Absent Data: The Empty-Analysis Trap and Lessons from Professional Chess

On December 12, 2026, in Singapore, on the 55th move of Game 14, Ding Liren pushed a rook to a square that no one could justify afterward on the analysis board. Fourteen days earlier, he was still the reigning world champion. After that move, he signed the scoresheet and left the board. Gukesh Dommaraju, born May 29, 2026, became the youngest world champion in history at eighteen, ending the match 7.5–6.5. In the press room, a reporter asked Gukesh how he felt. He answered briefly. But what kept me sitting for two more hours after I turned off the screen was not that answer, but the way dozens of analytical pieces were pushed out just hours later. One ran twelve pages long, beautiful, full of charts. But when I scrolled to the raw data section, an entire column was left blank, printing the letters N/A over and over like a mantra. What, I asked myself, were those charts drawn from? And I realized I had seen myself in that report, in a season many years ago. There is a paradox the professional sports world, and especially the chess world, has not yet named. The more analytical tools, the more raw data, the more analyses are produced that contain nothing but form. I call it empty analysis. It is not technically wrong in its presentation. It is simply a skeleton with no flesh, a suit of armor with no one inside. And in the transfer window, when hundreds of rumors flow through the pipeline every day, this kind of empty analysis is the most dangerous thing, because it looks exactly like the real thing. I have been in this profession for twenty-eight years. I started as a chess player, then a tournament organizer, then a commentator for VTC for seven years, covering almost every classic final of the era of players my generation names on a first-name basis. Then I moved into sports data. In 2026, at thirty-five, I sat inside a sports-data company in Shenzhen and was assigned to analyze the performance of a Brazilian striker. I thought I understood data until the summer of 2026 taught me a lesson I still remember vividly. That is why I am writing this piece, and why I will tell it through chess first, then through the transfer market. Let me start with a line I wrote in my notebook and reread every morning: When the data does not lie, it is we who lie to ourselves. The context of this story is bigger than one game. Over the past two decades, professional chess has undergone a revolution similar to the data revolution in football. Before computers were strong enough to beat humans at long time controls, chess analysis relied on feel, on memory of classic games, on the intuition of people who had breathed opening systems for twenty years. After Stockfish and Leela Chess Zero became the standard, everything changed. People began measuring in centipawns, in ACPL—average centipawn loss—in percentage accuracy, in engine-match rate. A game was no longer judged by feel; it was judged by a number. That was a great advance. But every advance breeds a trap. The trap here is this: once you have numbers, you start to believe everything can be measured, even the things that cannot. And when something cannot be measured, you do not admit it cannot be measured; you invent something shaped like a measuring stick. I have sat for thousands of hours in front of analysis boards in the backrooms of many tournaments. I learned that chess data has three layers. The first is the mechanical layer, where everything is absolutely clear: move count, result, Elo rating, total centipawns lost. The second is the contextual layer, where the number is still correct but its meaning depends on the situation: losing two hundred centipawns in a completely winning position is not the same as losing two hundred centipawns in a balanced position on move thirty. The third is the interpretive layer, where the number is only the starting point for a story the analyst must tell with intuition. Empty analysis is born when someone flattens these three layers into one, and draws a chart for the third layer even though there is no data in the first. In the transfer window, this trap works harder. Because contracts, release clauses, wage bills, transfer fees, agent fees, and performance bonuses are all real numbers, verifiable, signed by someone. But most transfer analysis never touches them. It touches rumors. And rumors have no first layer. They only have a third layer dressed up in a very technical-sounding name. Back to the game in Singapore. When I rewatched Game 14, the first thing I checked was not the final move. I checked the sequence of roughly fifteen moves before it, where Ding Liren had a chance and let it slip. His ACPL for the whole game was within a threshold that, by the numbers, remains acceptable. That is exactly why an empty analysis can say: “Ding played solidly; the mistake came in a single moment.” But when you place ACPL next to the time-pressure curve and next to the quality of the endgame, the story is entirely different. Ding’s death was not in one move. It was in an entire process of accumulated strain that the average number concealed. That is the lesson I paid for in the summer of 2026. In 2026, I predicted Germany would defend their World Cup title based on their possession data and passing accuracy in qualifying. I believed the number. I presented it to a group of colleagues, confident to the point of arrogance. Germany were eliminated in the group stage after losing to South Korea. My numbers were not wrong as numbers. But my model was wrong, because I had ignored pressure-conversion metrics and the speed of their wide attack, things that never appear in a possession table. After that shock, I spent three weeks rewatching all forty-eight group-stage matches, hand-calculating field-tilt and high-turnover metrics. I built a private dataset for underrated