The Midrange Theory

1. Basketball Analytics: Augmenting Understanding, Not Replacing Artistry

All that analytics is, at its core, is a way to show how the things that people have been searching for in players and teams for decades can be distilled into statistics.

Beyond the numbers. Analytics isn't about reducing basketball's artistry to spreadsheets or replacing experienced judgment; it's a mode of thought that uses data to enhance understanding, challenge cognitive biases, and help decision-makers be "less wrong" in an uncertain environment. It augments, rather than dismisses, the hard-won expertise of professionals, fostering constant curiosity about the game.

Context is everything. The influx of data has irrevocably changed basketball, from high school offenses mirroring NBA strategies to the widespread adoption of advanced metrics like offensive and defensive ratings. However, data without context is meaningless; understanding the unique circumstances of each season, game, or player performance is crucial to avoid misinterpretations and derive actionable insights. For example, the 2020-21 season's unique context due to COVID-19 means its data might not be directly comparable to previous years.

Aesthetics and ethics. While data can optimize competitive advantage, it must also serve the aesthetic and entertainment aspects of the game; if analytically-indicated styles turn off fans, rules should change. Furthermore, the ethical implications of commodifying players and the demographic imbalance in analytics departments highlight the need for inclusivity and trust-building, ensuring the discipline broadens understanding rather than being seen as exclusionary.

2. Stats as Language: Decoding the Game's Complexities

Numerical and probabilistic information is hard for the human mind to take in and process in that form.

The power of narrative. Just as chess masters recall game positions, basketball statistics gain power when they become a language, tying numbers to familiar narratives. The traditional box score, with its points, rebounds, and assists, serves as a shorthand, but these abstractions don't always tell the whole story. New technologies, like player tracking data, demand a richer vocabulary to capture the game's dynamic events.

From dots to dialect. Player tracking data, initially just moving dots on a screen, becomes meaningful through "fancy counting" and machine learning. Algorithms are taught to recognize basketball actions like pick-and-rolls, cataloging variables and outcomes. This process distills complex movements into a statistical taxonomy, allowing for precise descriptions like "six dribbles, five off-ball screens, four passes, three points scored, and two defensive switches," which combine detail with conciseness.

Qualitative meets quantitative. The perceived dichotomy between the "eye test" and "the numbers" is often false; even qualitative assessments involve numerical comparisons, and "the numbers" themselves are products of qualitative choices about what to measure and how. The challenge lies in finding the right words and concepts to communicate these insights, ensuring that data-driven understanding is shared and easily grasped by everyone in the basketball world.

3. Beyond the Box Score: Measuring True Player Impact with Advanced Metrics

RAPM’s approach is to measure, once we consider the circumstances in which the player appears, whether his team performs better or worse with him on the floor, and by how much?

Individual vs. team stats. Traditional box score stats often misattribute credit, conflating a player's individual output with their actual impact on team success. For instance, a player's defensive rating reflects the team's performance while they are on the court, not their individual defensive prowess. This distinction is crucial, as being the "fifth wheel" on a great team can inflate individual plus/minus, while a star on a struggling team might appear less impactful.

Adjusting for context. To overcome the limitations of raw plus/minus, advanced metrics like Regularized Adjusted Plus/Minus (RAPM) were developed. RAPM uses regression analysis to estimate a player's average impact on victory margin by accounting for teammates, opponents, and home-court advantage. This method reduces "lineup noise" and overfitting, providing a more robust estimate of a player's effectiveness in their given role, as seen with Jimmy Butler's high RAPM despite modest traditional stats.

Explainability vs. precision. While RAPM-style metrics offer powerful insights into overall player impact, they often lack explainability, struggling to pinpoint how a player achieves their impact. This led to the development of "statistical plus/minus" (SPM) models, which use box score and tracking data to derive weights for various actions. These hybrid models, like ESPN's Real Plus/Minus (RPM), offer a more granular view, but still require careful interpretation, as a stat like "steals" might proxy for general basketball IQ rather than just defensive ability.

4. Goodhart's Law: Why Chasing Stats Can Undermine Winning

When a measure becomes a target, it ceases to be a good measure.

Playing the drill. Goodhart's Law highlights a critical pitfall in using statistics: when a metric becomes a goal in itself, it loses its reliability as an indicator of true performance. Players, responding rationally to incentives, might prioritize accumulating specific stats over actions that genuinely contribute to winning. This "playing the drill" phenomenon can distort the meaning of seemingly impressive numbers.

