When the Data Falls Silent, People Still Tell Stories

In April 2026, I sat in the seventh row of the United Center after a game the home team lost by twelve. On the jumbotron, the stat sheet appeared. One player: 24 points, 9 rebounds, 7 assists, minus-18. A colleague sitting beside me turned and said, "He played well." I pulled up the tape. Yes, 24 points. But 14 of them came in the fourth quarter, when the team was already down 20, when the opposing defense had already let go. The numbers sat there, complete, and empty. That was when I started thinking about what I call the "empty data sheet" — a perfect structure, every field labeled, and nothing to read. The basketball analytics industry has come a long way. From the hand-written notes of 1980s coaches, to Second Spectrum and Synergy Sports, to plus-minus estimation models that can isolate a player from his teammates. Today, every NBA possession is logged by 25 tracking cameras, every touch is tagged, every gap is measured in inches. Data is no longer a supporting tool. It is the primary language. The regular season is where the volume of data swells largest. Eighty-two games, more than a thousand possessions per team, millions of tracking data points. And it is precisely there that the gap between data and meaning becomes most visible: a ten-game sample can look like a trend, a four-game winning streak can look like a turning point. But as that language thickened, a new problem appeared. The new problem is not about volume. It is about data beginning to pretend it means something. Three years ago, I sat in a postgame press conference and listened to a coach talk about his team's "heat map." He pointed to a bright red zone on the screen — the area where his player moved most — and explained that it was proof the offense was running correctly. No one asked him a simple question: is that red zone where the ball is being delivered, or where that player is trapped? A heat map does not answer that question. It just paints a beautiful picture and lets the viewer write the story. Over the years, I have watched analytics departments issue confident conclusions from maps like that — and I have seen them be right, and I have seen them be completely wrong. The deeper problem is this: a poorly designed analysis system can still return a polished output even when the input is entirely empty. I once watched a data pipeline return a report with every field filled in, every field named, every name properly formatted — and not a single field containing real information. No title. No source. No facts. Only a skeleton. The emptiest thing is not the emptiness itself. It is emptiness that looks exactly like fullness. In basketball, we meet this phenomenon every week. I call it the pressure to conclude. An analysis piece must have a conclusion. A report must have a judgment. A column must have an opinion. And when the real data is not enough to support that conclusion, what gets produced is not silence — it is a story that sounds very reasonable. Take Andre Drummond. In his peak years with the Detroit Pistons, he posted seasons averaging better than 15 points and 15 rebounds a night. On paper, those are the numbers of a top-tier center. But when you watch the tape, you see something else: rebounds that came from standing and waiting for the ball to bounce out rather than contesting, points that came from possessions that created no advantage for the team, and a defense that collapsed every time he was pulled away from the paint. Drummond was never a bad player. He was a player that basic data described completely wrong. Or take Russell Westbrook in 2026-17. He became the first player since Oscar Robertson in 2026-62 to average a triple-double: 31.6 points, 10.7 rebounds, 10.4 assists. The Oklahoma City Thunder finished 47-35, the sixth seed in the West, and were eliminated by the Houston Rockets in the first round. Westbrook won MVP. And ever since, no topic in basketball analytics has been argued over more: were his triple-doubles real, or empty? I do not want to take a side in that argument. I want to talk about the dash in the middle. Because what is scarier than empty numbers is empty conclusions. Even if all the data vanished, people would still tell a story. And that story would sound plausible. Here is the point I want to make, and it runs against popular intuition. When people argue about "empty stats," they usually ask the wrong question. They ask: are these numbers real? But they are always real. Drummond really did grab 15 rebounds. Westbrook really did record a triple-double. The right question is not whether the number is real. It is: what question is this number answering, and is that the question we need answered? I remember Ben Simmons in the 2026 Eastern Conference semifinals. In Game 7 against the Atlanta Hawks, he posted a stat line that was not bad: 8 points, 13 rebounds, 8 assists. If you only read the numbers, you would think he had done his job. But the whole game, and perhaps the rest of his career, was defined by a moment that appears in no box score: the fourth-quarter possession when he had a dunk available against a smaller defender and chose to pass the ball out. No column in the data sheet records hesitation. Here I have to say something blunt about my own profession. We — the people who report — bear part of the responsibility for creating these "empty data sheets." When readers demand a conclusion right after every game, when editors ask "so is this player good or bad," when a headline needs a number to become compelling — we come under pressure to say something. And saying something, when there is nothing to say, means inventing a story. I have done it. In 2026, I wrote a piece about a young player whose metrics spiked over ten games. I called him "the turning point of the season." Ten games. A sample far too small to say anything. He faded over the next fifteen and dropped out of the rotation. My piece was not wrong on the facts. It was just empty. And I think about it every time I sit down to write a new story. But here I also have to defend the data. Because a backlash is becoming popular: rejecting all numbers, retreating to the pure "eye test," believing that the intuition of a longtime viewer is more reliable than any model. That is another mistake. The human eye also produces empty data sheets. We just never see its fields. When I watch a game, I am swept along by rhythm too. I remember the beautiful shots and forget the terrible defensive possessions between them. I am led by the emotion of the crowd, by the sound of the arena, by the story I carried into the game before it began. Intuition is an analysis system — just one that is never forced to publish its methodology. And when it comes to load management, the gap between number and meaning is clearest. Kawhi Leonard sits out games across the regular season, and analytics departments produce a string of data-driven justifications: distance covered, physical-impact indices, a dense schedule. But that data does not decide who rests. People decide, then use data to justify it. That is another empty data sheet — one filled in after the conclusion already existed. So what actually needs to change? I think it is an attitude. In basketball analytics, we need to value saying "I don't know" as much as saying "I know." We need to build processes that can return an empty result — and accept that result as a valid one, not a failure to be papered over. An honest analysis system must be able to deliver the verdict: "insufficient information, cannot assess." And an honest sports press must have the same capacity. We do not need a conclusion for every game. We do not need a turning point for every week. We need patience. In the regular season, patience is the scarcest resource. The standings change daily, and with them, every story. A team wins three straight and becomes a title contender. A player misses four shots and becomes a burden. But the real tactical current — how a team changes its defensive scheme after a losing streak, how a player adjusts his shooting mechanics after an injury, how a coach experiments with a small lineup for 20 games before unveiling it in the playoffs — all of it moves far more slowly than the news cycle. That is what I want to find in every game I watch: not the number, but the shift. Not the conclusion, but the trace of a story forming. Where the ball rolls, we begin to tell the story — but I have learned that the best storyteller is not the one who tells the most, but the one who knows exactly when to be quiet. Years ago, stranded in Moscow after a World Cup match, I sat talking with an old Senegalese fan named Ousmane. He had followed his national team through five World Cups and had never once seen them win an opening match. I asked him why he kept going. He smiled and said something I still remember: "Because the next one could be the first." I think about that line every time I am tempted to write a rushed conclusion. A good sports story is not a story that has ended. It is a story still open — and an honest writer is one who admits he does not yet know the ending. On the pixel screen, I hear the heartbeat of the court. But I do not need to translate it into a number. Sometimes I just need to let it beat. Every time I put pen to paper, I ask myself: is this a real touch of the ball, or just a field filled in to fill the space? And you — the last time you read a complete stat sheet and felt you had understood something, are you sure what you understood was real, or only a conclusion someone needed you to believe?