teams. It took me three months to learn that a beautiful chart is no substitute for a correct process. I wrote that line not as a signature but to remind myself every time my hand hovers over the mouse to copy a chart into an article. Empty analysis has a recognizable structure. I will describe it so you can test any analysis you read, whether about chess, football, or the transfer market. First, it has a very professional-sounding headline but never states any fact with a source. Second, it uses correct technical vocabulary but never explains where a number comes from. Third, when it faces a gap in the data, it does not say “insufficient information”; it says “needs further monitoring” without saying what to monitor, how, or for how long. Fourth, it always has a decisive conclusion at the end, because a decisive conclusion is what gives the reader a sense of satisfaction. The analysis I held that day is a perfect example of all four. Every field was marked N/A—short for insufficient information to assess—yet the scaffold remained intact: game and technical analysis, player and data analysis, tournament-system analysis, competitive-landscape analysis, rules-and-governance analysis, risk analysis, public-narrative and expectation analysis, industry-transmission analysis. Let me be clear so I am not misunderstood. Keeping the scaffold and filling every cell with N/A is, in itself, correct discipline, if the writer truly admits there is nothing there. That is honesty. The problem only arises at the next step, and this is the step many of us in the profession have committed: the step of turning an empty scaffold into a product that looks plump enough to deliver to a client, a newsroom, a fan. That is the line between discipline and deception. One side says “I do not yet know.” The other says “I know, but I pretend I am saying something.” And in the transfer market, these two sentences are mixed so thoroughly that people can no longer tell them apart. When a club negotiates for a striker, many things are genuinely unknown: undisclosed injuries, personal terms not yet settled, dressing-room reaction, family preference, income tax in the destination country. A single trap among them can collapse the whole deal. The honest analyst says he does not know three of those four things. The empty analyst draws an arrow pointing up and calls it long-term valuation. Now I want to tell you a more concrete story, the one that shaped my entire approach to the work. In 2026, in Shenzhen, I analyzed the performance of Luis Fabiano while he played for Tianjin Quanjian. He scored twenty-two goals in the Chinese Super League. Twenty-two goals is a number that makes any recruiter nod. But when I broke it down with xG and shots inside the box, the picture changed color. His actual efficiency was about eighteen percent below expectation, largely because he depended far too much on set pieces. I presented this to the club’s leadership. I argued their attacking system was too predictable. As a result, the club changed tactics and signed a younger striker with a better pressing index. I tell this story not to boast. I tell it to show that twenty-two goals is a true number, and my conclusion was also true, but they were not saying the same thing. If I had stopped at twenty-two goals, I would have produced empty analysis. If I had stopped at xG, I would also have produced empty analysis, because a single league’s xG is too small a sample to conclude anything about a career. What I did right was to place the two numbers side by side and read the gap between them. After that episode, I became better known in analytical circles. But what changed me was not the fame; it was the lesson about the difference between a number and an event. I once told a young colleague: data is a mirror, but only those who dare face themselves see the truth. Three months later, I understood that statement applied to me too. In 2026, COVID swept through the sports industry. Tournaments were postponed, stadiums empty, and an entire sports-media industry reeled. I lost income. But precisely at that moment, I recognized the value of historical data. While no one had live football to watch, I launched a project analyzing ten years of Premier League transfer data. I found that Brazilian wingers had a roughly forty-two percent higher successful-integration rate if they had previously played in Portugal, thanks to linguistic and cultural proximity. I built a player-valuation model based on a cultural-adaptation index and shared it on my personal blog. The piece spread and helped me escape the income collapse. COVID did not destroy football; it merely exposed who was living on illusion. But I will not tell this part without its downside. My model later failed several times when applied to other cultures. The forty-two percent figure is a correlation, not a causal law. Correlation is not causation. That is a line every analyst knows and almost no one actually behaves as if they know it. I once used that model to talk about West African players and the results were far worse, because my cultural variable was designed for a very narrow context. I learned that a model is only valid within the exact data frame that produced it. This is where I want to return to chess, because chess gives us a cleaner laboratory. Chess is the sport with the most perfect data that has ever existed. Every move is recorded. Results are binary. There are no fake injuries, no controversial red cards, no refereeing blunders. If empty analysis can survive here, it survives far more strongly everywhere else. And it survives here. I have read long analyses of an opening system whose author never ran it through an engine, relying only on feel and memory of old games. I have read assessments of young players based on a single tournament. I have read predictions about a player’s future simply because