Sunday, March 6, 2011

The Celtics: Looking at the Numbers on Pairing a Center with the Big Four




On the night of the Kendrick Perkins trade, Danny Ainge made some intriguing statements regarding Shaq's role replacing for Perkins.  His remarks indicated that he understood the simple analysis of the situation -- the Celtics went 33-10 before Perkins' return to the team.  His words, however, also implied the team's usage of more advanced, lineup-based analytics:

Shaq has proved to be excellent with our starters.  The numbers actually show that he has been better with our starters.  We beat all the good teams in the league while Kendrick was out.


With that type of analysis in mind, let's take a look at the Celtic lineups.  Ranked by playing time, six out of the top nine Boston five-man units feature the Big Four (Rondo, Allen, Pierce, Garnett), along with one extra big man.   The data for those six lineups are shown below:


Click to expand.  Data, sorted by minutes, was obtained from
basketballvalue.com and includes games through March 4.


There's clearly a dividing line between the Davis/Shaq/Perk and JO/Semih/Krstic, because the number of minutes played for the bottom trio is really too small of a sample size to draw any reasonable conclusions.  The standard error in the second column bolsters that idea. But it also doesn't take a huge leap of faith to say that the adjusted +/- and overall rating numbers for Jermaine and Erden tell the same story as our eyeballs:  they are players of middling ability.

Given the health of the other centers Boston has used, trading Erden to Cleveland seemed like a risky move.  Semih showed flashes of being able to finish at the rim, and adequacy in other areas. Plus, he was playing injured, so his ceiling may be even higher. But getting a 2nd-round pick of Cleveland's is a strong move -- not only should it be near the top of the round, but 2nd-round selections afford more financial flexibility that 1st-round picks.

The Celtics offense has performed magnificently with Krstic on the floor in his short tenure in Boston.  The team is 4-0 since his arrival, and the Rondo-Allen-Pierce-Garnett-Krstic lineup has moved the ball to open players with great skill.  Again, it's a ridiculously small sample size, but the short-term level of success is lofty enough that, even if it falls a bit, it will still be above-average.  Even before Nenad donned the Celtic uniform, Ainge spoke highly of him:

Krstic is a terrific shooter.  I think he complements Rondo's game very well in that he's another guy that can knock down that mid-range jump shot.

The other bonus for Boston is that Krstic is currently healthy while Perk, Shaq, Big Baby, Jermaine, and Erden wait on the sidelines to recover from their injuries.

When healthy, Shaq logged time with the Big Four as a starter;  Glen Davis has spent most of the season as the player finishing games with them.  Kendrick Perkins got heavy minutes upon his return as Shaq, Jermaine, and at times, Erden, were injured.  As a result, we have more reasonably-sized sets of data to study than we did for Erden/Jermaine/Krstic.

What do the data show?  They back up Ainge's conclusions.  The defensive ratings for the three players are similar, but Shaq's offense stands out noticeably.  His ability to finish inside gives the Celtics more potency than what Davis or Perkins offer.  In fact, no other five-man lineup in the league (with equal or higher minutes) has a ratings differential higher than the 18.81 posted by Shaq + the Big Four.  Clearly, Ainge knew this information before completing the trade.  And looking at adjusted plus/minus, even if we take into account the standard error, Shaq is the top producer.  He'll be the starting center in the playoffs, if (and what a huge 'if' it is) healthy.

Of course, the impetus behind the Perkins trade is that the crunchtime lineups may not feature any of these centers with the Big Four.  Instead, Ainge wants to be have the flexibility to employ Jeff Green as a 'stretch-4', in a manner similar to the way James Posey was used during Boston's 2008 title run:  as an offensive threat who can shoot from outside to stretch the floor and as a defender capable of matching up with sizable forwards anywhere on the court.  The remaining 22 regular-season games may preview Green's ability to succeed in that role, but the true litmus test will come in the playoffs.  Given Boston's lofty expectations, anything short of success in the NBA Finals will be viewed as a failure.



No comments:

Post a Comment