When the Data Pipeline Returns Zero: Verification Discipline in Deep Basketball Analysis

2:47 AM, New York time

Three monitors, seven data tabs, a cup of coffee that had gone cold hours earlier. I was waiting for the extraction pipeline to return the Stage-1 deconstruction for a deep piece on the NBA trade market. What came back was an empty shell. Article title blank. Source blank. Article type blank. Core viewpoints blank. Entities involved blank. Time sensitivity blank. Source quality blank. Every core field carried the same value: N/A.

I sat still for about three minutes. Then I did what I always do when something does not line up: I rewound. I checked the pipeline logs. I checked the source path. I checked whether the original article was paywalled, video, or podcast format that a text extractor could not read. I checked three times. The result did not change.

What interested me most came next. The Stage-2 analysis handled the empty payload the only correct way possible. It did not fabricate. It rebuilt the full nine-dimension analytical frame, then in every cell, every table, every criterion, it wrote plainly: insufficient information to assess. It flagged process risk as High. It recommended a hard stop and a rerun of Stage-1. And it refused to issue tactical conclusions, player conclusions, salary-cap conclusions, or league-landscape conclusions — all the things a model trained to answer would desperately want to answer.

I read that whole document four times, and each time I became more certain of one thing: it was the best piece of basketball analysis I had read in months, and it contained not a single line about basketball.

I once recounted the tape four times, and the error belonged to the source, not to me.

---

Context: an industry built on unverified numbers

To understand why an empty report is worth writing about, you need to understand how professional basketball analytics actually operates.

On any given NBA night there are ten to fourteen games. Each game generates thousands of data points from optical tracking systems that record the position of every player and the ball twenty-five times per second. Those raw positions are transformed into hundreds of derivative metrics: offensive rating per hundred possessions, defensive rating, pace, true shooting percentage, usage rate, estimated plus-minus, and many more.

Technically, this is a marvel. Cognitively, it is a trap.

I call it a trap because I lived inside it. In my first year as a statistical research assistant, I believed that where there was a table, there was truth. If a center had a good defensive rating, he defended well. If a guard had a high true shooting percentage, he shot efficiently. I built entire arguments on the assumption that official data is structurally honest.

Then I discovered that official data is wrong in very small, very hard-to-see ways, and in ways that directly affect conclusions.

A simple example. A ball caroms off the rim; two players touch it within a fifth of a second. The system automatically credits the rebound to the last player who touched it. But the last player to touch it may have been the one who tipped it out so a teammate could collect it. The player who created the play gets nothing. An entire tactical act disappears from the official record, and nobody notices, because nobody recounts.

A rebound the league recorded wrong still counts — if you bother to rewind.

That is why I cross-check every number against two independent sources before writing. Not out of paranoia. Because I once published a wrong number, and people believed me.

---

One weekend in February, and the cost of a single bad number

On my first weekend as a freelance reporter at a college tournament, I was assigned to cover a game between a major program and a conference opponent. I logged the rebound total of a freshman who was drawing heavy attention at the time. I wrote the number in my notebook, filed the story, and went home.

That night, rewatching the tape, I saw that my number did not match what I had just watched. I rewound. I counted. I rewound again. I counted again. On the fourth pass, I sat down with a sheet of paper divided into four columns and marked every single possession. The result: the official tournament data feed had missed one play. I had not made the error. The source had.

I wrote a correction on my personal blog. The blog had two hundred and forty readers. Among those two hundred and forty was an editor at a major online sports publication. He shared the post, messaged me, and three weeks later I had my first statistical research job.

My career began with an error I did not cause, but that I was the only one willing to rewind and find.

I tell this story here because it explains my entire reaction to that empty report. When someone hands me an empty data frame, my first reflex is not to fill it. My first reflex is to ask where the pipeline broke.

---

The nine-dimension frame, and why preserving it matters

The Stage-2 analysis built nine dimensions: tactical and technical, player data, team operations and salary cap, league landscape and team positioning, rules and governance, coaching staff and locker room, risk, media narrative and expectation, and basketball industry ripple effects.

These nine dimensions are not decorative. They are the product of years of iteration, and they encode a very specific view of basketball: that a basketball event is never only a basketball event. A trade sits at the intersection of five different systems — the tactical system, the financial system, the rules system, the internal power system, and the media expectation system. Analyzing a trade while ignoring four of those five is not analysis. It is guessing with decoration.

