2008-2009 Advanced Splits

The three stats we will be looking at are pace, offensive rating and defensive rating. Pace is essentially possessions per game, offensive rating is points scored per 100 possessions, defensive rating is points allowed per 100 possessions. The formula I used to calculate the possessions was from Dean Oliver's book, Basketball On Paper, which again, every basketball fan interested in something more meaningful than PPG should read.

Here's a visual representation of the Sixers' pace throughout the season. The black line represents the overall trend.
And below you will find the visual representation of the team's offensive and defensive efficiency ratings throughout the season. Take a quick look and then we'll really dive in.
 
I'll be honest with you guys, I just wanted to throw together a couple of graphs. You can sort of get the picture of what happened during the season from the two above, but looking at the splits is really where the rubber meets the road. This chart should open some eyes.
 
Conventional wisdom seems to be that Tony DiLeo engineered the turnaround of the team by getting back to the running style the Sixers played so well the year before. If you look at the numbers, though, you'll see that the team's pace actually slowed once DiLeo took over. Take a look at the Elton Brand splits as well. The margin was even greater when he was on the floor vs. off, and again, the shift was a slower pace without Brand.

For the life of me, I simply cannot figure out these splits, at least not in any way that I can back up with numbers. With Brand, both teams combined for nearly two more offensive rebounds per game. Offensive rebounds typically slow the pace down, because they extend a possession and take more time off the clock. Without Brand, the teams turned the ball over a tad more frequently, which should lead to a faster pace. However, the trend went in the opposite direction in both cases.

The only thing that I can come with that makes any kind of sense is that the Sixers style actually slowed the pace, intentionally in some cases, and I think there is empirical evidence to back that up. Under DiLeo, the Sixers frequently used pressure on the ball to make the opposing teams burn some clock before they could get into their half-court sets. On the other side of the ball, yes the Sixers ran at every possible opportunity, but when they weren't running, the ball could usually be found at the top of the key in a guard's hand as the clock wound down. This would probably explain the pace split in wins vs. losses as well. Under DiLeo, the Sixers pushed the pace through their running game, exclusively. They never took quick shots in the half court. The easy correlation is that when they ran, the pace went up and they won. When they couldn't get out in the open floor, the pace slowed, and they lost.

That explains why the pace slowed after DiLeo took over, but I also believe the Brand splits (8 of Brand's 29 games played were played for DiLeo, so we're talking about an overlap of 28% of Brand's games) are meaningful, for another reason. It's been written a million times that Brand slowed the Sixers down, but that's simply not true. They were a fine running team with him on the floor, in fact I'd say they were a better running team with him on the floor, simply because having him on the floor equated to more opportunities to run. The opposing teams missed 2 more shots per game with Brand on the floor. The Sixers grabbed 3 more defensive rebounds. In a nutshell, they played far superior defense, thus creating more opportunities to get out on the break.

Now, looking at the offensive and defensive rating splits it's pretty clear what changed under DiLeo. The offense began to work. While the isolations may have been maddening to watch, they were actually a much more effective offense for the personnel to run. They opened the floor up for driving lanes, and they kept the ball in hands of the playmakers. Under Cheeks, the ball was dumped into the post so Elton could go against a double team, every time, with no outlet, no spacing and no threats from the outside. If you look at the difference between the Cheeks/DiLeo splits and the with/without Elton splits, you can see that under DiLeo with Elton the team performed better than under Cheeks, with Brand (I should've run this split as well. I'll do it later today.)

As for the defensive rating, I think we can attribute the horrific falloff under DiLeo to a couple of things. First and foremost, the loss of Elton Brand and the domino effect of moving Thad to the four and Willie Green to the two. The personnel was simply not as good. We also have to place some blame on the rotating, pressure defense. Under Cheeks opponents shot 30% from three against the Sixers. Under DiLeo, they shot a whopping 40%. Add in the rebounding disadvantage, and it's not hard to connect the dots.

That's all the analysis I have in for me at the moment. Take a close look at the graphs and chart above and let me know what you can come up with in the comments.