Empty calories vs. added value. Russell Westbrook's triple-double seasons exemplify this; while statistically remarkable, his pursuit involved actively "gobbling" uncontested defensive rebounds, which, while adding to his totals, didn't significantly enhance his team's overall defensive rebounding efficiency. In contrast, a teammate like Steven Adams, despite lower individual rebound numbers, often had to "earn" his boards, providing more marginal value to the team's defensive efforts.

Incentives matter. Contractual bonuses, like Maurice Harkless's three-point percentage clause, can also lead to stat-chasing that works against team goals; Harkless intentionally stopped shooting threes to secure his bonus, becoming an "anti-shooter" and potentially hurting his team's spacing. These examples underscore that statistics are indicators, not the goal itself, and when incentives shift from winning to individual numbers, the data becomes less trustworthy.

5. The Three-Point Revolution: A Strategic Evolution, Not a Decline in Skill

Rather than sacrificing dunks and layups to facilitate threes, a good chunk of the jumpers the league was already taking have simply become jump shots from a little farther out.

The math of 3 > 2. The explosive growth in three-point shooting is often misunderstood as a decline in basketball artistry or a shift away from attacking the rim; however, the data reveals a more nuanced evolution. The fundamental insight that three points are worth more than two, even if slightly less accurate, has driven a strategic realignment of offensive principles, forcing defenses to guard a much larger area of the court.

Midrange migration. The "three-point revolution" hasn't eliminated shots at the rim; their proportion has remained remarkably stable over decades. Instead, the decline has come almost exclusively from long midrange attempts—two-pointers outside the lane. These "in-betweeners," particularly assisted ones, have largely been replaced by three-pointers, meaning the league isn't taking more jumpers, but rather taking them from a more efficient distance.

Rule changes and stretch bigs. This shift was accelerated by rule changes, notably the liberalization of zone defense in 2001 and the re-emphasis on "freedom of movement" in 2004. These changes made post-ups less effective by allowing defenses to "shrink" the court and made perimeter creation more valuable. This, combined with the rise of "stretch bigs" who can shoot from beyond the arc, has forced defenses to spread out, creating more space for superstars to operate and showcase their "midrange artistry" when needed.

6. Chasing Ghosts: The Enduring Challenge of Defensive Measurement

How do we count all the things that don’t happen?

The off-ball dilemma. Measuring individual defensive impact remains the "holy grail" of basketball analytics, largely because defense is inherently collaborative and often involves actions away from the ball. Unlike offense, where primary ball-handlers drive much of the action, all five defenders are typically involved in a possession, and success often means preventing an opponent from getting an open look in the first place, which is difficult to quantify.

Context and scheme dependence. Defensive statistics are heavily influenced by coaching schemes and team context; a player's defensive numbers might reflect their team's strategy (e.g., Syracuse's zone defense inflating guard steals) rather than pure individual ability. Assigning credit or blame for defensive breakdowns is complex, as the "fault" for an open shot often lies in an earlier, off-ball action or a systemic failure in help and rotation, not just the closest defender.

Measurable vs. unmeasurable. While metrics like RAPM can estimate a player's aggregate defensive impact, they struggle to explain how that impact is achieved. Some defensive aspects are more measurable: rim protection, for instance, shows consistent individual effectiveness in deterring or contesting close-range shots. However, "three-point percentage defense" is largely noisy and unreliable at the individual level, and "visible efforting" can be mistaken for effective play, highlighting the need for deeper understanding of defensive tradeoffs.

7. Mind the Cap: Roster Construction as a Multi-Dimensional Puzzle

The desirability of acquiring a player is almost as much a function of their salary and contract status as it is their ability and on-court fit within a team.

The salary cap's iron grip. Building a championship-caliber team in the NBA is a complex optimization problem constrained by money, minutes, and roster spots. The salary cap, which guarantees players 51% of Basketball Related Income, forces teams to prioritize salary efficiency. Every dollar spent on one player reduces what's available for another, making accurate player valuation—not just in terms of on-court production but also contractual value—paramount for competitive success.

Valuing wins and marginal gains. Player value can be quantified by estimating the "wins" a player adds, typically by converting their plus/minus impact into a dollar figure; for instance, a player adding 3.6 wins might be worth around $10.4 million. However, wins are not additive; the marginal value of adding a win decreases as a team gets better, making it easier to add wins to a bad team than to upgrade a solid playoff squad, as the "replacement level" bar rises.

