The Golden State Warriors are about to return from their little December break, which means it’s a great time to partake in one of my favorite activities: dragging a fine comb through a tiny set of data.
With 25 games in the books, I decided it’s time to take a look at some of the best and worst lineups that the team has put on the court this year.
Using the NBA’s Advanced Stats tool, I looked at the best and worst five-player, four-player, three-player, and two-player combinations for the Dubs, based on net rating (point differential per 100 possessions). On the year, Golden State has a garbage-time adjusted net rating of +1.6, which is 14th in the NBA. They have an offensive rating (garbage-time adjusted) of 113.6, which is 21st in the league, and a defensive rating (garbage-time adjusted) of 112.0, which is fourth in the NBA.
The following lineup data is not garbage-time adjusted. Enjoy!
The best 5-player lineups (minimum: 20 minutes)
- Podziemski/Hield/Butler/Green/Post — +19.8 in 24 minutes (105.8 offense; 86.0 defense)
- Curry/Podziemski/Butler/Kuminga/Green — +16.9 in 65 minutes (119.7 offense; 102.8 defense)
- Podziemski/Hield/Moody/Butler/Horford — +15.9 in 21 minutes (118.2 offense; 102.3 defense)
Honestly, my biggest takeaway here is how few lineups the Warriors have given significant run to, in large part because of their frequently changing starting lineup. In total, there were only six five-man lineups that fit the minimum of 20 minutes, with the Opening Night lineup — featuring the previously presumed unplayable trio of Jonathan Kuminga, Draymond Green, and Jimmy Butler III — topping all of them with 65 minutes played.
To pick a random team, the Los Angeles Lakers squad that they played in that Opening Night showdown has used 11 different five-man lineups, with the most frequently-used lineup accruing 80 minutes. The world-beating Oklahoma City Thunder also only have six such lineups, but their main one tops out at 112 minutes.
The worst 5-player lineups (minimum: 20 minutes)
- Podziemski/Moody/Butler/Green/Post — +4.1 in 61 minutes (103.3 offense; 99.2 defense)
- Curry/Butler/Kuminga/Green/Post — +13.7 in 44 minutes (122.3 offense; 108.6 defense)
- Curry/Richard/Moody/Butler/Green — +13.8 in 55 minutes (122.2 offense; 108.4 defense)
As previously mentioned, the Warriors only had six lineups that hit the 20-minute qualifier, which means you’re seeing all of the frequently-used five-man rosters between these first two sections. The main takeaway here is that Steve Kerr is doing a good job of not lingering on bad lineups: all six of these lineups have had good results, and five have had elite outcomes.
The best 4-player lineups (minimum: 30 minutes)
- Spencer/Richard/Hield/Post — +40.0 in 34 minutes (132.9 offense; 92.9 defense)
- Podziemski/Hield/Moody/Butler — +28.0 in 34 minutes (121.1 offense; 93.1 defense)
- Curry/Podziemski/Moody/Green — +21.7 in 34 minutes (105.1 offense; 83.3 defense)
There’s really only one takeaway here: you should spend at least five minutes absolutely howling at the fact that two-way contract Pat Spencer, second-round rookie Will Richard, second-round sophomore Quinten Post, and lovable goofball Buddy Hield have a +40.0 net rating in a not-insignificant number of minutes. It’s mostly the Chicago game that is driving that, but it’s still utterly phenomenal theater.
The worst 4-player lineups (minimum: 30 minutes)
- Podziemski/Moody/Santos/Jackson-Davis — -17.8 in 32 minutes (95.6 offense; 113.4 defense)
- Podziemski/Richard/Butler/Green — -12.5 in 35 minutes (100.0 offense; 112.5 defense)
- Podziemski/Richard/Moody/Jackson-Davis — -9.9 in 38 minutes (98.8 offense; 108.6 defense)
It’s kind of funny how many of these lineups — positive and negative — Brandin Podziemski is already featured in. We’ll see what the three-man lineup data says, but I’d uhh … maybe think twice about playing him alongside Moses Moody and Trayce Jackson-Davis!
The 30-minute cutoff (which covered 32 lineups) narrowly saved a few bangers from making the list, notably the quartet of Podziemski, Moody, Jonathan Kuminga, and Draymond Green, which has produced a truly stupefying net rating of -60.9 in 27 minutes.
