How Effective to Allocate Minutes for Starter and Bench Majority Lineups?
Giving huge minutes to starters has been an interesting topic throughout the years in NBA. In the last decade, Tom Thibodeau has been the number one coach who allocates his starters more than 40 minutes at any given night. In this post, I have tried to understand how teams deploy their lineups over the looong 2019–2020 regular season. Hopefully, it will give an insight on how much teams should give their starters during playoffs.
The inspiration for this post is raised by Thomas Bassine’s post on his website, titled as “How Important Is A Good Bench To NBA Success?”. As he opened his post with this piece by Seth Partnow(former Bucks Director of Basketball Research and now at The Athletic), I have tried to apply my knowledge to understand the coach decision’s as possible as I can.
I would like to explain the definitions of lineups I’ll be using to avoid any conflicts. For this analysis, I parsed lineups(with defined starter slates) from pbpstats.com and tried to analyze the outcomes. As Thomas already stated, I also considered lineups which have 3 or more starters as starter majority lineups and the rest(2 or less starters) as bench majority lineups. It is not a perfect distinction, but it serves the best for the sake of simplicity for this analysis.
The below table is representing the minute allocation by any given number of starters.
It is fair to say all of teams in the league tend to use their all starters with the biggest share, which is 27.0%. For starter majority lineups, teams allocated about 63.0% of the available minutes and for the bench majority lineups have been accounted for nearly 37.0% of the total minutes.
The interesting point in this table is the following: Teams played with almost all bench players(0 or 1 starter) more than they played with 2, 3 or 4 starters. Moreover, some teams(e.g LAC, LAL) used “bench mob” lineups more than their 5-starter lineups.
The deviation of minute allocation is mostly because of the strong bench which few teams have and/or the injuries occurred during the season.
- Atlanta Hawks and Charlotte Hornets were the dead-last in the proportion of total minutes for lineups consist 5-starters with 21.78% and 21.59%, respectively.
- Denver Nuggets, Portland Trail Blazers and Toronto Raptors gave the most minutes to their starters with 1316(37.08%), 1189(33.33%), 1199(34.49%).
The thing that really matters is how effective are those lineups and how can teams maximize their efficiency as some of them start to begin playoffs and some to plan next year(s). In order to quantify this, I checked their net points(lineup points/(100poss) — opponent points/(100poss) at each number of starters.
In above graphs, I tried to illustrate how teams fared in different number of starters on court and no surprises for 5-starter lineups as Bucks smashed everyone else with a whopping +21.3 points per 100 possession(NetRtg). The weird thing is the Clippers and the Celtics managed to maintain +14.94 and +10.63, respectively, when they have 2 starters on the court. For the Clippers, +14.94 mark ranks them at the 2nd place among the league in any given lineup combination and it is fair to say they are very well-equipped with not one but two 6th Man of the Year candidates. As for the Celtics, the important feature is the amount of time they have used exactly 2 starters on the court with 807 minutes which accounts 23.1% of total minutes. As playoffs around the corner, maximizing the amount of starters’ minutes would be an obvious move for the coaching staffs.
As Thomas mentioned in his article 1.5 years ago, the Spurs are strange again with their bench majority lineups surpassed the starter majority lineups as it is clearly represented their individual graph. Dallas Mavericks are the team with highest Net Rating(+12.42) where there are no starters on the court, but this is mostly because of the garbage time effect(nearly one third of their games were blowouts).
Although playing with starter majority lineups enable to take the competitive advantage, the importance of bench remains valid in the NBA. In below chart, the point spread between two types of lineups is shown. As I previously stated, starter majority lineups consist 3 or more starters and bench majority lineups have 2 or less starters.
The Bucks are at the top of starter majority lineups with 13.38 net points per 100 possession and the Heat are leading the bench majority lineups with 6.18 net point differential. Maybe Miami should use their regular season rotation in order to take an advantage oftheir strong bench unit, since they were the only team qualified for the playoffs in the top 7 for bench majority lineups.
On the other hand, the bottom 5 teams for net points from bench majority lineups were PHI, BOS, HOU, BKN and POR, respectively. It means especially for the Blazers, they should dominate the starter minutes to narrow the gap between their starter and bench majority lineups, which are +2.66 and -11.68 respectively.
Lastly, I tried to look at the situation where majority of starters play against majority of starters, in which every teams’ statistics when 3 or more starters on the court. With this approach the results were intriguing since 15 out of 16 playoff teams(only ORL has a negative net rating) have had positive net rating when their starter majority lineups face against starter majority lineups. The dominance of the Bucks proved one more as they finished with +13.93 net points per 100 possession when they use their starter majority lineups against opponents’ starter majority lineups. You can see the full list in here.
It is fair to say the contenders have distinctly placed almost every type of lineup setting, thanks to their star power and starter majority lineup play. But the remaining 37.0% of minutes are still valuable for teams to surpass their opponents with slight edges that can make the difference.
