When the Numbers Argue With Your Eyes

A Familiar Basketball Argument

Every basketball fan has heard the exchange. One person points to the numbers — efficiency, shot distribution, lineup data — and insists the evidence is clear. Another shakes their head, having watched the same games, and says the numbers are missing something obvious.

The debate usually gets framed as analytics versus the eye test, as if the disagreement were simply about whether spreadsheets or eyesight should run the sport. Yet the deeper question is not really about numbers at all. It is about what should justify belief.

When someone says a player helps winning, or that a team’s offense makes sense, what counts as a good reason for that belief?

Philosophers call one answer evidentialismEvidentialism is the view that a belief is justified only to the extent that it is supported by the evidence available to the person holding it. Feelings, traditions, or authority alone are not sufficient grounds for rational belief. : the idea that beliefs should be guided by the evidence available to us. In basketball terms, the principle is straightforward. If you claim something about the game, you should be able to point to reasons that support it — reasons that other people can examine, question, and test.

The eye test, by contrast, often begins with an intuitionIn philosophy, intuition refers to a judgment or belief that arises immediately and without conscious reasoning. It can be a valuable starting point, but philosophers debate whether it can serve as reliable evidence on its own. . You watch a player move, react, improvise, and a judgment forms almost immediately. Sometimes that instinct is perceptive. Other times it quietly inherits all sorts of biases about how basketball excellence is supposed to look.

The interesting question is not whether intuition exists — of course it does — but whether intuition by itself is enough.

The Rockets and the Discipline of Evidence

A useful place to begin is Houston in the late 2010s, when the Rockets reorganized their offense around a very simple evidential claim: over time, some shots produce more points than others.

Three-pointers, layups, and free throws consistently outperform mid‑range jumpers when measured across large samples. That fact does not rely on taste or aesthetic preference. It emerges from thousands of possessions.

Under Daryl Morey and Mike D’Antoni, Houston pushed that logic to its extreme. The team almost eliminated mid‑range attempts, filling games with threes and drives instead. The approach looked strange to many viewers who had grown up associating the mid‑range pull‑up with basketball craftsmanship.

But the Rockets were not chasing style. They were following evidence.

From an evidentialist perspective, the key point is not that every individual three-pointer is wise. The justification lies in the pattern across the season. If a shot profile reliably produces more points per possession, then the belief that this style of offense is better rests on stronger grounds than the intuition that certain shots simply feel like better basketball.

In other words, the Rockets were not replacing thought with numbers. They were demanding that beliefs about offense answer to evidence rather than habit.

The Game That Fueled the Backlash

Then came Game 7 of the 2018 Western Conference Finals.

Houston missed twenty‑seven consecutive three‑pointers against Golden State and lost the game. The image of those missed shots quickly became a kind of cultural proof that the analytical approach had gone too far.

Yet the reaction reveals something about how intuition works.

A vivid failure can feel more convincing than a long stretch of quieter evidence — an example of what psychologists call the availability biasAvailability bias is the tendency to judge the likelihood or importance of events based on how easily vivid examples come to mind, rather than on the full body of evidence. Dramatic moments are overweighted because they are memorable. . Watching a team miss shot after shot creates the sense that something must be fundamentally wrong. The mind latches onto the drama of the moment.

But evidential reasoning treats events differently. One game, even a catastrophic one, is a tiny sample compared to an entire season of possessions. The Rockets’ system had produced sixty‑five wins that year. A single streak of misses may reveal the cruelty of variance, but it does not automatically overturn the evidence that supported the strategy in the first place.

The tension here is philosophical. Intuition tends to overweigh the vivid. Evidence insists on looking at the full body of information.

When the Eye Test Goes Wrong

Sometimes intuition fails not because it ignores numbers but because it notices the wrong things.

Consider Nikola Jokić.

Early scouting reports often focused on what he lacked. He was slow-footed, heavy, and visually unconventional for a dominant NBA player. To many observers he did not resemble the kind of athlete who usually becomes a franchise centerpiece.

