Chapter 1 — What the Scoreboard Leaves Out

What the Scoreboard Leaves Out

The box score is waiting before the players have finished leaving the court.

Someone has written the final numbers onto a sheet clipped to a board near the entrance: points, rebounds, assists, fouls, turnovers. Parents and spectators gather around it in small groups, scanning columns while the sounds of the game still echo from the gym.

One player scored eighteen. Another scored twelve. A third played nearly the entire game and finished with two.

The page makes comparison easy. It places every player inside the same narrow grid and allows the eye to move quickly from one performance to another. The player with eighteen appears important. The player with two does not.

But anyone who watched closely may remember something different.

The player with two guarded the opponent’s best scorer. She called out screens before they arrived, moved a teammate into the right position, and recovered when someone else was beaten. On offense, she set the screen that opened the lane for the final basket. None of these actions required her to touch the ball for long. Some did not require her to touch it at all.

The box score has not made an error.

Everything written there may be true.

The trouble begins when someone taps the page and says, “She didn’t do much.”

The Number After the Game

Sports depends on measurement.

A game produces too much movement to remember without reduction. Players run, turn, hesitate, collide, recover, communicate, and anticipate. Some actions succeed because of what happened immediately before them. Others matter because they prevent an event that therefore never appears.

The score gathers this movement into a result. The box score divides the result into individual contributions. Together, they make the game portable.

Without measurement, comparison would belong almost entirely to memory and reputation. The loudest moment could become the most important one. A spectacular play could displace a steady performance. A familiar player could receive credit for work someone else performed. Statistics do not eliminate these distortions, but they resist them. They force at least part of the story to answer to a record.

The player either made the basket or did not. The runner reached base or did not. The goalkeeper stopped the shot or did not. Measurement gives the game a memory less vulnerable than ours.

This is why complaints about statistics often miss the point. The number is not an intrusion into the game. It is one of the ways the game becomes visible.

But visibility always has a shape.

To record one thing is to distinguish it from everything surrounding it. The made basket becomes countable because the screen, the decoy cut, the defensive rotation, and the hesitation that opened the lane are treated as context. The statistic isolates an event from the conditions that made it possible.

That isolation is useful.

It is also incomplete.

The Player Whose Best Statistic Was Missing

Bill Russell played thirteen seasons for the Boston Celtics. During those seasons, Boston won eleven NBA championships. Russell was named the league’s most valuable player five times and accumulated 21,620 rebounds, an average of 22.5 per game.¹ The NBA remembers him not primarily as a scorer but as the defensive center of a dynasty—a player whose shot-blocking, rebounding, and ability to begin fast breaks changed how defense could organize an entire team.

Yet one of the actions most associated with Russell does not appear in his official career totals.

The NBA did not begin recording blocked shots as an official statistic until the 1973–74 season, several years after Russell retired.² The historical leaderboard therefore begins after one of the game’s most influential shot-blockers had already left it.

This creates a peculiar record.

Russell’s rebounds can be counted. His championships can be counted. His awards can be counted. The shots he prevented at the rim cannot—not comprehensively, and not within the official statistical system used to compare later players.

Even a blocked-shot total would reveal only part of what his defense did.

A defender can alter a shot without touching it. A player who expects Russell to be waiting may release the ball sooner, change direction, pass to a teammate, or decide not to enter the lane at all. The possession changes because of an action that never occurs.

This is one of the recurring difficulties of measurement: absence can be evidence of success.

The cornerback whose receiver is never targeted may disappear from the defensive statistics. The offensive lineman who prevents pressure produces a pocket in which nothing dramatic happens. The defensive midfielder closes a passing lane early enough that the ball goes elsewhere. The catcher guides a pitcher away from the pitch the hitter was waiting for, leaving no visible event to count.

Failure announces itself. Success may remove the thing that would have made success visible.

Russell’s legacy survived this absence because the outcomes around him were too large to ignore. Eleven championships created a record that pointed back toward the value the ordinary box score could not fully contain.

Most players whose contributions are difficult to count do not receive that protection.

They simply appear smaller on the page.

The Shape of Attention

A statistic does more than preserve information. It directs attention.

Once a category exists, people learn to look for it. A fan notices the player approaching a triple-double. A broadcaster mentions how many yards a running back still needs. A scorer near a milestone receives the ball on possessions that might otherwise unfold differently.

The number begins by describing the game and gradually becomes one of the forces shaping it.

This does not mean the players are behaving dishonestly. Measures help make goals legible. They allow athletes to track improvement, coaches to recognize patterns, and teams to make choices that memory alone would handle poorly.