The best 3-player lineups (minimum: 80 minutes)
- Curry/Green/Post — +20.8 in 80 minutes (126.2 offense; 105.4 defense)
- Curry/Podziemski/Butler — +19.1 in 113 minutes (121.0 offense; 101.9 defense)
- Curry/Richard/Green — +18.3 in 81 minutes (122.4 offense; 104.0 defense)
- Moody/Butler/Post — +17.5 in 104 minutes (113.8 offense; 96.3 defense)
- Curry/Podziemski/Green — +17.1 in 131 minutes (113.3 offense; 96.2 defense)
With so many more lineups to choose from as we move from quintets and quartets to trios and duos, it’s time to start looking at the five best/worst instead of just three lineups. It should come as no surprise that Curry is a staple of these lists. Even for a team that is making its mark on defense, Curry’s offense is so transformative that everything just works better when he’s on the court.
The minutes cutoff robbed us of another banger: the threesome of Pat Spencer, Will Richard, and Quinten Post is +57.1 in 57 minutes.
The worst 3-player lineups (minimum: 80 minutes)
- Curry/Podziemski/Kuminga — -3.6 in 98 minutes (108.8 offense; 112.4 defense)
- Podziemski/Hield/Horford — -2.5 in 81 minutes (112.5 offense; 115.0 defense)
- Podziemski/Kuminga/Green — -2.3 in 125 minutes (105.5 offense; 107.8 defense)
- Podziemski/Moody/Green — -2.0 in 164 minutes (97.4 offense; 99.4 defense)
- Richard/Butler/Green — +0.4 in 102 minutes (113.5 offense; 113.1 defense)
The main takeaway here (other than that Brandin Podziemski fans should look away) is the same as in the five-player lineups: Steve Kerr is doing a good job avoiding bad lineups. The Warriors have used 33 different three-player lineups for at least 80 minutes, and only four such configurations have a negative net rating.
I’ll be curious to see how that holds up over time. The cutoff is saving a few disastrous trios from making an appearance — Brandin Podziemski, Buddy Hield, and Jonathan Kuminga are -38.0 in 51 minutes, for example — so we may have to wait to see if Kerr is actually staying away from these rough lineups, or if it’s just because of how early in the season we are.
The best 2-player lineups (minimum: 125 minutes)
- Spencer/Post — +27.6 in 133 minutes (124.0 offense; 96.4 defense)
- Butler/Post — +17.8 in 242 minutes (117.0 offense, 99.2 defense)
- Curry/Green — +15.5 in 321 minutes (121.4 offense; 105.9 defense)
- Green/Post — +14.4 in 208 minutes (113.9 offense; 99.5 defense)
- Moody/Post — +12.8 in 193 minutes (109.6 offense; 96.8 defense)
I expected one name to populate at least four of these duos. I just, uhh … well … expected that name to be Steph Curry, not Quinten Post.
There’s no need to overcomplicate this one. Within the systems that the Warriors play, Post is one of their very best players, and he should be both starting and playing heavy minutes (Curry and Post together, by the way, would have been the sixth duo on this list if you’re wondering). If you want a reason for optimism that the Warriors might have turned a corner and are ready to make a run, Post turning a corner and the coaching staff realizing it should give you that optimism.
The worst 2-player lineups (minimum: 125 minutes)
- Hield/Kuminga — -15.2 in 172 minutes (100.0 offense; 115.2 defense)
- Richard/Hield — -15.1 in 158 minutes (105.7 offense; 120.8 defense)
- Curry/Horford — -13.0 in 141 minutes (108.1 offense; 121.1 defense)
- Podziemski/Hield — -12.8 in 236 minutes (98.2 offense; 111.0 defense)
- Payton/Hield — -11.6 in 128 minutes (103.2 offense; 114.8 defense)
If the takeaway from the best two-player lineups is that Kerr should be playing Quinten Post more, then the takeway from the worst two-player lineups is that Kerr should be playing Buddy Hield less, which seems likely to occur given De’Anthony Melton’s return and Pat Spencer’s breakout (plus the addition of Seth Curry).
Hield probably doesn’t deserve as much blame as the numbers suggest, as he’s certainly hurt by the amount of time he spends on the court while Steph Curry is off it. And speaking of the older Curry brother, they desperately need to reverse those numbers with All Horford sometime between now and April.
Well, there you have it, folks. What stands out to you from these still-very-small samples?