he won two games in a row against a stronger opponent. One of the biggest blind spots in modern chess analysis is engine-match rate. When you read an assessment saying a player has an eighty-seven percent engine-match rate, you tend to trust that as a good metric. But engine-match rate measures agreement, not quality. A player can move exactly like the engine in an already fixed position and still understand very little about it. Another can move differently from the engine while actively setting a psychological trap the machine cannot see. The number cannot distinguish these two cases. It only speaks to the degree of similarity. I often think of this when I hear people discuss the finals I once commentated on over many years. The great players left their mark not through engine-match rate. They left their mark through moves the machine did not choose but later had to admit were sound. Moves that created problems for the opponent in time, in psychology, in the position. That is the part the number cannot touch. That is why I say data is not the destination. A Chinese club taught me that data is not the goal but a walking stick. You hold it to move, not to stand still in place. So if the data is complete and perfect, why does empty analysis still exist? The answer lies in the economics of the trade, not in the technique. Empty analysis exists because it is cheap to produce, fast to publish, and hard to disprove immediately. A real analysis takes dozens of hours to gather data, dozens more to cross-verify, and can take a week. An empty analysis can be written in two hours, with a beautiful scaffold and ten N/A cells, appearing more professional than most other things. In a content environment that competes on speed, this is an enormous economic advantage. And in the transfer window, when hundreds of rumors and millions of reads accumulate daily, that advantage grows. That is why I want to say this plainly to younger people in the trade. If you choose to write faster rather than write correctly, you are not competing. You are burning the credibility others spent years building. From a raw data warehouse to a data monastery—the journey is not only technology. It is discipline. Here I want to use this section to talk about what I believe is the most common mistake, and also the part where I deceive myself most easily—not you, but me. The most common mistake is confusing having a scaffold with having an analysis. A good scaffold is necessary. But a scaffold filled with N/A is not an analysis, and the danger is that it looks very much like one. The reader only sees the structure. The reader does not see that in every cell, the author dodged a decision. I will put it another way. When an analysis has ten cells and all ten rest on a single source, it is not a deep analysis. It is one source repeated ten times. This is the second dangerous trap, and it is far subtler than leaving a cell blank. The third trap is the trap of consensus. When five articles say the same thing, people think it is the truth. But five articles may simply be reading one original. This is a phenomenon I see in both chess and the transfer market. In chess, an assessment of some player spreads, and ten reports repeat it as gospel. In the transfer market, a rumor is reposted by five outlets, and suddenly people believe it is fact. But five sources are not five pieces of evidence. They are one piece of evidence counted five times. After 2026, I no longer trust predictions. I trust only early-warning systems. The difference between the two is huge. A prediction says what will happen. An early-warning system says that if this happens, I will have to pay attention. One asserts. The other monitors. I choose the second, because the second never demands that I know what I do not know. Now let us talk about how a real analysis is built, so you can compare it with what you read. A real analysis begins with a testable question. Not “is this player good,” but “under what conditions does this player perform above his own average.” The first question cannot be answered; the second can. It continues by defining the metric clearly. Not “good efficiency,” but “expected goals per ninety minutes in matches where the opponent is in the top ten.” This specificity is what separates analysis from feeling. It continues by admitting limits. Every number has limits. An honest analysis states plainly what it cannot measure. An empty analysis fills the gaps with pretty words. And it ends with a signal, not a conclusion. It says: monitor this index. If it crosses this threshold within this many matches, then my hypothesis needs review. That is a system, not a verdict. It took me years to build this habit, and I still make mistakes. People of my type—those driven by efficiency and fond of quick conclusions—are the most likely to settle a game before it ends. I had to set a rule for myself called long-term valuation: whenever I am about to conclude about a player, I force myself to consider two more seasons, even seasons that have not happened. Not to predict, but to test whether my conclusion can withstand that stretch of time. Most hasty conclusions collapse the moment you extend the time frame. There is one thing I want to say specifically about the transfer market, because we are in a transfer window and this is when the traps work hardest. The transfer market is not a chess game; it is a synchronized performance of thousands of algorithms. Thousands of clubs, thousands of agents, thousands of investment funds, each optimizing for its own interest. No single side sees the whole board. When you read a report saying Club X will sign Player Y, you are hearing one side of them. Not the whole system. One side speaking may be the agent negotiating. Another may be the club applying pressure. Another may be the league needing good news. Understanding whose voice is being