Post-game Reactions

What are the implications of this?

If you click that link—and you should, so here it is again—it will take you to a knockout piece in which NBA.com’s John Schuhmann breaks down how various teams (and the league as a whole) fare on a quarter-by-quarter basis.

And the centerpiece of his must-read (MUST-READ) piece is this: The Celtics have played poorly in the 2nd halves of games.  And, more specifically, the Celtics’ offense has played very poorly in the second halves of games:

In fact, only two teams, the Nets and Sixers, have more losses than the Celtics do when leading at halftime. Boston has led 52 of their 71 games at the half, but 16 of their 25 losses have come in that situation.

Is it an age thing? Ray Allen, Kevin Garnett, Paul Pierce and Rasheed Wallace all shoot worse in the second half than they do in the first. Wallace’s dropoff, from 47.8 percent in the first half to 33 percent in the second half, is by far the largest of any player in the league who has attempted at least 200 shots in each half this season. Yet, for some reason, he’s attempted more shots in the second half (294) than he has in the first (278). Pierce suffers the next worst dropoff of the four Celtics vets, from 49.9 percent to 44.2 percent.


Sophisticated quarter-by-quarter stats are tough to come by, so Schuhmann’s piece is invaluable. He includes Boston’s quarter-by-quarter efficiency numbers, and the trend is clear: The C’s play declines in the 2nd half, mostly on the offensive end. Here are the quarter-by-quarter offensive efficiency numbers (i.e. points per 100 possessions):

1st: 106.6

2nd: 109.6

3rd: 100.9

4th: 103.2

Only one team in the entire league scores fewer than 100.9 points per 100 possessions, and they play in New Jersey and have won 8 games this season. Only four teams score fewer than 103.2 points per 100 possessions, and none of them will make the playoffs.

Now, context is important, and Schuhmann offers it by explaining the league as a whole scores less efficiently as games go on. (Note: Please go read the whole story if you’re interested. Schuhmann offers some reader-friendly bullet points about most of the key teams in the league as well as league-wide trends).

But the steepness of the C’s in-game offensive decline is problematic.

The most important question is: What, if anything, can the Celtics do about  this? Should the coaching staff tweak the rotation at all? If so, how?

Do you play the starters fewer minutes in the first half? Do you play the bench more in the 3rd quarter? Do you play Sheed fewer minutes early if you plan to use him late? The possibilities are almost infinite, and perhaps dangerous to try so late in the season.

But you’ve got to try something, right?

Or are these numbers the product of a small enough sample size that you don’t pay them much heed?

The following two tabs change content below.

Zach Lowe

Latest posts by Zach Lowe (see all)

Share →
  • Joel W

    Just don’t play ‘Sheed in the second half. Give those minutes to Sheldon and Baby. If he has a problem with it, tell him to do wind sprints for the next three weeks or STFU.

    The other possibility is that it’s tight legs coming out of half-time, where they stop moving. Is it possible that during half time the team should be riding exercise bikes in the locker room or something?

  • Joel W

    I’m trying to find half-by-half splits to see how Baby and Sheldon and TA perform in the 3rd quarter.

    I really do think the cold legs thing is relevant. I remember Dennis Rodman used to keep his legs warm while on the bench as he got older, and the amphetamines could only keep him going so much. It’s also consistent with the theory that they’re worse in the 3rd than the fourth. The starters warm back up in the fourth.

    Perhaps Doc needs to be aware of this in his resting patterns also. Have the starters play more minutes in the second half than in the first, rather than the other way around, that way there’s less sitting at the end of the third, early fourth.

  • Cptn Bubbles

    This problem has happened far too many times to be ignored. It does not seem to happen all the time, but it is recurrent. At the core, the #1 problem is Doc RECOGNIZING it when it is happening & then actively doing something DIFFERENT.

    What you see is a massive fail at BOTH ends of the floor. Doc should be super sensitive in 3rd quarters (wary in other quarters) to the Cs lack of production & use time outs & SUBSTITUTIONS to try & nip this thing in the bud. They should DELIBERATELY try to go inside & get to the foul line & change the momentum/pace of the game when the ship is taking on water.

    Doc has a philosophy of sticking with ALL the starters no matter what, but he doesn’t have to sub all the starters, probably just 1 or 2 will spark a change. I was impressed with how TA came in & brought tons of energy. You could see the rest of the starters kick it up a notch (Paul had been playing some decent defense, but Quis was just reaching).

    TA is the best at sprinting like a madman to the other end of the floor when there is a turnover. This gives Rondo a ready target for an ez bucket. There is no one else who gets out as fast in transition with Rondo as TA. It leads to ez buckets because 80% of the defense is left in the dust. You can see that it really pumps Rondo up when he gets that fast break assist (which means Rondo with more energy on D). I like Quis, but TA is a better defender (look at what he did to Kobe etc.). TA is tougher than Quis & will put a body on people. TA is also the best on the team (along with a motivated Paul) at ball denial (Billups had to go somewhere else when TA was all over a hot Melo that 1 possession). The Cs are supposed to be all about defense, right? If you have good shooters on the floor then you should have 1 of your best defenders out there with them.

