DFDepressed FanDepressed Fan



, all the time

Game 22 Stats: Reverse Boxscore and PER

Well, I was all set to do the full Differential Production stats for this game, but then the final play happened and I lost my enthusiasm for it (kind of like Brian).  If there's one saving grace (for those who want to see the team do well this year, that is), it's that many of these same players were around in 08-09, when they proved to be resilient in the face of several stomach-punch games (besides the Ray Allen buzzer-beater and the Devin Harris buzzer-beater, there were also buzzer-beaters by Nowitzki and Tony Parker, for example).

Anyhow, I did watch the game again and compiled the reverse boxscore and PER numbers for the Sixers. As a reminder, the reverse boxscore attempts to capture the (conventional) stats accumulated by whoever was opposing a given player (guarding or being guarded by that player).

Here is the reverse boxscore for the game.

AI 40 5-9 2-3 4-4 0-5 3 1-4 0 4 2 16
EB 39 6-13 0-0 1-2 0-5 1 4 1 2 0 13
SH 28 4-7 0-0 6-6 1-6 1 1-3 0 2 1 14
JH 34 5-10 0-0 1-2 0-2 7 3 3 2 1 11
JM 40 6-9 1-2 3-3 1-4 4 1 0 0 0 16
TY 28 7-11 1-2 2-2 3-8 3 1-3 1 4 1 17
LW 22 4-8 3-5 0-0 0-3 3 2 3 0 1 11
ET 6 1-1 0-0 2-2 0-0 0 1 0 1 0 4
AN 0 0-0 0-0 0-0 0-0 0 0 0 0 0 0
Tot 240 38-68 7-12 19-21 5-33 22 3-21 8 15 6 102

A few notes on the boxscore (I'll have full comments below the PER numbers):
  • In the official boxscore, there is an assist credited to Garnett in the first quarter that is wrong.  I watched the play several times and he never touched the ball.  If that assist gets counted, then Brand's opposing line has one more assist.  Also, this doesn't show up in the reverse boxscore (but will in the PER numbers below), but I credited a block to Iguodala that doesn't show up in the official boxscore (his clear block on Rondo in the 1st quarter).  If that block doesn't get counted, his offensive PER goes down to 20.0.
  • I added "OF" for offensive fouls (drawn), and "PF" now means total personal fouls drawn (offensive and defensive).
  • In re-watching this game, I noticed that the Celtics either utilize picks more than other teams or set more effective picks than other teams.  This is the fifth game I've re-watched closely this season, and it was by far the one where the Sixers were forced to switch the most.  A lot of that was because of Rondo, who penetrates effectively with even the slightest opening.  As far as the reverse boxscore, all the switching made the scoring more subjective than usual, because it wasn't always clear who had most responsibility for a shot.

Now for the PER numbers:

AI 21.0 10.9
EB 21.1 6.3
SH 13.3 22.6
JH 5.3 17.2
JM 12.1 19.1
TY 21.1 24.0
LW 12.3 23.2
ET 5.2 18.4
AN -11.3 0.0
Tot 14.3 16.1

Comments upon re-watching the game:
  • The PER numbers support Brian's pick of Brand as player of the game, but I would actually go against the stats and pick Iguodala.  Iguodala's passing was superb all game (11 assists to 1 turnover, and the turnover was not a passing turnover -- it was a 3-second call that came after a nice pass to Hawes, who didn't shoot), and he assisted on all 4 of Meeks' 3-pointers (and 6 in all, for an impressive 28 points set up in addition to his own 14).  Iguodala also played good defense on Pierce the whole game; Pierce's one basket against him was a wild penetration in traffic.  (Where Iguodala wasn't as good on defense was on switches, accounting for 3 of his 5 baskets allowed, all in the 2nd quarter.)  And down the stretch, Iguodala not only scored three of his own baskets but directly set up two others (penetration and kick to Brand for an open jumper, great pass to Thad for a lay-up).
  • Brand was clearly another big plus for the Sixers, and I don't want to minimize his contributions.  As I've said before, the rehabilitation of Brand is Collins' best work this season (Thad's resurgence being a close second).
  • On re-watching the game, I found that Holiday's defense was not as poor as Rondo's line implied.  Jrue made several subtle plays to help on defense, and one wonders if the 9-0 run at the end of the 3rd might not have happened had Jrue been in there.  One note, however: Jrue seems to have gotten lazy fighting through picks (this afflicts all the Sixers to some degree).  Often he'll get hung up even if the pick isn't "well set," expecting help which doesn't always come.  Against someone like Rondo, P/R defense is paramount.  In this game, though, it was Jrue's offense that was worse than his defense.
  • Speaking of the 9-0 run, on re-watching it I can't complain about the offense they got during that stretch.  It consisted of an aborted penetration by Thad and four open jumpers.  This year especially, I'll take a penetration by Thad anytime, and I don't mind an open 3 by Nocioni (44% this year) or even open mid-range jumpers by Hawes and Lou.
  • However many times I calculate opponent PER this year, I think I'm going to see similar numbers from Meeks and Thad, especially now that both have played well enough to be playing against opposing starters:  good offensive PER, bad defensive PER.  Meeks tried hard but rarely challenged Ray Allen (or Rondo for stretches).  And Thad was in one of his "no contact" post defense modes. Both of them, though, have generally been bringing enough on the offensive end to offset their defensive deficiencies.
  • Regarding the fateful play at the end, I think it's clear that Rondo would not have been able to penetrate baseline had Jrue gone under the pick, because there was a third defender in the area (Meeks in the corner).  But I'm almost certain Thad would have helped and Garnett would have gotten an open jumper (which of course would have been preferable to an open lay-up).  In the end, it was truly a "pick your poison" type of play.
  • Finally, for those who want to torture themselves more, take a look at the inbounds play after Garnett's basket.  It seems to me that if Jrue hadn't thrown it directly to Garnett (it looks like the ball slipped from Jrue's hands as he threw it), Jrue had Thad streaking behind everyone for a possible winning basket.  But it wasn't to be ...