transmitted matters more than understanding its content. In the transfer window, where is the real signal? It lies in release-clause structure, in remaining contract length, in leftover wage budget, in the days before the market closes, in whether a player is negotiating personally. These are verifiable. Whether a player likes a certain city is not verifiable, and therefore it is often used to replace what the writer did not check. I want to return to Game 14 in Singapore once more, this time from another angle. What struck me most about Ding Liren’s wrong move was not the move itself. It was how it was explained afterward. Some experts said he was in time trouble. Others said he was under psychological pressure. Some said it was the result of playing too much. Each explanation is plausible, but none was proven by data. They are stories attached to a move whose true cause may never be known. This is the boundary between analysis and interpretation. Analysis can say: in this position, this move loses so many centipawns compared to the best move. Interpretation can then say: why. And much of what is called chess analysis is actually interpretation dressed in the look of analysis. I do not say this to dismiss the value of interpretation. I say it to distinguish the two, because both are necessary and both are different. What an honest writer does is state clearly which side of the line he is on. A line I often write in internal analyses is: “Here the data ends and judgment begins.” This line matters because it protects the reader from believing that a judgment is being called data. I have tracked niche tournaments and underrated young players for many years. That is a good habit, because it is where empty analysis appears least, since not enough people care to write it. But I also know it is where I deceive myself most easily, because I see players others do not see, and my ego wants to say we spotted what the market missed. Treat your own data as harshly as your opponent’s. That is a line an old colleague and I reminded each other of. Because when I am the one who discovered a young player, I have an incentive to defend my discovery with weaker data than I would use to judge another player. I want to spend the final part of this analysis on what I call the next-cycle signal. In chess, the next cycle will be decided by how the young cohort such as Gukesh, Praggnanandhaa, and Firouzja handle their playing load. Not by the openings they choose, but by the physical and psychological structure that lets them withstand a dense calendar at high quality. This is the signal to track, not individual results. In the transfer market, the next cycle will be decided by how clubs handle the tension between financial reporting and sporting decisions. When shareholder pressure is high, sporting decisions are often overridden. This is something I believe strongly: a club IPO turns fan emotion into money, and financial-reporting pressure often weighs on sporting decisions. That is a long-term signal, slower than one transfer window, but stronger. On refereeing and VAR, I have spoken many times in previous pieces. Millimetre offside lines are killing the attacking instinct. Referees are becoming editors of the match instead of officials running it. In chess, something similar happens with anti-cheating rules. Every rule protecting fairness carries a price, the freedom of players to express instinct. This balance is the next-cycle signal. Let me return to the question at the start of this piece. Why does an empty analysis exist in a sport with data as perfect as chess, and in a market with verifiable contracts like the transfer market? My answer is: because data is not what produces conclusions by itself. Data is what humans must read. And humans have incentives to misread it, or to not read it and still get paid. That is why I no longer trust predictions. I trust early-warning systems. An early-warning system does not tell you what is about to happen. It tells you when you need to reread your assumptions. In an industry where information moves faster than understanding, that is the only tool truly worth building. And that is what I bring from what I have lived: an analysis room in Shenzhen, a striker who scored twenty-two goals and whom I had to call out as depending on set pieces, a German national team eliminated when every model of mine said otherwise, a pandemic season that forced me to find value in historical data, and a game in Singapore that ended with a move I rewatched no fewer than twenty times. Each time, I learned the same lesson. Not about chess, not about football, not about the transfer market. About holding discipline when you do not know. I think about that every time I hold a twelve-page analysis and see a column full of N/A. That day I did not throw it away. I read it to the end. Because I recognized it as a version of myself, at another moment, when I had the right scaffold but no data, and what I chose to do was not to go find data, but to make the scaffold look full. When the data does not lie, it is we who lie to ourselves. I have written that line many times. But there is a line I have never written, and now I write it: if you are holding an empty scaffold, tell the reader it is empty. That is the first step, and for many in the trade, also the hardest. I will keep tracking this cycle. Not to predict who will win. To watch three signals: the quality of passes under high pressure among the top teams, the chance-conversion rate of young players in balanced positions, and the release-clause structure in the big deals of this transfer window. If there is one thing I want you to take away after reading this far, it is this: do not ask whether the analysis you are reading is beautiful. Ask whether it has enough data to exist. If it does not, then no matter how long it is, it is only a decorated silence.