But the categories also teach us what a contribution looks like.

A player who scores appears productive because scoring has a visible endpoint. The ball passes through the basket, the scoreboard changes, and a number attaches to a name. A player who creates space for someone else performs a more ambiguous kind of work. The value moves through another person before it becomes visible.

The box score prefers actions with clear ownership.

The game often depends on actions whose ownership is distributed.

A pass may receive an assist, but the defender who drew help before the pass does not. A rebound belongs to the person who controls the ball, though a teammate may have cleared the space. A touchdown belongs to the runner, though its possibility may have been produced by a series of blocks whose success appears mainly as open grass.

Adding one more column does not repair the incompleteness. Each additional measure creates another way to isolate contribution, and each isolation leaves relationships around it.

The question is not whether a better statistic exists.

It is what becomes visible when the statistic changes—and what disappears once we begin trusting it.

When the Old Numbers Were Not Enough

Baseball supplied one of the clearest modern examples of measurement correcting inherited judgment.

The approach associated with Moneyball did not begin from the belief that numbers were more real than baseball. It began from a practical observation: the baseball market did not value every contribution accurately. The Oakland Athletics, operating with fewer financial resources than many rivals, used statistical analysis to identify skills that contributed to winning but were comparatively inexpensive to acquire. On-base percentage became an important example because traditional evaluation had often given greater cultural attention to batting average, while the ability to avoid making outs could be purchased more cheaply than its value suggested.³

The insight was not simply that scouts were wrong and spreadsheets were right.

It was that baseball had been seeing through established categories for so long that those categories had begun to look like the game itself. Certain bodies looked like prospects. Certain statistics looked like production. Certain kinds of confidence looked like future success.

Analysis interrupted that familiarity.

A walk was already part of baseball. On-base percentage did not invent it. The measure changed how the event was valued. It gathered walks and hits beneath a different question: how often does this player avoid making an out?

Once the question changed, players who had appeared ordinary under one description became more valuable under another.

This is what good measurement can do. It does not merely assign numbers to what everyone already knows. It can make an old game newly visible.

The correction matters beyond the players who benefited from it. It protects sports from the authority of tradition. Without evidence, phrases such as “he looks like a ballplayer,” “she has the right instincts,” or “he knows how to win” can preserve reputation without requiring precision. Intangibles may be real, but the language of intangibles can also become shelter for unexamined preference.

Measurement asks the judgment to leave tracks.

That is one of its moral strengths.

When the Correction Becomes the Standard

A market does not remain surprised forever.

Once teams recognize that a skill is undervalued, they pursue it. Once enough teams pursue it, the price rises. The advantage created by seeing what others ignored begins to disappear. After the publicity associated with Oakland’s methods, major-league organizations increasingly hired analysts and incorporated statistical approaches; the specific inefficiency changed because the act of measuring it changed how the sport valued it.⁴

This is the part of the Moneyball story that is easiest to compress away.

The lesson was never merely “on-base percentage is the correct number.” If every team knows the value of reaching base, then the advantage no longer lies in knowing it too. The frontier moves.

The deeper lesson is that every settled account of value creates opportunities at its edge.

Baseball believed it understood what a good player looked like. Analysis found contributions that the inherited view had priced poorly. The new view spread. Teams reorganized around it. What had been a correction became conventional wisdom.

Then the new convention began leaving out different things.

A model built to identify inexpensive production may struggle with development, adaptation, injury, communication, or the interaction between a player and a particular environment. The fact that some of these conditions are difficult to measure does not make them mystical. It means the model was built to answer another question.

This pattern repeats across sports.

A new statistic reveals that a familiar player was overrated. Another reveals that a quiet player created unusual value. Tracking data captures movement away from the ball. Cameras estimate spacing, pressure, route separation, defensive positioning, or shot quality. Teams learn more, and the game changes in response.

The measurement succeeds.

Then success tempts people to forget the conditions of its success.

A tool built to correct one blind spot becomes the authority through which all other claims must pass. Work that does not appear inside the model begins to feel anecdotal. Judgment that cannot translate itself into a number begins to sound sentimental.

The correction becomes another compression.

The Game Changes to Fit the Measure

Players do not remain passive while they are measured.

They learn what the system rewards.

A basketball player may change shot selection because certain locations produce greater expected value. A baseball player may alter launch angle, plate approach, or defensive positioning in response to analysis. Coaches design around the efficiencies the model makes visible.

