Table Tennis Has No xG: Why Data Still Cannot Tell the Whole Story of a Match

Table Tennis Has No xG: Why Data Still Cannot Tell the Whole Story of a Match There was a match in the WTT system in which the winner scored only 41 points while the loser scored 49. The set score was 3-2. The winner walked away with a ticket to the next round; the loser sat staring at the scoreboard, wondering what had just happened. In football, such a result would immediately be placed beside xG to be dissected. In table tennis, we have no xG. We have serve-point win rate, third-ball attack conversion, and the number of rallies extending beyond five exchanges. But all of these numbers added together still do not produce a single answer: who actually played better over the last forty minutes. Every analysis session, I begin with an uncomfortable question. If numbers do not lie, why do we keep getting it wrong? Since 2026, World Table Tennis (WTT) has changed how the sport operates. The tour is tiered — Grand Smash, Champions, Star Contender, Contender and Feeder — with a ranking ladder that rolls over within 52 weeks. Points are protected on a cycle, meaning a player can drop in the rankings without losing a match, simply because old points have expired. For a player inside the top 10, this pressure cuts two ways. First, they must play more to accumulate points. Second, they must choose tournaments wisely to avoid an early loss to a lower-ranked opponent. I once tracked a player who competed in four events across six weeks. By the fourth, lateral movement speed had dropped noticeably, visible in the number of missed steps on backhand rallies. The ranking still looked good. But the legs said otherwise. This is where data becomes useful, and also where it becomes dangerous. Rankings measure accumulated points, not current form. A pretty number on the ladder can conceal a physical platform in decline. To understand table tennis today, you have to understand the rule changes that shaped it. In 2026, the ITTF increased the ball diameter from 38mm to 40mm. A bigger, heavier ball means less spin, stripping an advantage from players who lived on pure spin. In 2026, the scoring system changed from 21 points per game to 11. More games, shorter games, more luck variance. In 2026, the hidden-serve rule took effect, removing a weapon from many classic servers. In 2026, speed glue containing organic solvents was fully banned. In 2026, the celluloid ball was replaced by plastic. Each change benefited one group of players and hurt another. Historical data therefore cannot be compared on a straight line. When someone says player A was stronger than player B of an earlier era, I always ask three questions: which ball did they play with, under which scoring system, and under which service rule. At Paris 2026, this picture became clearer. Wang Chuqin, the top seed in men's singles, was eliminated by Truls Moregard in the round of 32. Fan Zhendong went on to win gold. In women's singles, Sun Yingsha dominated the early rounds but lost to Chen Meng in the final. Look only at the rankings, and these results seem absurd. Look at the point-conversion rate in decisive rallies, and the story is entirely different. Men's and women's table tennis have different degrees of openness. In the men's game, the gap between the leading group and the rest is narrowing, evidenced by the rising number of players outside the top five reaching the semifinals of major WTT events over the past three years. In the women's game, the stability of the top group remains higher, though young Asian players are steadily closing the gap. On equipment, when the 40+ plastic ball replaced celluloid, ball speed dropped and spin characteristics changed. Fast two-winged attackers benefited, while far-table defensive styles lost ground. This is a textbook case of equipment becoming a tactical variable, not merely a technical matter. On development, China still maintains a talent pipeline through a multi-tiered selection system. But the gap in the 18 to 23 age bracket is narrowing. Young players from Japan, South Korea and Europe are appearing more often in major knockout rounds. On the commercial side, WTT is pushing digital media with real-time data. But commercial value and competitive value do not always move together. A player famous on social media may not have a high win rate at Grand Smash events. Serve points are the most reliable metric, and also the easiest to misread. A high serve-point win rate does not necessarily reflect good serving, because it depends on third-ball attacking ability. I always separate the two numbers: serve efficiency and third-ball efficiency. Blending them is the fastest route to a wrong conclusion. The biggest risk for a player is not the direct opponent, but the schedule. A dense tournament calendar makes shoulder and wrist injuries a standing variable. A minor wrist strain is enough to completely change the quality of a backhand, and that never shows up in the rankings. On the public side, media narrative usually runs ahead of data. A shocking loss spreads far faster than an analysis of third-ball point-conversion rate. The gap between media heat and underlying strength is where forecasting errors are born. This is where I have to say plainly what many in the industry avoid. Table tennis has too much variance for small data to reflect the true picture. An 11-point game ends too quickly. Two missed serves, one edge ball, one lucky net cord, and momentum flips. Compared with football, where 90 minutes generate hundreds of events, table tennis generates too few events for any metric to reach statistical reliability. A sample of 40 to 60 points per player in a single match is far too small to conclude anything about long-term form. More seriously, the psychological factor is almost impossible to quantify. I once sat in an arena and watched a young player lead 10-6 before losing 12-14. My data sheet could only record two words: loss of focus. But paper cannot capture what happened inside his head at 10-7. In 2026, I looked into the players' eyes before I looked at the score sheet. That experience taught me something still true today: numbers do not lie, but the people reading them do. We unconsciously select the figures that confirm what we already want to believe. That is also the limit of data analysis inside the locker room. A model's conclusions are often out of step with the actual rhythm of a match. An algorithm can compute the exact win probability of a serve, but it cannot compute how a player feels when he realizes his opponent has read his signature serve. The data monk does not pray for victory, but for correctness. So do I. In table tennis, data is a tool for narrowing the blind spots, not a final verdict. The right question is not which metric looks better, but which metric helps you make a better decision. The next season will answer an old question: is a transparent ranking system enough to predict outcomes, or will the table tennis table — where everything happens in a split second — remain the place where data must bow to human instinct?