Though coaching orthodoxy has long maintained that 2 points at any moment in a given game are equivalent, the choices of many NBA coaches at the end of close games this season would beg to differ. Perhaps LeBron James was right when he said “2 points ain’t 2 points.”
Across the league, many NBA offenses change in clutch time (which the NBA defines as any game time when there are five or fewer minutes remaining in the game and the scoring margin is within 5 points). But those changes aren’t uniform across the league. Most teams pass less and commit fewer turnovers, but the plays through which they achieve that are different. Every team increases the frequency with which it runs isolation plays. Some teams also run more pick and rolls or initiate out of the post. But the teams that win the most in the clutch, it turns out, often change the least.
Clutch minutes have a variety of caveats — most importantly, that there’s not a whole lot of them. The Los Angeles Lakers played the most clutch minutes up to this season’s All-Star break, at 159 — fewer than four games’ worth. Three-point shooting, for example, doesn’t stabilize until approximately 750 attempts, and only four teams had reached 100 attempts in clutch minutes in that same time period. As a result, clutch data from a single year can include unexpected results that certainly describe what has happened but don’t necessarily predict what will happen.
So what do we know about what has happened so far? Looking through the 5,932 clutch offensive possessions played through the All-Star break, we can separate these teams into four categories: those that increase their pick-and-roll frequency in the clutch, those that increase their post-ups, those that increase their isolations and those that change little. Some teams may fall into multiple groups, as an individual player can drive more post-ups and his teammates more isolations, for example. But most teams coalesce around a singular identity in clutch offense.
To separate the teams, we found the average leaguewide difference between play type frequency over full games versus in the clutch. Any team that fell outside a half standard deviation of that mean difference for an individual play type was considered to exist within that category. Teams could fall into multiple categories.
|Offensive Team||Clutch record||Overall||In Clutch||Increase|
|Los Angeles Lakers||17-18||58.45||68.09||16.50|
|Golden State Warriors||20-10||54.74||62.63||14.41|
Combined record through the All-Star break: 149-147, .503
The NBA is already a pick-and-roll league, but a number of teams turn up the dial in crunch time. Per Second Spectrum, pick and rolls are both the higher-frequency play and have higher points-per-chance averages than either post-ups or isolations in the clutch, so teams are only following the numbers.1 Pick-and-roll frequency has been similar in clutch time and full games over the full league, but that’s of course not true for every team. The Hawks, Hornets, Cavaliers, Grizzlies, Raptors and Warriors employ All-Star point guards who spend a sizable share of their time in the pick and roll. Their reliance on such plays only increases as the game gets close.
|Ball-handler||Team||Overall||In Clutch||Overall||In Clutch|
The pick-and-roll teams run by those star guards have been among the best within this group. The Pacers and Pistons have been among the worst teams in the clutch by points per clutch chance, and they don’t have the same caliber of guard running the show in the clutch.
Meanwhile, the Grizzlies and Wizards are among the class of this group. The Grizzlies ramped up their assault on the rim in the clutch; they drove at the sixth-highest rate, and they were in the top three in clutch points off turnovers, second-chance points, fast-break points and points in the paint. Morant’s usage rate was over 40 percent, and his self-creation is critical for the Grizzlies, as almost 90 percent of his clutch field goals were unassisted — only DeMar DeRozan surpassed him in self-creation among players with consistent clutch roles.
The Wizards, on the other hand, shot 32.8 percent from deep as a team, third-worst in the league. That jumped to 38.9 percent in the clutch, fourth-best. Their highest-volume shooter, Kyle Kuzma, saw his accuracy jump over 10 percentage points.
