The Credit Goes to the Wrong People
- Bruin Sports Analytics
- Mar 27
- 10 min read
By: Van Avanesian, Ricky Cai, Jayden Kim, and Justin Herr
Two seasons. Eight Western Conference teams. One question: when you measure not just what happened while a player was on the floor, but how much of the offense he was running and how efficiently he ran it — does the player the league celebrates match the player the data identifies? Usually, it doesn’t
WESTERN CONFERENCE TOP 8 · TWO-SEASON EFFICIENCY ANALYSIS
Kawhi Leonard played 68 games last season. He averaged 23.5 points. He was the best player on a team that won 50 games. And the dominant frame around him, in every national conversation, was durability: injury history, load management, will he make it to May. What rarely came up was an honest accounting of what he actually does when he’s out there.
By the measure in this analysis, Kawhi Leonard posted +47.0 in 2024–25 — the second-highest score in the entire dataset, behind only Nikola Jokić. He came back in 2025–26 and posted +41.0. Still second. Different opponents, different supporting cast, same result. Two full seasons. The conversation around him has not caught up to this.
He is not the only one. Across eight teams and two seasons, this analysis found a consistent gap between the players the league treats as engines and the players who are actually running these offenses. Some of the gaps are small. Several are startling. A few of them should change how you think about the franchises you believed you understood.
How we measured this
Most stats measure output — points, assists, rebounds. This one measures impact: when a players was on the floor, did the team’s offense get better or worse? We built a single number from five components, all measured per 100 possessions so a player logging 15 minutes and one logging 35 are judged on equal footing.
Offensive Points Differential: Points the team scores per 100 possessions with you on the floor versus off it.
Assist Differential: How much ball movement improves. Captures the pass before the assist, the screen that freed the cutter — playmaking that never shows up in a box score.
Defensive Points Differential: Points the opponent scores per 100 possessions with you on the floor. Subtracted from your score. Offensive production does not cancel out defensive damage.
Turnover Differential: Weighted at 2.3×, because a turnover hands the opponent a transition opportunity on top of the lost possession.
Those four produce a Net Impact Score. But a player running 12 percent of possessions and one running 30 percent can post the same net number for very different reasons. Hence, we add one more layer:
Usage × True Shooting Bonus (USG% × TS% × 100): Usage rate multiplied by true shooting percentage, then multiplied by 100, and added to the net impact score. Rewards players who are carrying a large share of the offense and converting it efficiently. Penalizes players who dominate the ball without producing.
In 2024-25, Kawhi's net impact was already strong. He was running 28 percent of the Clippers’ offense at 58.9 percent true shooting. The bonus: +16.5 points. He finishes at +47.0. One number that accounts for both what happened on the floor and the load he was carrying to make it happen.
Minimum threshold: 15 games played. Net impact data from PBP Stats (pbpstats.com). True shooting percentage and usage rate from NBA.com.
Jokić: The Number That Makes Everything Else Legible
Start here, because everything else in this dataset needs a reference point. Nikola Jokić’s 2024–25 Efficiency Score is +66.1. The next-closest player is Kawhi Leonard at +47.0. That is a 19-point gap between first and second place in a conference of elite offensive players. It is the largest margin between any two adjacent scores at the top of the rankings. In a sport where single-digit differences are considered meaningful, Jokić is playing a different game.
His 2025–26 score fell to +45.1. This, in the context of this dataset, as a decline. It still led the conference. It still led Denver by double digits. The decline is real only in the sense that the 2024–25 number was historically anomalous — and +45.1 is not.
Denver’s depth is the secondary story, and the hardest one to interpret cleanly. Christian Braun scored +42.8 in 2024–25. Jamal Murray +29.7. Michael Porter Jr. +24.0. Aaron Gordon +23.7. Four players posting elite scores on the same roster. Whether those numbers reflect genuine individual impact or whether they are, in part, the product of playing alongside the most dominant offensive player in the dataset, this metric cannot fully separate those effects. What it can say is that the offense ran well for all of them, consistently, across two seasons. That is either a historically deep roster or the most powerful single-player gravitational field in modern basketball. The honest answer is probably both.
