Orlando Magic Preseason Lookback, Vol. 3: Five-Man Unit Data

I've mentioned some of the Orlando Magic's lineup data in a few other posts on this site, but for what may be the final installment in the Preseason Lookback series, I wanted to attend to those data in greater detail.

The Magic's preseason spanned 336 minutes over seven games, during which time coach Stan Van Gundy deployed 58 distinct lineups. However, just 18 of those lineups spent more than 6 minutes on the court together; the other 40 ranged in use from 18 seconds to 5 minutes, 47 seconds, so I'll spend considerably less effort explicating them. Mickael Pietrus is the only rotation player who missed more than the customary one game for rest, which means Van Gundy indeed had a mostly full complement of players to mix and match.

First and foremost, the five likely starters steamrolled their competition, as I documented last week. With Jameer Nelson and Vince Carter in the backcourt, with Quentin Richardson, Rashard Lewis, and Dwight Howard up front, the Magic outscored their opponents by 41 points in 39:29. Though the numbers are not adjusted for opponent quality, they're nonetheless impressive, and demonstrate the ease with which the free-agent signee Richardson has assimilated with the team.

The next most used lineup, at 26:01, featured Lewis sliding to small forward to make room for Brandon Bass at power forward. This group, too, proved effective: it had a plus-27 scoring differential and did most of its damage on the defensive end. Its offensive rating of 109.6 is more "good" than "great," but the D yielded an incredible 60 points per 100 possessions. Bear in mind that, prior to taking the court this preseason, Lewis and Bass had never manned the forward positions together; their shared time came with Lewis at power forward and Bass at center.

As a result of Orlando's frequent blowouts--it won its final five games by an average of 31.8 points, with its smallest victory margin in that span, 13, coming with Carter, Lewis, and Howard resting--Van Gundy often had to turn to his second unit. It proved ineffective due to its shortcomings on offense, something which he might need to address, given that he might actually need to use this lineup in the regular season. In 16:59 on the floor, Magic opponents outscored the lineup of Chris Duhon, J.J. Redick, Mickael Pietrus, Bass, and Marcin Gortat by a 30-18 margin. This group scored its 18 points on 36 possessions, strongly indicating that it needs an additional scoring presence. Duhon isn't a threat, and with Pietrus' shot off and Gortat rarely sought, the task fell to Redick and Bass to generate offense here.

Subbing Ryan Anderson for Brandon Bass in that lineup makes sense, as the Magic scored 115 points per 100 possessions with Anderson in Bass' place with those other four reserves. The group with Anderson played 10:31 this preseason, making it the fifth most-often used lineup.

After the jump, a look at the most effective players at each position.

POSITION-BY-POSITION PLAY, BY INDIVIDUAL

Here's how the team performed, on both ends of the court, with the given player on the floor at that particular position. The table is sorted by time spent positionally, and then by time, in descending order. The team leaders' lines are shaded in gray, with a minimum threshold of 50 minutes required:

OBSERVATIONS

The team was at its very best with Nelson running the show on offense, and wasn't too bad with him on defense, either. As such, he owns the team's second-best efficiency differential.

Lewis is the team's best option at either forward slot, which lends credence to Van Gundy's idea that it's best to move Lewis around based on matchups, which in turn gives the Magic comfort with different, versatile offensive approaches.

I'm surprised we saw so litle of Carter at small forward, given how often he and Redick paired up on the wings in the postseason. And Gortat got hardly any time alongside Howard in a big lineup, though when together, they proved effective.

Please note that I'm still compiling positional pair data--for example, which guard combinations worked the best?--and will post on them when I have more time.