Building Lineups in Cleveland

October 25, 2012

By Ian Levy


Byron Scott worked has worked his way through the canyons, and the stinging talent drought that marked the beginning of his time in Cleveland. For the first time as Cavaliers’ head coach he has a roster made up almost entirely of legitimate NBA players. Player development needs to be an ongoing focus, but it doesn’t need to dominate his agenda in the same way it has the past two seasons. The pieces are there for him to assemble rotations than can compete for a playoff spot in the Eastern Conference. Doling out minutes is one of the chief responsibilities of a head coach, but one that is often forgotten in the murky stew of motivation, player and media relations, and tactical aptitude.



Last summer I devised a quick method for measuring a coach’s success with this one small area of their professional responsibilities. The premise is simple. The best coaches should put their best player combinations on the floor, for as much time as possible. My evaluation tool was, for each head coach, to look at the correlation between the effectiveness of each lineup and the number of minutes their head coach played them. I used Net Rating (ORtg. – DRtg.) to measure effectiveness and weighted that number for each lineup by the number of minutes they actually played. This method has its flaws, but is also powerful in that it is constricted by a coach’s options. It doesn’t matter if a coach has good or bad players, what matters is that they do the best they can with what they have.


Interestingly, the correlations for each coach tended to vary quite a bit from season to season, indicating that this particular skill may not be implemented consistently. In my initial analysis, which covered 2008-2011 Scott was one of the coaches who’s numbers changed quite a bit. During the 2008 and 2009 seasons with New Orleans he was among the most effective coaches at managing rotations, with correlations of 0.889 and 0.470 respectively. In 2011, his first year in Cleveland, that number fell to -0.336 (remember that a negative correlation indicates that the worse a lineup performed, the more minutes it saw, with a correlation of -1.0 taking that situation to its complete extreme). In last year’s lockout-shortened season, the situation was roughly the same with Scott’s correlation coming out to -0.239.


Scott hasn’t found a rhythm to his rotations yet in Cleveland, but the jump in talent on this roster should make many decisions easier. He’ll no longer be forced to rely so heavily on a collection of misnomers - Antawn Jamison, the power forward with no power. Daniel Gibson, the shooting guard who can’t shoot. Samardo Samuels, the post scorer who can’t score in the post. Samuels and Gibson are still on the roster, but additions like C.J. Miles, Jon Leuer, Tyler Zeller and Dion Waiters mean both can be pushed down in the rotation.


As Scott experiments through the preseason there are a few issues he’ll need to resolve in deciding who plays and when. The first is how he’ll pair players in the frontcourt.


Judging by the lineups he used in preseason, Anderson Varejao and Tristan Thompson will be the frontcourt starters. There is plenty of promise with that pairing, but in limited minutes last year they were a horrible combination. In the 47 minutes they played together the Cavs were outscored by 28 points and shot just 36.5% as a team. They were able to collect a ton of their own misses, with an OReb% of 37.2% but they also turned the ball over on 30.2% of their possessions. The Varejao and Thompson pairing is probably the best defensive and rebounding frontcourt that the Cavs can put forward, but it puts them at a decided disadvantage at the other end of the floor. Neither player provides much on offense besides the ability to hit the glass and finish at the rim. Playing both for extended minutes also parks extra bodies in the lane, which brings me to the next concern.



Who will space the floor for Irving? As inefficient as Jamison revealed himself to be during his time in Cleveland, he was still an offensive threat respected by defenses. His ability to shoot from the outside gave the offense much-needed flexibility, even when he wasn’t the one shooting. Last season Irving shot 46.9% from the field and 39.9% on the three-pointers. 47.8% of his field goal attempts came from inside of 10 ft. and he averaged 7.3 assists per 40 minutes to 3.8 turnovers. However if we just look at the 162 minutes he played without Jamison the numbers change drastically – 35.8% from the field, 38.9% on three-pointers, 35.9% of field goal attempts coming from inside of 10ft., 4.7 assists per 40, 6.4 turnovers per 40.


Irving’s ability to get into the lane was incredibly special last season but it may not be sustainable if the Cavs can’t get shooters on the floor to open the lane for him. This is easier said than done, and looking at the wings there may not be much help. Gibson made threes last year, but nothing else. Both Alonzo Gee and Omri Casspi struggled to consistently make shots from the corners, and Waiters has some proving to do in that area as well. Miles sports a career FG% of 41.9%, with a 3PT% that has declined in four consecutive seasons, bottoming out at 30.7% last year.


The key may be how Scott is able to use newcomers Zeller and Leuer. Both have much more versatile offensive games than Thompson and Varejao and with the ability to pick and pop or knock down a mid-range jumper on a kick out, can help keep those driving lanes open. While Varejao and Thompson may be the starters, it seems probable that the Cavaliers’ most effective lineups may involve mixing and matching with Leuer and Zeller.


Hope and talent have both arrived in Cleveland, in quantities that just two years ago felt like a dream. If those hopes are to be realized than Byron Scott will need to find ways to use that talent in the most effective combinations possible.