The Team That Was Never Given Credit
In 2024–25, Los Angeles placed four players in the top 18 scores in the entire conference: Kawhi at +47.0, Ivica Zubac at +36.0, Norman Powell at +31.5, James Harden at +24.1. Four players, same roster, same season. The Clippers were discussed as an injury-dependent playoff team. The data describes a top-tier offensive organization with multiple elite contributors at every level of the rotation.
Zubac at +36.0 is the single most counterintuitive number in this study. He runs a modest share of possessions, converts them at 64.1 percent true shooting, and the Clippers’ offense ran measurably better with him on the floor in both seasons. He is not in any serious conversation about the conference’s best centers. By this measure, he is a top-five offensive contributor in the Western Conference. The gap between his actual impact and his public perception may be the largest of any player in this dataset.
Harden at +24.1 is the finding that most directly confronts received wisdom. He was written off — aging, declining, a player whose best years were clearly behind him. His number says the Clippers’ offense ran significantly better when he was on the floor than when he wasn’t. He maintained that in 2025–26 at +24.3. Whatever the league decided about James Harden, his offense did not get that memo.
Oklahoma City: One Player, One Championship
Shai Gilgeous-Alexander posted +35.7 in 2024–25 and +34.3 in 2025–26. He ran 33 to 34 percent of Oklahoma City’s possessions at 63 to 67 percent true shooting. His two-year consistency — the same elite score at elite usage in back-to-back seasons — is matched in this dataset only by Jokić and Kawhi. Three players, across 60-plus tracked athletes, managed it. Shai is one of them.
In 2024–25, no other Thunder player cracked the top 15 in the conference. OKC won the championship. That is, in part, a story about defense and depth and coaching. But it is also a story about how wide a margin one exceptional offensive player can create — wide enough to carry teammates whose numbers are nowhere near his into a title. In 2025–26, the supporting cast improved: Alex Caruso posted +22.5, Ajay Mitchell +19.5. Shai held. OKC held the top seed.
The number worth sitting with: Jalen Williams posted +5.7 in 2024–25 and +9.3 in 2025–26. Both positive. Both years in a championship-caliber system. And yet: Williams is discussed as Shai’s co-star — equivalent billing, near-equivalent usage, major contract. His number is that of a strong complementary player. OKC’s offense ran around Shai. Williams contributed. The championship narrative framed them as co-leads. This data does not support that framing.
The Players Who Were There the Whole Time
Alperen Şengün posted +31.4 for Houston in 2024–25 — ninth in the conference — running 25.9 percent of the Rockets’ possessions at 54.5 percent true shooting. The national conversation about Şengün is a development story: can he become a star, when will he put it all together, is he the real deal? The data says that the conversation started a year too late. By the only measure that actually counts — did the offense run better when he was on it — Şengün was already a star. Houston finished as a playoff team with a young roster and no obvious star-level talent by conventional evaluation. The talent was there. It just wasn’t being evaluated correctly.
Dorian Finney-Smith posted +25.6 for the Lakers in 2024–25, third on the roster behind Luka Dončić and Austin Reaves. Every piece of public analysis on Finney-Smith describes him as a 3-and-D piece — a role player, a stretch-and-defend guy, interchangeable with a dozen other players in the league. This number says something categorically different. The Lakers’ offense ran dramatically better when he was on the floor. He was traded. The Lakers’ offense declined noticeably in his absence — none of the players who absorbed his minutes approached the same impact level when measured the same way. That is a loss the Lakers almost certainly did not price correctly because the label — 3-and-D piece — made his value invisible.
Gui Santos posted +21.4 and +15.5 across both seasons for Golden State. In those same two years, Draymond Green fell from +23.1 to +5.2, Podziemski from +19.1 to +6.6, and the Warriors fell from the second seed to the seventh. Santos held. He was the most stable offensive contributor on his team across two seasons. He does not appear in trade analysis or contract discussions or most coverage of Golden State. He has been there, quietly making the offense work, the entire time.
The label a player carries — role player, 3-and-D piece, promising young big — determines how his contributions get evaluated. Several of these labels are wrong. The players are paying for it.
What One Summer Can Do
The most instructive case for understanding year-over-year change is Golden State, because the change is large and the circumstances are controlled. In 2024–25, Draymond Green posted +23.1. Same role, same coach, same system, one year later: +5.2. A drop of nearly 18 points. Curry fell from +31.8 to +23.9. Podziemski from +19.1 to +6.6. The one player who improved was Jimmy Butler III — the one new player. The Warriors fell from a top-two seed to seventh. The standings eventually confirmed what the numbers were already showing.