The Bird Rights Trap and superstar scarcity. Provisions like "Bird Rights," designed to help teams retain stars, can inadvertently lead to the "Bird Rights Trap," where teams overpay to keep players, sacrificing future financial flexibility. This, combined with the extreme scarcity of true superstars (only 3-7 players per season are "supermax" level), means that finding and retaining elite talent, often through the draft, is the most critical factor. The "Cap Spike" of 2016, which allowed the Warriors to sign Kevin Durant, was a rare confluence of circumstances that temporarily circumvented these systemic constraints.

8. The NBA Draft: A Wicked Problem of Prediction and Bias

The difficulty in post facto evaluation of a draft pick goes well beyond injury luck.

A high-stakes crapshoot. The NBA draft is an intensely difficult endeavor, characterized by more failure than success; despite immense effort and data, a high pick offers no guarantee of a superstar, with fewer than half of draftees making a meaningful impact. This "math problem" makes the draft a "wicked" environment for decision-making, lacking the regularity, practice, and immediate feedback necessary for reliable intuition.

Tools vs. feel. Prospect evaluation often overvalues "tools"—physical measurables like wingspan and raw athleticism—at the expense of "feel for the game," which encompasses perception, quick situational evaluation, and proper execution. A player might look impressive in a workout, but if they lack the "connective tissue" to make micro-decisions effectively in game action, their tools are wasted, as seen in players with great physical gifts but limited on-court impact.

Bias and comps. The draft process is rife with cognitive biases; confirmation bias can skew the interpretation of qualitative information (e.g., a player's "leadership" vs. "being an asshole"). Player comparisons ("comps") are particularly destructive, as they anchor evaluators to an idealized version of a past player, ignoring the unique developmental path and survivorship bias. Overcoming these biases requires a disciplined process, combining quantitative models to narrow the prospect pool with qualitative scouting to assess elusive traits like feel and character.

9. The Power of Small Edges: How "Little Things" Drive Big Wins

The inches we need are everywhere around us!

Marginal gains in a competitive league. The NBA is incredibly competitive, with many games decided by a few possessions; while "game-changing" discoveries grab headlines, consistently chipping away at small margins can accumulate into significant impacts over a season. Identifying and optimizing these "little things"—seemingly trivial adjustments or strategies—is where an analyst can make subtle but substantial differences, ultimately swinging games and even playoff series.

Optimizing possession value. Examples of these small edges include "splitting the pair" at the free-throw line (calling a timeout between attempts to "ice" a shooter), which can lead to an extra miss every 20 times. Another is the nuanced trade-off between offensive rebounding and transition defense; while often conservative, teams can find specific situations where crashing the glass for an extra possession outweighs the risk of giving up a fast break. Even avoiding turnovers, particularly "live ball" ones, significantly boosts offensive efficiency and prevents defensive compromises.

Timing and end-game strategy. Strategic timing, like the "two-for-one" at the end of quarters, aims to gain an extra possession, but its value depends on execution and shot quality. More critically, end-of-game scenarios, often subject to fallacies like the "quick two," can be optimized using win probability models. These models reveal that playing for a tie or a "safe" two-pointer often dramatically reduces win probability compared to taking a three-pointer for the win, even if it's a lower-percentage shot.

10. The 82 and The 16: Mastering Playoff Game Theory

In the playoffs his first move is gone…whatever they know he does well is gone.

The playoff transformation. The NBA playoffs represent a fundamentally different competitive environment than the 82-game regular season; with fewer games, more rest (in early rounds), and intense opponent-specific preparation, the game changes. Depth becomes less critical, rotations tighten, and the overall level of play ratchets up, highlighting the distinction between "82-game players" who thrive in the regular season grind and "16-game players" who elevate their performance when it matters most.

GTO vs. exploitative play. Regular season strategies lean towards Game Theory Optimal (GTO) play, maximizing performance against a wide range of opponents. In the playoffs, however, teams adopt highly exploitative strategies, meticulously dissecting and targeting an opponent's specific weaknesses over a multi-game series. This means a star's "first move" might be neutralized, forcing them to dig deeper into their repertoire, as seen with players like Robert Horry thriving while one-dimensional scorers like Carlos Boozer struggled.

Talent, role, and the "rubber band effect." Playoff success isn't just about raw talent; it's about how that talent fits a role and performs under intense scrutiny against top competition. The "rubber band effect," where large leads tend to shrink in the regular season, is significantly weaker in the playoffs, leading to more blowouts as coaches keep their stars on the floor longer to secure wins. Ultimately, building a playoff-ready roster means acquiring versatile players who can adapt, execute, and maintain their effectiveness when opponents are specifically trying to take away their strengths.