What impressed me about the empty report was that it preserved all nine dimensions, including the cells that were impossible to fill. It did not trim the frame to look tidier. It let the emptiness show up fully, structurally, numbered, tabulated.

To me, that is a high-grade editorial decision. A structured empty frame tells the reader a clear story: we know what we need, we do not have it, and we are not going to pretend.

---

Dimension one: tactics, or why you cannot say anything when you do not know who you are talking about

Tactical analysis begins with a subject. A team. A player. A game. A series. Without a subject, there is no tactic.

This is something many basketball writers do not understand. They believe tactics are a set of concepts that can be discussed in a vacuum: pick-and-roll, switch defense, drop coverage, five-out spacing, sideline traps. They can write three thousand words about pick-and-roll without needing to know who is running it.

But pick-and-roll does not exist in a vacuum. It exists as a specific sequence of decisions by two specific people against two other specific people, in a specific context of score, clock, and fatigue.

A six-foot point guard with a short wingspan running pick-and-roll with a shooting big creates an entirely different defensive problem than a six-foot-five, two-hundred-ten-pound guard with a long wingspan running pick-and-roll with a rim-running big.

Same name. Two different games.

So when the analysis receives an empty frame, refusing to issue a tactical assessment is correct behavior. No team is named. No player is named. No offensive rating, defensive rating, or pace figure is supplied. No lineup is described.

Any sentence like "this team needs to improve its transition defense" would be a fabricated sentence. It sounds professional. It has enough terminology to pass the review of a tired editor at eleven at night. But it has no basis.

And this is the part I want to stress most in this entire article: in sports analysis, the most dangerous fabricated sentence is always the one that sounds most reasonable. An obviously nonsensical sentence gets cut immediately. A sentence with the right terminology, the right structure, but the wrong data, goes straight into the broadcast, into the podcast, into the feeds of hundreds of thousands of people.

I have watched that happen. Many times.

---

Dimension two: player data, and the problem of a stat sheet with no person in it

A professional player profile has four tiers of data.

Basic tier: points, rebounds, assists per game. This is the tier mass media uses most, and the tier that carries the least information.

Efficiency tier: true shooting percentage, player efficiency rating. This tier begins to say something, but is still easily misleading, especially with small samples.

Impact tier: plus-minus, estimated plus-minus, metrics adjusted for opponent and teammate quality. This is where real analytical work begins.

Usage tier: usage rate, touch rate, seconds of possession per touch. This tier explains who is actually controlling the game.

A complete profile needs all four tiers, plus age-curve position, contract status, and injury history.

In the empty frame, all four tiers were blank. No player was named. No number was supplied.

Croatia was not the team that ran the most — it was the team that ran in the right direction.

The correct response is to preserve the frame, mark every cell as insufficient information, and not extrapolate.

I want to pause here, because this is where I see many of my colleagues slip. When data on a specific player is missing, writers tend to pivot to talking about "that type of player." They write about "the modern center archetype," "the pure scoring guard," "the three-and-D wing." Those pieces read smoothly. They feel deep. But they answer no specific question, and they cannot be wrong, because they assert nothing.

In my profession, a piece that cannot be wrong is a piece worth nothing.

A good analytical piece must be falsifiable. It must make a claim specific enough that weeks later, when new data arrives, someone can reopen it and say: this person was right, or this person was wrong.

That is why I always state my data limits at the end of every podcast episode, even a forty-minute one. Listeners deserve to know how confident I am.

---

Dimension three: the salary cap, and the math you are not allowed to guess at

No dimension invites fabrication more than the salary cap.

The reason is simple: cap structure looks dry, so writers assume few people check. Max contracts, mid-level exceptions, rookie-scale surplus, luxury tax, bird rights, trade exceptions, sign-and-trade mechanics. Every concept is a trap for a careless writer.

One example I encountered. A piece discussed a team's "financial flexibility," asserting the team could generate a max contract slot in the next trade window. It sounded reasonable. The problem: that team had exhausted its exceptions, sat near the tax line, and had two contracts that could not be traded for six months. The flexibility did not exist.

The writer did not check. The editor did not check. And three weeks later, when the team signed nobody, readers concluded the team was incompetent. In truth, the writer had been wrong.