Memphis is the mirror image. Zach Edey posted +3.2 in 2024–25 — not a liability, but not a contributor either. One summer later: +30.9. A swing of 27.7 points, the largest individual year-over-year improvement in this entire dataset. Memphis did not trade for a star. They did not sign a difference-maker in free agency. They developed a player who was already on the roster from unremarkable into one of the ten best offensive contributors in the conference. That is the kind of internal development most front offices say they prioritize and rarely achieve at this scale.
Denver and Minnesota are the quieter versions of the same story — and the more important one for what it says about roster building. Peyton Watson went from −3.6 to +19.1. Aaron Gordon from +23.7 to +33.2. Minnesota’s Julius Randle went from +9.0 to +25.4. Donte DiVincenzo from +18.1 to +28.5. Both teams got dramatically better by doing nothing externally and everything internally. Golden State spent the summer making one big addition and got worse everywhere except where the addition played. Denver and Minnesota stood pat and got substantially better. The offseason narrative almost always favors the active team. This data suggests the teams worth watching are the ones developing what they already have.
What Holds Up Over Two Years
One season of this data is a measurement. Two seasons, same player, same team, is evidence. Three players held elite scores in both years: Jokić, Kawhi, and Shai. All three led their teams offensively both seasons. None dropped below +34. Consistency at that level and at that score is not a product of sample size or circumstance. It is the clearest signal this dataset produces.
Ja Morant posted +11.2 in 2024–25 and +5.8 in 2025–26. Both years, his score was the lowest of any featured player on his own team — a team that featured Edey and other contributors who scored higher. His number is positive, meaning the metric does not rate him as a liability. But the gap between him and his teammates is the story. The player the franchise is built around is consistently the smallest offensive contributor among the players most evaluated alongside him. Two years of the same result on the same team is harder to explain away than one.
Then there are the players who swung hard in one direction and need a second season to interpret. Christian Braun at +42.8 in year one and +18.1 in year two. Alperen Şengün at +31.4, then +17.7. Both are still positive. Both are significant drops. For Braun, the two-year structure is clarifying: 2024–25 was a season in which Jokić was historically dominant, and his teammates floated on that tide. +18.1 in a more normal season may be closer to Braun’s actual level. For Şengün, there is no clear explanation yet. That is what a third season is for.
The players the league has already decided on — Jokić, Shai, Kawhi — are confirmed here. The more interesting exercise is applying the same scrutiny to the players the league is still making up its mind about.
What You Now Know That Most People Don’t
Kawhi Leonard is the second-most consistently dominant offensive player in the Western Conference. He has been, by this measure, for two straight seasons. The conversation treats him as a health risk with elite upside. The data treats him as a near-certain elite contributor when available, which he has been for the majority of two full seasons.
And Jokić is so far ahead of everyone else in this dataset that even his off-year — +45.1 in 2025–26 — leads the conference. The gap between him and the second-best player is larger than the gap between second and tenth. He is not playing at a different level. He is playing a different game.
The credit has been going to the wrong people. Two seasons. Eight franchises. Sixty-plus players. The pattern is consistent enough that it stops looking like a media failure and starts looking like a structural one — the way the league evaluates players, the labels it assigns, the narratives it builds around stars and role players, systematically undervalues certain kinds of contribution and overvalues others. The numbers have been there the whole time. The question is what gets built differently now that you’ve seen them.
Sources
PBP Stats — https://www.pbpstats.com Net on/off impact data: assist differential, offensive points differential, defensive points differential, and turnover differential (per 100 possessions)
NBA.com — https://www.nba.com/stats True shooting percentage (TS%) and usage rate (USG%) for all tracked players
A NOTE ON METHODOLOGY
Efficiency Score = Net Impact Score + (Usage Rate × True Shooting % × 100). Net Impact Score = Assist Differential + Offensive Points Differential − Defensive Points Differential − (Turnover Differential × 2.3), all per 100 possessions. Minimum threshold: 15 games played. Small-sample outliers excluded. Net impact data from PBP Stats (pbpstats.com). True shooting percentage and usage rate from NBA.com. Dataset covers rotation players from the eight teams that finished in the top eight of the Western Conference in 2024–25, tracked across both 2024–25 and 2025–26.
Comments