    TA will get you some ez points with offensive rebounds or running very hard on breaks. TA’s cuts in the Orlando Christmas game made a huge difference, & he got out with Rondo in transition (remember that behind the back to TA slam?). While all the other Cs are trying to process what to do, TA already has a 3 or 4 step sprint lead to the other end of the floor. It is sad because we can’t see Rondo’s full potential due to the lack of speed of fellow teammates to keep up in transition—especially in the 2nd half of games.

    I have to believe that one of the starters knows when one of these meltdowns is occurring. If Doc is just going to continue to ignore what is happening right before his eyes then one of the players should call a time out. Right now, everyone is just pretending that it is not happening. Ignoring it is just contributing to your own self destruction. There should be a number in point differential in Doc’s mind which triggers an automatic time out or substitution instead of this ‘let the starter’s work it out’ mantra which has cost the team far too many games already. Doc has made some good coaching decisions, but he HAS to recognize that he is not right about everything. This is not some fluke. It reoccurs because the same old same old is not going to fix it.

  • DRJ1

    There have to be statistical anomalies in these numbers. The system they’re studying is too complicated.

    For example, a top scorer like Pierce is going to have his 4th quarter numbers skewed artificially downward, relative to earlier quarters, because he tends to play more 4th-quarter minutes in close games than blowouts. Blowouts, or even good Celtics games, are typically games where Paul is playing great ball. If he and the team play well enough, Paul DOESN’T PLAY in the 4th quarter, or plays little. The net effect is that Paul’s totals are skewed TOWARD closer games, and AWAY from games where he is playing his best. By removing the most positive statistical weight from his 4th quarter season totals, this helps make it seem as if Paul is less productive in 4th quarters, whether he really is or not.

    This same concept applies to the team’s quarterly stats too. The best games are when the starters don’t play in the 4th, so naturally the team’s overall performance will be skewed downward then (assuming our secondary units perform less well than our starters).

    I’m pretty sure there are other anomalies we can find, if we think about it enough. So I’m not sure how much can be learned from these stats, in the end. (And to fix the numbers to account for each potential anomaly is a nightmare task no one (sane) is going to even think about.)

  • @Bubbs: I like the idea that the foul line is your friend when all else is failing;

    @DRJ: I agree the #s can be skewed in the way you indicate, particularly the 4thQ numbers. But the 3rd Q offensive efficiency #s worry me.

  • Pingback: What are common treatments for postpartum depression? | ISelfCenter.com()

  • DRJ1

    Btw, the change in defensive eff. numbers is just as bad, if not worse, than the change in offensive numbers for the latter quarters.

    Anomalies aside, it all speaks of lack of effort and focus… the 3rd (and 4th) quarter collapses we’ve seen so much this season. I think of it as ‘regular season malaise’… which is not unexpected, and probably not even inappropriate, given the ridiculous length and grind of those 82 games.

    So in terms of ACTION to take… they’re already doing it. Nothing specific related to these particular numbers… just playing better, and for 48 minutes. When they do that, the numbers take care of themselves.

  • @DRJ: But the def efficiency #s in our “bad” quarters–102-104 in range–are still elite by league standards. The off eff #s are piss poor.

  • DRJ1

    @Zach– Got it, thanks.
    So as a “defense-first” team, they were starting games with great D and good-to-medium O, then ending them with mediocre D and poor O. Both drop-offs were problems.
    But I agree with Doc that they need to focus on the D, and let the O flow from the D. Instead, what they did in their bad stretch was get all bent out of shape over their bad O and let the D suffer as a result. Now it’s changed. In the Nuggets game, we saw many instances of O flowing from D… and it was beautiful.

  • Pingback: Should my silicone baking sheet turn dark? Will silicone affect baking time? | Silicone Baking Sheet()

  • I love Green

    I think in the third quarter when we have leads, we don’t come out as agressive. We don’t feel the need to score every possession, which causes us to take more jump shots.

  • Ray Leighton

    Lots of good thoughts here. But this reminds me of an earlier post commenting that Paul has been having his best shooting games when he spent a chunk of time on the bench because of foul trouble. Then he comes out later in the game and shoots the lights out. That sounds to me very much like he simply needs to be resting for another five minutes every game. My sense is that Ray, Paul, and KG all typically have a drop in production per minute as their minutes get high. It would be nice to see some data on this, but it would take a lot of work.

    In any event, I agree with the Capt’s idea (and mentioned this in the TA post earlier) — put TA in for those extra minutes. I like the idea of a smaller, quicker, defense- and transition-focused unit coming in early in the 3rd quarter to harass and wear down the other team.

  • GranTur

    It’s always about Pierce.

  • Pingback: Learning the Difference Between NBA and NCAA Basketball | The Jump Manual Review: How To Increase Your Vertical Jump()

  • aarongoil20100408
    I think in the third quarter when we have leads, we don’t come out as agressive. We don’t feel the need to score every possession, which causes us to take more jump shots.