An idea who's time has come, and went."

    Not a fan?

19. Greyberger Says:
    May 24th, 2011 at 2:12 am

    I might not have a grasp on what the VI is measuring and what it means. The only context I've seen it in is coefficient tables for SPM, where it seems arbitrary and questionable to include it. To me

20. marparker Says:
    May 24th, 2011 at 3:12 am

    Isn't it important to know who took over that player's minutes and what kind of usage/efficiency they were able to produce. Can we really lump in trades with injuries? All sorts of bases were left uncovered. This is a reach. It almost seems disingenuous.

21. Rob P. Says:
    May 24th, 2011 at 3:29 am

    Seeing Worthy on the list for '91 and '92 made me wonder about him. A "high-ankle injury in the '91 playoffs" and a "season-ending knee injury in '92" (wiki) really seemed to alter his game. Interesting how much he trended toward taking shots from deep in his later years:

    In his first 6 seasons (471 games): 59 3PTA and 3 3PTM (.017 3PT%).
    In his final 6 seasons (455 games): 427 3PTA and 114 3PTM (.266 3PT%).

    Side note: I wondered who led L.A. in scoring (PPG) in Worthy's final two seasons:
    Sedale Threatt with 15.1 PPG (1992-1993) and Vlade Divac with 14.2 PPG (93-94)

    Threatt edged out Worthy by 14 points; meaning Worthy came really close to having another lousy, inefficient season appear on this list.

22. huevonkiller Says:
    May 24th, 2011 at 10:44 am

    Wow this is some great research.

23. Jason J Says:
    May 24th, 2011 at 1:19 pm

    I actually think VI is a minor component of PER as well... that may be wrong... I do believe it is a Hollinger stat in any case.

    I suggested it because we're trying to isolate the inefficient scorers who also do a lot of other things, and VI measures other things without any concern for efficiency.

24. Vjl110 Says:
    May 24th, 2011 at 3:08 pm

    The fact that losing a starting player (who happens to be an inefficient scorer) lowers efficiency differential 1.2 points really isn’t interesting by itself. Absolutely no negative change is not an acceptable null hypothesis for this analysis. That said, there are ways you could get at answering the proposed question using this data-set.

    1)
    Try randomly selecting 100 starting players and see how losing those players affects efficiency differential. Now you have a control group which gives you a null value to compare against. If losing a random player is no different in effect than losing an inefficient scorer, that inefficient scoring isn’t necessarily important. If losing inefficient scorers hurts the team more than losing a random starter, you now have much better evidence for the import of inefficient scoring.

    2)
    Alternatively, build a regression model using some composite metric like WP48/WS/PER for each player, and another using shots/usage or some combination thereof. Then try using these two models to account for the team efficiency differentials that you found. The results of this study wouldn't be as clean as the one above, but they would still be interesting. If usage or shot attempts does not explain the strength of effect on efficiency differential within the above group of inefficient volume scorers, then I am not sure the study supports the value of inefficient scoring.

    This is a great study, but I don't feel like I have learned much until you at least find a better null value than "no effect of losing starting player".

25. yariv Says:
    May 24th, 2011 at 9:50 pm

    As #24 noted, and as discussed in the original post, the problem is comparing losing the leading scorer to losing any high MPG player (I would prefer "high MPG" over starter). Any intention to work on this?

26. Owen Says:
    May 25th, 2011 at 10:53 am

    "I think this shows that WoW simply does not understand usage curves."

    Not that this will get answered, but I am a WOW type, and I have to admit, I don't get usage curves. The premise is obvious. The more you shoot the less efficient you should be. The less you shoot, the more efficient you should be.

    But to me it rarely works out as predicted. For instance, shouldn't Bosh have seen a huge bump in efficiency this year playing next to two stars? He was more efficient from the perspective of turnovers but his scoring efficiency actually went down to it's level of two years ago.

    You look at Kobe's numbers and the pattern is counterintuitive as well. He basically has been in a ts% band of 54.4% to 56.4% for most of his career (all but two years) despite changes in usage, shot attempts, teammates etc. His highest usage year (in his age 27 season granted) wasn't basically right where it should have been at 59.9%. If anything though his efficiency seemed to increase with higher usage (although adjusting for age probably would flatten things out).

    Clearly there is some relationship between usage and efficiency, but it strikes me as being very weak.

27. Owen Says:
    May 25th, 2011 at 10:55 am

    "wasn't basically right where it should have been at 59.9%."

    should read

    "was basically right where it should have been at 55.9%."

28. David Says:
    May 25th, 2011 at 11:24 am

    #26 - Neil did a recent post on usage curves. Well, not exactly, but he looked at how per minute stats change when minutes change. This, to me, is the heart of the matter. There really was no conclusive take home message. You can find players who get better or worse and also those that do not meaningful change. I've seen this elsewhere too. I also wonder where this conventional wisdom stuff come from. I know Hollinger likes this idea. But there is simply too much scatter for this idea to, say, be a parameter for player acquisition by the FO. Maybe I've missed _the_ study on this too.

    #24 - Yes, this is huge. A meaningful null should not be "no change". And I would still love to see the following: Replace every team's leading scorer by an average player using WinShares and simulate the current season. What does eff diff look like then? The conditional analysis Neil did certainly points in the right direction but there is too much variability in terms of how a team makes up for the missing chucker.

29. Vjl110 Says:
    May 25th, 2011 at 1:04 pm

    #28 - I think using Winshares or anything else to predict the affect of an "average" player would simply replicate Berri's method.
    The virtue of using a group of randomly selected high-mpg/starting players as a control. You don't need to make any assumptions about what features contribute to wins. This would be a really easy addition the Neil's study, and would give us a really good look at the value of "inefficient" scoring.

30. Anon Says:
    May 25th, 2011 at 1:05 pm

    "Not that this will get answered, but I am a WOW type, and I have to admit, I don't get usage curves. The premise is obvious. The more you shoot the less efficient you should be. The less you shoot, the more efficient you should be."

    This is based on some key assumptions, however (like running/playing an offense that maximizes your play, etc.). Not to mention that the James/Wade/Bosh trio is something that is unique in NBA history.

31. Bill Says:
    May 26th, 2011 at 5:42 pm

    How James/Wade/Bosh came together is unique, but three star players on one team is certainly not unique in NBA history. Did Larry Bird ever play on a team with fewer than three HOFers?

32. David Says:
    May 27th, 2011 at 5:33 pm

    #29 Agree. And I'd love to see this for above and below average players for PPG, REB, and AST. Or maybe by terciles. I think a larger context would help.

33. Nathan Walker Says:
    May 28th, 2011 at 5:36 pm

    Usage curves explain that a player's PERSONAL efficiency decreases when usage increases. Whether or not his TEAM's efficiency increases depends on the player, but we can predict it with their usage. Eli Witus found that lineups containing players with high-usage will fare better than lineups containing players with low-usage.

    The general trend I have found is that (Dean Oliver) efficiency * usage is the most significant descriptor (rather than separating the two) and produces the best predictions.