Often this is exactly what the measurement is for. A team should learn from evidence. An athlete should avoid a low-value habit merely because tradition made it familiar.

But optimization changes the environment the measure originally described.

A statistic may have identified value in a game where most participants behaved differently. Once everyone pursues the same value, defenses adapt, rosters change, and previously rare patterns become ordinary. The context moves beneath the number.

The measure can also narrow development.

When a young athlete learns which outputs receive attention, other forms of play may begin to feel wasteful. The extra pass, the decoy run, the patient possession, or the defensive risk taken to protect a teammate may matter less if its contribution does not attach clearly to the person performing it.

This does not require anyone to manipulate the system consciously. Attention itself is an incentive.

What gets displayed gets discussed. What gets discussed gets coached. What gets coached gets repeated. What gets repeated begins to define the sport.

The measure that once expanded sight can eventually train everyone to look in the same direction.

The Refuge of the Unmeasurable

There is an easy response to all this: trust the eye.

People who love sports often know that the game contains qualities the statistics cannot hold. They speak of leadership, timing, courage, intelligence, chemistry, sacrifice, and feel. The language can point toward something genuine. Teammates know who calms a huddle, who notices a tactical change first, who accepts an unglamorous assignment, and who becomes unreliable when pressure arrives.

But the unmeasurable is not automatically true.

A player can be credited with leadership because observers already admire them. A veteran can receive praise for presence while a younger teammate does the actual relational work. A coach can describe an athlete as lacking character when the athlete simply resists the coach’s preferred style.

Vagueness creates room for perception.

It also creates room for power.

Statistics can challenge stories that have become too comfortable. They can show that a celebrated scorer uses possessions inefficiently, that a supposedly careless player creates more chances than peers, or that the role everyone dismisses has a repeatable relationship to winning.

The right response to an incomplete measurement is not to retreat into a world where nothing must be demonstrated.

It is to hold measurement and remainder together.

The number deserves authority over the question it can answer. It deserves less authority over the questions it cannot.

What the Result Records

The scoreboard is the most successful measurement in sports because its limits are so clear.

It does not claim to record beauty, justice, effort, or future potential. It counts the actions the rules have designated as scoring and tells us which side produced more of them.

The trouble comes later, when the clarity of the score spreads into everything around it.

The team won, so its best-known player must have been the cause. The team lost, so its approach must have failed. The player scored little, so the player contributed little. The metric improved, so the underlying performance must have improved with it.

The scoreboard settles the contest. The box score distributes pieces of the outcome. Analytics offers increasingly precise accounts of how those pieces were produced.

Each layer can reveal something the previous one missed.

None can carry the whole game.

This does not make understanding impossible. It makes understanding cumulative. We see through several partial accounts at once: the score, the statistics, the film, the movements that do not become events, the choices that alter what another person can do.

The goal is not to find the final measure.

It is to remember that every measure has an edge.

The Sheet Near the Door

The gym is nearly empty now.

A custodian folds the final row of chairs while the sheet remains clipped to the board near the entrance. The numbers have not changed. Eighteen points. Twelve points. Two points.

Soon the paper will become the durable version of the game. Someone may enter the totals into a website. The eighteen-point performance will be easy to recall. The two-point performance will require someone who was there.

The player who scored two walks past the board without stopping. She is carrying a bag over one shoulder and talking with the teammate who made the final basket.

On the page, the basket belongs to one of them.

In the game, it belonged to more.

The box score remains correct.

The game remains larger.

Often the larger game still gathers around one face.

Footnotes

1. Basketball-Reference, “Bill Russell,” https://www.basketball-reference.com/players/r/russebi01.html; NBA.com, “Legends Profile: Bill Russell,” https://www.nba.com/history/legends/profiles/bill-russell. Career: thirteen seasons with Boston; eleven NBA championships; five MVP awards; 21,620 rebounds (22.5 per game). ↩

2. NBA.com, “Legends Profile: Bill Russell.” The NBA began recording blocked shots as an official statistic in 1973–74; Russell’s final season as a player was 1968–69. ↩

3. Michael Lewis, Moneyball: The Art of Winning an Unfair Game (New York: W. W. Norton, 2003); Jahn K. Hakes and Raymond D. Sauer, “An Economic Evaluation of the Moneyball Hypothesis,” Journal of Economic Perspectives 20, no. 3 (2006): 173–185. ↩

4. Dave Studeman, “The Triumph of Moneyball,” The Hardball Times / FanGraphs, January 23, 2014, https://tht.fangraphs.com/the-triumph-of-moneyball/. ↩