Season Preview: The Starters/Bench Divide



With the Grizzlies opening training camp Monday, now's the time to begin our season-preview coverage. Like last year, in addition to preseason game coverage, links, and league-wide predictions, I'll be doing a series of more in-depth blog posts on team-related topics leading up to my season-preview piece in the print edition of the Flyer.

I'm starting with what I think is the fundamental issue for last year's team and the biggest looming question for the coming season — the gulf between the team's returning starting lineup and its bench.

These five guys are playoff-ready, but they need a whole lot more help.
  • These five guys are playoff-ready, but they need a whole lot more help.

This post is rooted in a comparison of starting (or primary) lineups across the league and how teams fared with and without those lineups on the floor. I'm going to post a spreadsheet (of sorts) comparing relevant stats for all 30 NBA teams — numbers derived from the unit stats found at — and a key explaining what each of the columns of numbers is. But if you don't want to wade that deep into the numbers, you can skip over this part and just read the comments afterward based on these stats.


PL+/-: This is the plus/minus rating for each team's primary (most-used) lineup last season, per 48 minutes.

PLM: This is the total number of minutes played by each team's primary lineup.

NPL+/-: This is the per 48 minute plus/minus for each team when not using its primary lineup.


Dallas +18.6 342.35 +1.2
Orlando +15.6 773.15 +5.5
Portland +13.7 372.45 +2.5
Boston +12.4 1154.45 even
L.A. Lakers +12.3 649.78 +3.3
Milwaukee +11.2 257.90 +0.7
L.A. Clippers +10.5 406.77 -8.3
San Antonio +10.0 230.88 +4.8
Phoenix +8.9 828.95 +3.8
Miami +8.0 286.73 +1.8
Memphis +7.3 1474.80 -6.7
Denver +6.4 520.47 +3.4
Atlanta +5.0 1169.08 +4.5
Utah +4.1 327.92 +5.6
Indiana +3.3 426.38 -3.0
Detroit +2.8 304.77 -5.7
New Orleans +2.7 486.68 -3.2
Oklahoma City +2.6 1291.28 +4.5
Washington +2.4 163.07 -5.0
New York +0.9 447.23 -4.4
Chicago +0.6 451.82 -1.9
Houston -0.8 823.65 -0.2
Charlotte -1.5 454.07 +1.6
New Jersey -3.7 301.55 -9.4
Philadelphia -4.5 253.72 -3.9
Cleveland -5.2 384.82 +7.8
Sacramento -9.4 116.88 -4.3
Toronto -12.8 369.83 -0.6
Golden State -17.0 124.37 -3.2
Minnesota -21.0 295.52 -8.7

Statistical Noise: The total minutes played by each team's primary lineup ranges widely, from 1474.80 (your Memphis Grizzlies) to a laughably small 124.37 (Don Nelson's Golden State Warriors, whose most-used lineup last season included Mikki Moore; no joke), and the reliability of the insights to be gleaned becomes wobblier as the chunk of time gets smaller. Which makes looking at these numbers pretty meaningful for the Grizzlies, whose most-used lineup took the floor considerably more than any other team's.

But some caveats are needed for a few other teams: The real outlier on this list is obviously the Cleveland Cavaliers, who won 61 games but whose most-used lineup (Mo Williams/Anthony Parker/Lebron James/J.J. Hickson/Shaquille O'Neal) was the fifth worst in the league. That's no typo. It suggests that Cleveland could have been even better if they'd admitted defeat on the Shaq deal sooner. The team's second-most-used lineup (239.13) substitutes trade acquisition Antawn Jamison for Shaq and would be safely in the Top 10.

The Dallas Mavericks' league-best primary lineup, if you're interested, consisted of Jason Kidd/Jason Terry/Shawn Marion/Dirk Nowitzki/Erick Dampier. And the Mavs were pretty terrific with Juan Jose Barea substituting for Terry. (It seemed to be the small lineups with Drew Gooden in the middle that fouled up the Mavs.)

The San Antonio Spurs presumably preferred lineup (Tony Parker/Manu Ginobili/Richard Jefferson/Antonio McDyess/Tim Duncan) managed to take the floor together for only 64 minutes. That lineup's +/- is better than any of the lineups listed, but the sample size is small.

The Milwaukee Bucks' primary lineup (Brandon Jennings/John Salmons/Carlos Delfino/Luc Richard Mbah a Moute/Andrew Bogut) performed surprisingly well. The sample size is fairly small, but looking at other lineups, the Bucks seemed to be really good with any unit that included Jennings, Bogut, and Delfino.

The Toronto Raptors played slightly more minutes (369.83) with Jarrett Jack at the point than the hobbled Jose Calderon (334.68) in an otherwise stable primary lineup (DeMar DeRozan/Hedo Turkoglu/Chris Bosh/Andrea Bargnani). They were better with Calderon and would have been several spots higher if I'd used that lineup instead.

