5-Man Units After 25 Games

The Cavs have played 25 games now, and it’s safe to say that the lineup data available is at least semi-reflective of the teams’ true form. It’s really easy to just look at the data for five-man lineups and conclude that whichever one has the highest plus-minus per game (PM) is the best one. That’s true in some ways, but it doesn’t tell the whole story.

It’s a lot harder to look at all of the lineups and figure out, which ones (especially those that aren’t the clear cut best) are performing well. An effective way of examining 5-man lineup data is to look at how a given unit’s plus-minus per 48 minutes (PM/48) compares to the plus-minus per 48 minutes average of the players that comprise it. To do this, I gathered the actual on-the-court PM/48 data for lineups, then I looked up individual player’s PM/48. Finally, I added together each individual players PM/48 in a given unit to determine the average. That number was a unit’s expected PM/48. I compared each unit’s expected PM number to their actual PM number to figure out which five man groups were under-performing and which were over-performing.

The Numbers

The chart below reflects the exercise I performed for the Cavs top ten most used 5-man groupings. They are in order from most used to least used.

Some Quick Thoughts

1. Of the Cavs’ ten most used lineups, five of them are actually under-performing. The ones that are only outscoring their expected PM/48 by 1.56 points and 1.84 points should probably be counted as under-performers as well. So, basically seven of the Cavs’ ten most used lineups probably shouldn’t be seeing the floor with regularity. That’s cause for concern. After the big trades last season for J.R. Smith, Iman Shumpert, and Timofey Mozgov, just eight of the Cavs top 20 most used lineups were under-performing.

2. Most of the under-performing units have something in common (the exceptions being the Dellavedova, James, Jefferson, Jones, and Thompson group out-performing their expected PM/48 by 1.84 points, and the Cunningham, James, Love, Smith, and Thompson unit under-performing by 9.82 points/48 minutes): Mo Williams or Timofey Mozgov are featured in them. If you watch the Cavs regularly, this probably doesn’t come as a surprise to you. If Mo isn’t hitting his shots, he’s a negative for the team. When Mo covers opposing point guards, they often do a simple left-to-right dribble crossover for a shot at the rim, or an assist. And, Timo for most of the season, has been a glorified statue, incapable of catching a pass rolling to the rim, and unable to figure out how to contest penetration without leaving the hoop wide open.

3. Fortunately for the Cavs, Kyrie Irving is back, and Mo will be relegated to the bench for long stretches of time. If he isn’t hitting his shots, he won’t have to play. And, Mozgov is finally playing much better. Through the first 22 games of the season, Timo averaged just 7.3 points. 4.1 rebounds, and 1.5 turnovers on 51.5% shooting. The past three games, Timo has averaged 7.7 points, six rebounds, and .3 turnovers on 62.5% shooting.

4. The lineups that out-perform expectations also have something in common: they feature Dellavedova, Love, the King, and Thompson. Delly and the King sport individual PM/48 of 13.9 and 13.5, respectively. And, while there are still occasional rumblings that Delly isn’t a starter level player, or even above average player, people need to keep in mind that Delly has earned his PM/48. Early in the season, Delly was often playing with bench groups that were without Love and LeBron for stretches. Delly has figured out how to fit into any situation and make the most of it.

5. The Cavs are only going to get better. Think about this: The best 5-man group to date is Delly, Williams, James, Love, and TT. They are outperforming their expected PM/48 by 47.92. Now, think about swapping out Williams for Irving. Both players are score-first guards, and both can spot up for 3-pointers. Kyrie is a better shooter and scorer, however.

6. It’s a bit surprising that the Cavs’ best-performing group this season features three players that are offensive minded. Last season, the surprise group for the team was Delly, Shump, James, Tristan, and Mozgov. During the Finals, they put up a PM/48 of just 1.7. That’s not a great number; however, they were the only Cavalier group that played over twenty minutes together in the Finals and posted a positive PM/48. The groups’ sole source of offense was the Le-Iso machine, and they posted just 15.2 assists/48 and 21.2 turnovers/48. This seasons’ group of Williams, James, Delly, TT, and Love is averaging 21.5 assists/48 and just 16.1 turnovers/48. That’s an identity change for the ages.

The Death Lineup And The Cavs

I have to bring up the Warriors. Everyone is talking about their death lineup of Curry, Thompson, Iggy, Barnes, and Green. This grouping has an expected PM/48 of 19. Their actual PM/48 is 67.9. They are performing 48.9 points better than expected over a full game. That’s just .98 points more of an over performance than the Williams, James, Delly, Tristan, Love group though. The death lineup has a better overall PM/48, but these two units are more similar than people realize.

It’s also very interesting to look at how these two groups match-up, even without thinking of Kyrie Irving as a starter for the Cavs. The Warriors’ defense in the death unit has been astounding, allowing just 84.8 points per 100 possessions. Their offense is scoring over 150 points per 100 possessions. They’re obviously playing both ends of the floor.

Yet, the Cavs will be a formidable foe on the 25th of December.

The Warriors often rely on a 1-5 pick and roll to generate offense in their small ball lineup. The Cavs will need to switch on this play occasionally and Tristan Thompson will land on Curry. While Curry beat TT numerous times in the Finals, he didn’t beat him from beyond the arc. TT has the foot speed to at least run Curry off the three-point line to prevent an easy look, a huge part of the Warriors’ offensive dominance. Furthermore, tossing the ball into Draymond Green, when Delly is on him, won’t be easy work. Delly has shown time and time again, that he can front a bigger man and prevent the easy pass. Just go watch the Nets’ game from earlier this season where Delly denied Brook Lopez (a true seven footer) the ball several times.

And, Delly is decent at covering Curry. He at least makes him work, and won’t give up 25 foot 3-pointers.

Offensively, TT and Love are both very good offensive rebounders. TT is grabbing 2.9 offensive boards a game while Love is grabbing 2.5 per game (14th and 27th respectively). I’m not confident that Barnes and Green could keep them off of the boards. Ultimately, the Cavs will have to play chicken with the Warriors and see if they can out-rebound their death lineup by so much it won’t be worth keeping them on the floor. Thankfully, for all basketball fans, these two super lineups will do battle in a little less than three days time, and the world will see which unit is better.