In the empty frame, the cap dimension was handled exactly as it should have been: every cell on max contracts, mid-level tier, rookie-contract surplus, and luxury tax left blank, with a note that cap mechanics require at least a team identifier and a transaction type as a starting point, and neither exists.

That is a fine sentence. It shows the report understands what is missing, and understands why that missing element is a blocking condition.

I asked myself: if I received this empty frame on a busy day, would I have the courage to write exactly that? The honest answer is yes — but only after I had taught myself that lesson the expensive way.

A thesis can survive a rejection; numbers cannot argue back.

---

Dimension four: league landscape, and the danger of ranking in a vacuum

Landscape analysis sorts teams into four tiers: contenders, playoff teams, play-in teams, and teams actively tanking for picks.

This sorting appeals to writers because it creates a feeling of clarity. You place a team in a box, and suddenly you have a story.

But tiering only means something when you know three things: the average age of the core, the contract window of that core, and the financial flexibility over the next three years.

Without those three, tiering becomes a vibes-based ranking exercise.

I once tracked a women's team through a nine-game losing streak. Local media called it a crisis. In reality, looking at optical tracking data, the problem sat at one very specific point in the transition defense system.

The rookie center was being targeted fourteen times per game in pick-and-roll coverage. Opponents scored an average of one point one seven points per such attack. Multiplied out, that is roughly sixteen points per game from a single gap. The team was losing by seven on average.

The problem was not morale. The problem was where one specific person stood in one specific situation.

I built a podcast series on this using detailed tracking data supplied by an assistant analyst. The coaching staff declined an interview. Three weeks later, the team changed its scheme: that center was kept closer to the rim. The series drew eighty thousand listens, five times my usual episode.

I tell this not to brag. I tell it to make a point: a team's situation does not live in how the team feels. It lives in very small, very specific places that only data reveals.

And in the empty frame, the landscape dimension was handled correctly: no league, no conference, no trade deadline context. The only confirmed parameter was the broad domain label "basketball," which is insufficient to scope any analysis.

Again: keep the frame, state what is missing, do not extrapolate.

---

Dimension five: rules and governance, where carelessness costs real money

Rules are the least-read and most consequential dimension in long-horizon analysis.

A team can win every game in March and lose its entire future in July, because of one clause in a collective bargaining agreement nobody in the room read carefully.

Checklist items always include: salary cap and luxury tax provisions, draft and extension rules, disciplinary mechanisms, and load management and competition format regulations.

A classic example of using rules as leverage: a team can absorb a bad contract in a trade to acquire a future first-round pick. On the books, that is a loss. Strategically, it can be one of the most valuable moves of the entire season.

A writer who does not read the rules will call that "a confusing trade."

In the empty frame, the rules dimension was handled correctly: no rule system, no dispute, no governance event surfaced. And the report stated plainly: compliance risk cannot be rated without an applicable rule and a fact pattern.

I like that phrasing. It is a sentence any sports lawyer would sign.

---

Dimension six: coaching and the locker room, where data ends and people begin

This is the hardest dimension for me, and the one most writers handle sloppily.

Data on the locker room barely exists in public form. You can measure how many kilometers a player ran, how many shots he took, how many touches he had. You cannot measure the level of trust between a star and a head coach.

That does not mean the dimension is unimportant. It means it demands a different kind of evidence, and demands that the writer state his confidence level explicitly.

Factors to assess: owner investment and patience, front office operating level, coaching staff stability, leadership structure, coach-player relations, and multi-star compatibility.

None of these can be inferred from an empty data frame.

And the report did it right: keep the frame, state that organizational dynamics cannot be modeled without subjects, and do not extrapolate.

One more thing about this dimension. Over years in this job, I have learned that most team failures do not come from tactics. They come from one person no longer believing another. And that never shows up in a stat sheet.

That is why I always credit assistant analysts in my work. They are the ones who see what the box score does not display.

---

Dimension seven: risk, and the biggest risk is not on any team

The risk matrix in basketball analysis has six categories: competitive, contract and financial, personnel, rules, public opinion, and systemic.

The first five all require a specific subject to assess.

The sixth is different. And in the empty frame, systemic risk was the only category rated High. Specifically: the risk that an empty analytical pipeline output prevents all downstream assessment.

I consider this the most important sentence in the entire report. It distinguishes between two fundamentally different kinds of risk.