Finally, the Minnesota Timberwolves were bad with pretty much any lineup, but, unsurprisingly, were a lot more competitive with their second most-used lineup, which paired Kevin Love with Al Jefferson up front, than with their most-used lineup, which featured Ryan Hollins rather than Love.

There is one other outlier on the list, but I will wait and bring them back up in relation to the Grizzlies.

What It Means for the Griz: What does this spreadsheet tell us about what the Grizzlies did last season and their prospects for the coming season?

The Grizzlies were tied (with Toronto) for the league's 18th-best record at 40-42, but the team's primary lineup (Mike Conley/O.J. Mayo/Rudy Gay/Zach Randolph/Marc Gasol) ranked 11th with a per-48 minute plus/minus of +7.3. Griz starters performed better than their equivalents from 53-win teams in Denver, Atlanta, and Utah and 50-win Oklahoma City, and much better than the primary lineups of other 40-something win teams such as Charlotte, Houston, Chicago, and Toronto.

The Oklahoma City comparison is probably most interesting. While the Griz starters played the most minutes of any unit in the league at 1475, the Thunder starters were second in the league at 1291, so that's a meaningful comparison. And even with Kevin Durant emerging as a superstar, the Grizzlies starters were significantly better — nearly 5 points better per 48 minutes. So why did the Thunder win 10 more games than the Grizzlies? The Thunder were able to go to their bench, substituting James Harden for Thabo Sefolosha or Nick Collison for Nenad Krstic with little, if any, dropoff, and were able to play deeper reserves Eric Maynor and Serge Ibaka without their team falling apart. The Thunder was actually a little better — +4.5 — without their primary lineup.

By contrast, the Grizzlies were generally terrible without their full starting five on the court. While the Grizzlies were the 11th best team in the league with the starters on the floor, they were the fourth worst when going to the bench. Only the Nets (12 wins), Timberwolves (15), and Clippers (29) were worse than the Grizzlies -6.7 per 48 minutes without their primary lineup.

The Clippers, incidentally, were the most similar team to the Grizzlies when it came to this starters vs. bench divide, the only other team better than +5 with its main lineup and worse than -5 without it. The Clippers were actually better than the Grizzlies (+10.5) with its most common lineup (Baron Davis/Eric Gordon/Rasual Butler/Marcus Camby/Chris Kaman) and even worse without it (-8.3). But the Clippers, due to injury and an in-season trade of Camby, only had that unit on the floor for 407 minutes.

The lesson of all this for the Grizzlies next season is perhaps an obvious one, but these numbers confirm it. For the Grizzlies to be a playoff team, they essentially need two things to happen: They need the starters (all returning, all in or just entering their prime) to stay reasonably healthy and they need the bench to be better. The Grizzlies are not far from being a 50-win caliber team. But meeting these two requirements won't be easy.

Last season, Grizzlies starters missed a combined 18 games — 13 by Gasol, 2 by Conley, 2 by Gay, 1 by Randolph. A smart bet would be on the over for 18 this season. Randolph alone averaged 17 missed games a year over the previous four seasons. But even if the Grizzlies starters stay reasonably healthy, the team shouldn't expect the unit to log more than 1400 minutes together again. Last season, Grizzlies starters played nearly 200 more minutes together than any other unit. And in the previous two seasons, the most used lineups in the NBA totaled 981 (Celtics) and 1209 (Hornets) minutes. With Gasol and Gay coming off tours-of-duty in the World Championship tournament, the Grizzlies might want to lessen their load even if they stay healthy. Reasonably, the Grizzlies shouldn't expect to get more than 1200 minutes from the starters this year, and even that's optimistic.

So, if the bench torpedoed the team's playoff hopes with the starters playing an astronomical 1475 minutes together, getting passable play from the reserves will be even more important this season with the starters likely to play a few hundred fewer minutes. And the Grizzlies enter the season with massive question marks among its reserves yet again. Free-agent acquisition Tony Allen is a playoff-proven veteran. But Allen's only averaged more than 20 minutes or scored more than 10 points a game once in his career. As for the rest of the bench, it is once again made up almost entirely of unproven players on rookie contracts. There's potential there, particularly with Xavier Henry, Hasheem Thabeet (no, really), and Sam Young, but not much real reason for confidence. But I’ll get into more detail on the range of possible for the Grizzlies' bench in later posts.

Keep the Flyer Free!

Always independent, always free (never a paywall),
the Memphis Flyer is your source for the best in local news and information.

Now we want to expand and enhance our work.
That's why we're asking you to join us as a Frequent Flyer member.

You'll get membership perks (find out more about those here) and help us continue to deliver the independent journalism you've come to expect.

Comments (9)

Showing 1-9 of 9

Add a comment

Add a comment