There are some near-certain playoff teams in this category; the Cavaliers, for example, have an excellent record on the season, yet they were below .500 in games that went to clutch time. Some of that could be attributable to their poor shooting luck; they had the third-highest shot quality in high or very high leverage situations2 yet the second-lowest effective field-goal percentage. The Warriors have suffered a similar cold streak in clutch time, with Steph Curry shooting 32.8 percent from the field in the clutch. Golden State scored the 10th-lowest points per chance in the clutch yet still managed an impressive 20-10 record.
|Offensive Team||Clutch Record||Overall||In clutch||Increase|
|Portland Trail Blazers||13-14||15.69||24.24||54.52|
|San Antonio Spurs||8-18||9.16||14.05||53.31|
|New York Knicks||13-18||19.68||29.81||51.46|
|Los Angeles Clippers||18-13||21.05||31.34||48.87|
Combined record through the All-Star break: 139-173, .446
Every team ran a higher rate of isolations toward the end of close games in the first half of the season, which means the teams here ran far more isolations down the stretch; on average, the leaguewide rate of isolations rose from 17.29 per 100 possessions to 23.94 in the clutch, per Second Spectrum.
The Hornets, Mavericks, Pistons, Timberwolves, Knicks, 76ers and Kings all had longer average times of possession in the clutch versus full games, as isolation attempts could serve to slow down offensive possessions. Conventional wisdom is that teams play slower to limit turnovers, and turnovers did for the most part drop across the league in the clutch. Isolations saw fewer passes per possession in the clutch as well as lower turnover rates compared to pick and rolls or post-ups, per Second Spectrum. They were also the likeliest to see a ball-handler shoot rather than pass, so if teams want their best scorers to go it alone and get up a shot with no chance of a turnover, isolations are the way to go.
Luka Dončić used the majority of his team’s clutch isolations, deciding Dallas’s fate both in full games and the clutch. Similarly, between Feb. 8 and the All-Star break, when the Trail Blazers featured Anfernee Simons without either Damian Lillard or CJ McCollum, Simons was second in the league in clutch isolations with a frequency higher than 10 teams.
|Ballhandler||Team||Overall||In Clutch||Overall||In Clutch|
|Anfernee Simons||Trail Blazers||6.03||12.04||0.878||0.750|
Reggie Jackson has taken over as the Los Angeles Clippers’ closer with Paul George and Kawhi Leonard injured. He’s long been a low-frequency pick-and-roll guard, so when his usage rate jumps, so too do his isolations. Anthony Edwards is one of the premier drivers in the game, so isolations are intended to give him more room to attack. Cole Anthony, De’Aaron Fox and Dejounte Murray saw big leaps to their usage rate in the clutch, with all three clutch-time usage rankings falling in the top 10 in the league. Each used those extra possessions to approximately double his rate of isolation attempts.
While some teams recorded the increase in isolation frequency mainly from individual players, others received contributions across the board. The Hornets saw increased isolation attempts from far more than just Miles Bridges; while he was their isolation frequency leader over full games, both LaMelo Ball and Terry Rozier had higher frequencies of isolations in clutch time.3
|Offensive Team||Clutch Record||Overall||In clutch||Increase|
|New Orleans Pelicans||10-14||8.92||11.52||29.11|
Combined record through the All-Star break: 76-97, .439
Post-ups record a larger drop in efficiency in the clutch than pick and rolls or isolations. Perhaps as a result, among this group only the Pelicans scored within the top-10 highest efficiency in the clutch. Still, there are advantages down low to be mined when games slow down toward the end. The Mavericks and Pelicans have similar archetypes leading their rosters, with Dončić and Brandon Ingram functioning as enormous ball-handlers who can create for teammates and shoot over defenders in the post. Neither, however, has been very successful.4
Nikola Jokić and Joel Embiid are the premier centers in the league today, both over 7 feet tall with masterful scoring and passing skills. Jokić and Embiid were first and second in number of post-ups this year, per Second Spectrum, and they were also both in the top seven among high-volume players for efficiency in such scenarios.5 Yet the Nuggets and 76ers have seen their offenses become less impactful in the clutch. Both teams have slightly lower points per chance in the clutch than over full games. Both were above .500 in games that reached clutch time despite ranking outside of the top 10 in efficiency. To be fair, the Nuggets shot only 22 percent from deep in the clutch. Their defense has been perhaps more significant, as Jokić has three game-winning blocks at the buzzer. And with James Harden now a 76er, he may absorb some of Embiid’s clutch usage and shift the team away from the post.