Content risk is risk you can assess. This team depends too heavily on one ball handler. This scheme gets countered by a specific opponent. This regular-season style will not transfer to the playoffs. The new system is still gelling.

Process risk is risk you cannot assess, because you have nothing to assess. And if you insist on concluding anyway, you are creating a far larger risk: the risk of injecting false information into the system.

The report stated this at the highest priority level: analytical-integrity risk, where an empty input invites fabricated output, with a recommendation to refuse content-based conclusions, flag, and escalate.

I read that line several times. It reminded me of a principle I learned in my first year: when you do not know, say you do not know. It sounds simple. But in an industry that pays for confidence, it is the hardest principle to follow.

---

Dimension eight: media narrative, where rumors get promoted to facts

This is the dimension closest to my daily work, because I host a podcast. I live in the world of trade rumors.

Media analysis needs four things: what tier the source sits at, what the leak motive is, where market expectation stands, and the gap between that expectation and objective assessment.

Without those four, you are not analyzing media. You are repeating it.

I built my own source-classification ladder after years of being wrong.

Tier one: official team or league announcements. Near-absolute reliability, but usually arrives after everything is done.

Tier two: reporters with direct front-office or agency relationships and a track record of accuracy. High reliability, but motive still needs checking.

Tier three: reputable reporters without direct relationships, citing others. Medium reliability.

Tier four: anonymous social media accounts. Low to negligible reliability.

Tier five: content pushed out by the agent or the player's camp to create negotiation pressure. Reliability depends entirely on whether you recognize it as a negotiating move.

Tier five is the most dangerous and the most reported during trade season.

To me, player agents are the largest hidden cost of the trade market. Not because they do anything wrong. They do their job. But the noise they generate distorts player market value and skews the entire public conversation.

A player with average market value can be repriced as a strategic asset by three well-timed articles and two anonymous quotes.

That is why I rank rumors by evidence, follow the money, the contract structure, and the agent's moves, rather than by how compelling the story is.

In the empty frame, the media dimension was handled correctly: no source, no leak motive, no sentiment indicators. And the accompanying note was precise: the inability to identify even the source suggests the extraction pipeline received an empty or malformed article object.

That is a technical diagnosis, not a guess. I respect that.

---

Dimension nine: ripple effects, and why a small trade can shake an entire value chain

Professional basketball operates as a three-layer value chain.

Upstream: youth development, talent pipelines, agencies, international academies.

Midstream: teams, leagues, events, officiating.

Downstream: broadcast, sneakers and equipment, derivative markets, regional markets, international events.

A change at the midstream layer ripples downstream within days. A change upstream can take three to five years to appear midstream.

Ordinary writers only see the midstream, because that is where daily news lives. But most long-term value in the system sits upstream.

A team that builds a strong player development system sustains competitive position for fifteen years, regardless of which star it signs in any given summer.

A team that only buys stars without building foundations gets a short contention window, then collapses within two seasons.

I have watched enough cycles to believe this almost absolutely.

In the empty frame, the ripple dimension was handled correctly: no industry signal, no actor, no event to trace.

---

The contrarian angle: the most valuable part of an empty report is where it refuses to answer

This is the section I want to spend the most time on, because it runs against almost the entire industry's intuition.

The common intuition holds that the value of an analysis lies in the number of conclusions it produces. More conclusions, more value. More cells filled, more professional. More numbers in tables, more credibility.

I think that intuition is wrong, and wrong systematically.

An analysis with ten conclusions, three grounded in solid data and seven built on extrapolation, is worse than an analysis with three solid conclusions and seven cells marked "insufficient information."

The reason is practical: readers cannot distinguish a solid conclusion from an extrapolated one when both are presented in the same format, the same tone, at the same confidence level. They will believe all ten.

And seven wrong conclusions spread faster than three right ones, because wrong conclusions tend to be more compelling.

People see a mistake and laugh; I see a mistake and go looking for the source.

There is a concept in software engineering called the null-payload guard. When a function receives a null value, it is not allowed to continue as if the value were normal. It must stop, raise an error, and return a clear signal for the upstream system to handle.

The same should apply to basketball analysis.

When an analyst receives an empty dataset, the correct behavior is to stop, flag, and escalate. The wrong behavior is to fill the gap with what you know about similar topics.