Yet while those impressions were forming, other evidence was quietly present. Jokić manipulated defenders with passing angles that most centers never see. His timing was extraordinary. His decisions arrived half a second earlier than the defense expected.

Those qualities were visible on film, but they were subtle. They required a different kind of attention.

The problem, then, was not that scouts watched games. It was that the eye test had become superficial, guided by familiar visual markers like burst, verticality, or smooth isolation scoring. Those markers had become proxies for greatness even when they failed to capture what was actually happening on the court.

Once Jokić’s production accumulated — points, rebounds, assists, efficiency — the evidence forced a reconsideration. The numbers did not defeat vision. They corrected a poorly trained version of it.

Evidence Also Criticizes Its Own Tools

If evidential thinking demanded blind faith in metrics, the story would end there. But evidence applies to measurement itself.

Russell Westbrook’s 2017 season is a good example. Averaging a triple‑double across an entire year pushed statistical models into unfamiliar territory. One advanced metric, Box Plus-Minus, initially rated the season so highly that analysts later revised the formula to better handle extreme cases.

This was not a victory for anti‑analytics sentiment. Instead it showed something healthier: evidence correcting one of its own instruments.

Metrics are tools for tracking patterns. When a tool breaks under unusual conditions, the response is not to abandon evidence altogether. The response is to refine the instrument.

The philosophy is the same. Justified beliefA justified belief is one held for good reasons — reasons that can withstand scrutiny. In epistemology, justification is what separates knowledge from mere opinion or lucky guessing. requires evidence, but it also requires good methods for interpreting that evidence.

Turning Intuition Into Evidence

The eye test becomes most useful when it stops presenting itself as a verdict and instead becomes the beginning of an argument.

Take Tyrese Haliburton as a prospect. Some evaluators worried that his shooting motion looked unusual and that his thin frame might limit him. Those are intuitive impressions. On their own, they carry little weight.

But once the intuition is translated into a claim — perhaps about release mechanics, defensive physicality, or the translation of shooting form to NBA range — the judgment becomes testable. It can be weighed against other information: college production, assist numbers, efficiency, decision-making, and on‑court impact.

At that point intuition is no longer standing alone. It has been converted into evidence that can be evaluated alongside everything else.

That transformation is what separates disciplined observation from vague basketball mysticism.

Seeing the Game Differently

The analytics-versus-eye-test debate persists partly because the language makes the conflict sound simpler than it is.

Watching basketball is indispensable. The game is too fluid and contextual to be understood by spreadsheets alone. But watching becomes genuinely informative only when it leads to claims that can be explained and examined.

Evidence, meanwhile, is broader than people often assume. Statistics are one form of evidence, but so are carefully described observations, tactical patterns on film, and repeated behaviors across possessions.

Once the issue is framed this way, the conflict changes shape.

The real divide is not between numbers and eyesight. It is between beliefs that can show their reasons and beliefs that merely feel certain. The best evaluators — whether they sit in front offices or scout from the stands — move constantly between perception and evidence, turning what they see into claims that can survive examination.

In that sense, analytics did not replace the eye test. It forced the eye test to become more accountable.

And the moment intuition begins explaining itself, it stops being a rival to evidence and becomes part of it.

---

Footnotes / Philosophy Terms

1. Evidentialism ↩

Evidentialism is the view that a belief is justified only to the extent that it is supported by the evidence available to the person holding it. Feelings, traditions, or authority alone are not sufficient grounds for rational belief.

2. Intuition ↩

In philosophy, intuition refers to a judgment or belief that arises immediately and without conscious reasoning. It can be a valuable starting point, but philosophers debate whether it can serve as reliable evidence on its own.

3. Justified belief ↩

A justified belief is one held for good reasons — reasons that can withstand scrutiny. In epistemology, justification is what separates knowledge from mere opinion or lucky guessing.

4. Availability bias ↩

Availability bias is the tendency to judge the likelihood or importance of events based on how easily vivid examples come to mind, rather than on the full body of evidence. Dramatic moments are overweighted because they are memorable.