|Change in Clutch Time|
|Offensive Team||Clutch record||Picks||Isos||Posts|
|Oklahoma City Thunder||13-18||-22.96||+14.86||-37.92|
Combined record through the All-Star break: 140-116, .547
This is a group full of contenders, whether you look solely at performance in the clutch or overall. The two best teams in the East and the best in the West both reside here, as does last year’s champion. Just because these teams didn’t up the ante in clutch time doesn’t mean they didn’t shift more subtle aspects of their offenses. For the Bucks, Giannis Antetokounmpo’s usage actually dropped slightly in the clutch, as Jrue Holiday and Khris Middleton initiated more often. For Miami, Kyle Lowry’s usage jumped while Bam Adebayo’s dropped, perhaps for the same reason as the Bucks turned to Middleton and Holiday instead of Antetokounmpo — having capable pull-up shooters initiate plays is a safe option. DeRozan is winning fanfare as the league’s best closer and made two game-winning buzzer-beating jumpers on consecutive nights for the first time in NBA history.
Utah’s offense was not the problem in the clutch, as it scored at the seventh-best rate. It was already the highest-frequency pick-and-roll team in the league, so it had few spare possessions to convert to that play type. Mike Conley and Donovan Mitchell split clutch duties there fairly equally, and both were effective.6 While the Celtics were one of the best teams in basketball heading into the All-Star break, their middling clutch offense was largely driven by Jayson Tatum shooting 35.3 percent from the field. And the Rockets and Thunder had better records in the clutch than their overall marks. Kevin Durant took over in the clutch for the Nets, with his usage rate seeing the fifth-largest leap in terms of percentage points among team leaders, and he used possessions out of the pick and roll, isolation and the post at rates nearly identical to those in full games.
There are other exceptionalities for the teams in this category. The Celtics, Bulls, Bucks and Thunder are four of the five teams that played a half standard deviation below the mean for average time of possession, meaning they play faster in the clutch while most of the league slows down.
The Suns, though, are the real story when it comes to clutch basketball. They obliterated everyone in the clutch. Their net rating of +41.2 — in an admittedly miniscule 85 minutes7 — is the highest in the NBA Advanced Stats database, which extends to the 1996-97 season.8
Though most teams’ assist rates dropped in the clutch, the Suns’ actually increased. Chris Paul was the only player with more than two clutch games played who threw a full assist per clutch game. The Suns averaged 1.361 points per chance when Paul ran pick and rolls in the clutch, and he was virtually equally effective with any pick-and-roll partner. The Suns can cycle through looks to attack the weakest defender on the floor — a popular half-court offensive approach in the playoffs. The Suns scored a nonsensical 1.564 points per chance in the 76 clutch possessions with more than one pick and roll involving Paul, per Second Spectrum.
Devin Booker may be one of the premier isolation scorers in the league, but he only attempted 10 in clutch time through the first half of the season. Phoenix’s rate of isolation hardly budges down the stretch, with one of the smallest increases of any team. Clutch offense in Phoenix belongs to Paul, and he changes little. He will be out for the next several weeks with a fracture in his thumb, but he should be back by the playoffs.
Being the best in clutch time doesn’t guarantee anything come playoff time. But it doesn’t hurt, either. Each of the final three games of the 2021 playoffs — all Bucks wins — featured clutch minutes that the Bucks won. This year, the Suns will have a huge advantage if their record-setting dominance in close games continues going forward. If the rest of the league wants to catch up, perhaps they ought to mimic Phoenix in changing little come clutch time and simply playing offense the way they had all game.
Check out our latest NBA predictions.