I have seen the consequences of the wrong behavior. A preview of a team missing its number one star was written as if that star were playing. Nobody checked. The piece published. Three days later, after the team lost two more games, readers came back and found the piece meaningless. But the damage was done: the credibility of an entire column suffered.

What frightens me more are the cases where the error is never caught. Wrong numbers stay in old articles, still get cited, still get used as foundations for new analysis. Errors compound over time.

That is why I always attach a data-verification note at the end of every bulletin, even a short podcast one. Listeners need to know what I checked, and what I could not check.

This is an odd habit in this industry. Almost nobody does it. But I believe that in ten years it will be the standard.

Because as data volume grows, human verification capacity does not grow with it. The gap between data generated and data verified will only widen.

And that gap is the ideal habitat for error.

---

Case study: four recounts, and a lesson about unwelcome correctness

In 2026, while interning at a local radio station, I was assigned to analyze the defensive scheme of a national team at a major tournament.

I rewatched all seven of their matches. I logged the distance covered by a key wide player and found something I did not expect: he averaged twelve point three kilometers per match, but only thirty-one percent of that distance was oriented toward the opponent's goal.

Thirty-one percent of kilometers toward the opponent's goal is the number I want to talk about.

The rest was lateral running, backward running, position-holding, and creating space for others. Those runs had tactical value. But they never appear in the distance-covered statistic that media cites.

I wrote a nineteen-page internal memo on this imbalance. I wrote nineteen pages only to extract one sentence worth saying.

The editor did not use the memo. Too dry, too many numbers, not enough human element, he said.

After that team reached the final, he called me and admitted my assessment had been correct.

The lesson I took away was not "I was right." The lesson was: a correct assessment presented the wrong way will not be used, and therefore will have no value.

From then on I changed how I wrote entirely. I learned to embed data inside human stories rather than presenting bare statistics. Every analysis now opens with a specific moment involving a specific player, and only then places the statistic as supporting evidence.

If I had written that one sentence on the first page of a nineteen-page memo, the editor might have used it.

---

Second case study: a rejected thesis and the value of stating limits

In 2026, when leagues shut down, I defended a master's thesis on how playing without crowds affects free-throw efficiency.

I collected data from six hundred and twelve games over seven months. I found that free-throw percentage for players under twenty-five dropped an average of two point eight percent when arenas had no spectators.

Meanwhile, the largest European league showed no significant change.

When the crowd disappears, youth free throws disappear with it — unless you are in the EuroLeague.

The review panel judged the sample too small for conclusions. They were right methodologically. I did not argue.

But I used that thesis as the foundation for my first solo podcast episode. And I did something most sports content creators do not do: I stated the limits of my data explicitly, in that very first episode.

I said my sample was six hundred and twelve games, one league, one unusual period. I said the two point eight percent drop fell within an acceptable margin of error. I said my result was a hypothesis, not a conclusion.

That transparency made the difference.

Listeners do not come to my podcast to hear me be certain. They come to hear me point out where I am certain and where I am not.

In a market where everyone shouts, the person who speaks quietly but clearly gets believed.

---

Why an empty report is a high-craft document

Let me return to the opening question: why spend an entire article on a report that contains nothing?

Three reasons.

First, the report did the hardest thing in my profession. It distinguished between "I have no information" and "there is no information to have." It did not blur the two.

Second, it preserved professional structure under conditions of total data absence. A newcomer receiving an empty frame will either give up or fabricate. An experienced person builds the frame, marks every cell, flags process risk, and escalates. That is a skill, and it only comes after being wrong many times.

Third, it points to a truth I believe is central to all serious sports analysis: the value of a process lies not in its ability to answer every question, but in its ability to recognize which questions cannot yet be answered.

When I was young in this job, I thought an expert was someone who knew many things.

Now I think an expert is someone who knows exactly what he does not know, and says so before saying anything else.

---

What should happen next: a process, not a conclusion

The empty report ended with a blocking recommendation. It proposed supplying a corrected Stage-1 extraction containing at minimum four things: a populated article title and source, at least one factual information point about a team, player, transaction, or event, an entities list, and assessments of time sensitivity and source quality.

That recommendation is a checklist, not a complaint. And it is admirably short.

I want to expand it into a process anyone in sports analysis can apply.

Step one: identify the subject. If you cannot identify a subject, stop.

Step two: identify the source. If you cannot identify a source, downgrade confidence to the lowest tier and say so.

Step three: identify the timeframe. If you cannot identify a timeframe, you may not use relative expressions like "yesterday" or "this week."

Step four: cross-check every number against at least two independent sources.

Step five: state what you could not verify, at the end, in the plainest language possible.

Step six: if after five steps there is still nothing, write a piece about there being nothing. That is also a piece with value.

Step six is the step almost nobody takes, and the one that separates people who do this job from people who make content.

---

A few things about reading a stat sheet

Let me use this section for a specific skill I consider the most important in my job: reading a stat sheet and knowing where it is lying.

Four warning signs I always check.

First, small samples presented as large samples. A player shooting forty percent from three on twelve attempts is not a good shooter. He is a player with a small sample. The difference is enormous and almost always ignored in daily reporting.

Second, system-dependent metrics presented as individual metrics. A player with a high efficiency rating in a system optimized for him will lose most of that rating in a different system. That does not mean he is a system product. It means his metrics must be read in system context.

Third, automated data presented as verified data. Optical tracking is very good, but not perfect. It confuses players when two wear similar jerseys and stand close. It misses plays that are obscured. It assigns the wrong subject in contested situations.

Fourth, and most important: what is not measured tends to vanish from the discussion. A defender who is good because he makes opponents not want to receive the ball in dangerous positions will have no standout defensive metric, because no metric measures refusal.

Thirty-one percent of kilometers toward the opponent's goal is a metric I built myself, because no system supplies it. I had to count it myself.

And I counted a lot.

---

On the loneliness of the verifier

There is an aspect of this work few people discuss.

Verification is lonely work.

When you rewind a tape a fourth time to count a rebound, nobody is beside you. When you discover the official data feed is wrong, nobody applauds. When you decide not to publish a piece because you lack data, while a colleague published that same piece two hours ago and is pulling thousands of interactions, you feel like you are losing.

In the short term, you are losing.

Obsessive verification does not pay weekly. It pays annually.

It took me years to understand that. In my first two years I tried to publish faster, more, and louder. I wrote pieces I was not entirely sure of. Nobody caught it. But I caught it, and I did not sleep well.

Then I changed.

I set a rule for myself: never publish a number I had not cross-checked. If I ran out of time, I did not publish that number, and I said explicitly that I was missing it.

That rule cut my output by thirty percent in the first six months. It doubled my podcast audience over the following eighteen months.

Readers are not stupid. They can sense the difference between someone trying and someone certain.

---

Appendix: my verification method

I always attach this section at the end of every long analysis. Not to show off. To let readers know what I did and did not do.

Data sources: the league's optical tracking system, official statistical feeds, and internal data sheets from an assistant analyst on a women's team during the period discussed.

Cross-check method: every metric compared across two independent sources. When sources disagree, I prioritize the raw tape and recount myself. When the tape lacks resolution to conclude, I mark it as undetermined rather than choosing one source.

Limits: metrics on distance covered toward the opponent's goal were built by me and have not been normalized across leagues. Conclusions about the free-throw decline in crowdless conditions fall within an acceptable margin of error at two point eight percent and should not be used to predict individual outcomes.

What remains unverified: in the specific case of the empty report discussed at the top of this article, I have not identified the root cause of the extraction failure. The most likely hypothesis is that the source was not plain text, or that the pipeline received an empty object. I have no evidence to conclude with certainty.

And I will not write a conclusion I do not have evidence for.

---

A forward-looking thought

In the next ten years, sports analytics will have more data than anyone can process. That is certain.

But data volume is not the problem. The problem is the ability to distinguish verified data from merely generated data.

An empty frame handled correctly will become the standard in the future, not the exception. Serious media organizations will build null-payload guards, just as software engineers did long ago. They will state sources, state limits, and state what remains unverified.

And the fast, loud, unverified content makers will keep existing. They always will.

What I care about is whether there is enough room left for the other kind of person.

If every week I rewind one more tape, and every month I find one more error in a number everyone is citing, then the real question is not whether I am right. The question is: who will be the one to count a fifth time?

When you open the box score from last night's game and see a round, tidy, quotable number, ask yourself one question: has anyone rewound to confirm it.

If the answer is no, then that number is still waiting for the next person to check it. And in basketball, as in everything else, the next person is always the one who finds the mistake.

I will still be here, with three monitors, seven data tabs, and a cold cup of coffee. Rewinding for the fifth time.