Inside a 214-Match Dataset: The Korea–Japan Women's Football Gap Lives in Set Pieces

In the 23rd minute, Lee Min-a received the ball at the edge of the box, turned, and shot. The opening goal for Incheon Red Angels. The stands at Incheon that day held 347 people. The match's only camera, fixed at midfield, missed the entire sequence that led to that moment. I know, because I was sitting in the third row with a low-angle camera I had rigged up myself. It was the first match I handled during my internship at the women's sports channel Her Ball — round 12 of the WK League, between Incheon Red Angels and Gyeongju KHNP. On paper, an unremarkable fixture. But it taught me something I have carried through twelve years in this trade: in women's football, the most neglected thing is not the goal, but the way people record it. The secondary camera is not a low starting point. It is an angle the stands have never seen. For most of this sport's history, women's football has been documented through a system inherited from the men's game — one designed for matches with dozens of cameras, thousands of data points per minute, and an entire analysis department. An average WK League match has one or two cameras. A women's national team qualifier may have none beyond a fixed, non-zooming main frame. As a result, most of what we know about Korean women's football comes from a very thin dataset: goals, shots, possession percentage. Those three indices say almost nothing about how a team operates. They tell you what happened, not how it was produced. Twelve years of watching women's matches in Korea have taught me that the most important habit is not watching a lot, but watching again. A passage of play only fully reveals itself on the third viewing. On the first, you see the ball. On the second, you see the player with the ball. Only on the third do you see the players without it — and that is where the match is actually written. In 2026, as leagues worldwide paused, I had just finished a master's in sports management. I had no live matches to watch. So I did what a watchmaker does when there are no customers: I took the watch apart and counted every gear. I built a dataset of 214 matches of the Korean women's national team from 2026 to 2026, and re-watched every goal by hand to classify its origin. 214 matches, 214 problems: the pandemic did not stop football, it only changed how we read the game. The classification was deliberately simple. Each goal was assigned to one of four groups: organized open play, transition counter-attack, set piece, or individual error. I used no data provider's definition of a dangerous chance. I counted myself, cross-checked myself, and logged the exact minute of every sequence. For any passage I was unsure about, I watched it three times before entering it. Before concluding anything, I asked the question a demanding editor would ask: so what? The result made me stop. Across those 214 matches, the Korean women's national team scored only 23.7% of its goals from set pieces. Japan, over the same period and using the same classification, reached 41.2%. That is not a small gap. That is nearly double. And it does not lie in physique, speed, or finishing technique — it lies in how many times a team repeats the same combination until it becomes reflex. Set pieces are the most undervalued thing in football and the cheapest to improve. They do not require players to be faster or stronger than opponents. They require time on the training pitch, someone to design the drill, and the patience to repeat a movement until it is boring. For women's football — where training time is tighter and players often hold second jobs — this is the highest-return investment available. Yet it is the first item cut when the schedule tightens. When I dug deeper, the gap between the two teams was not in the quality of the cross. Japan did not deliver prettier crosses than Korea. They had more options for the same corner. A Japanese corner routine usually has three layers: a blocker running to the near post, a decoy run to drag a defender, and the true receiver at the far post. Those three layers only work if they are rehearsed at least a few dozen times a week. Korea, in most of the sequences I reviewed, had only one layer. I do not trust emotion, I trust data. Emotion can lie; a spreadsheet cannot. That difference never surfaces in any official stat sheet, because a stat sheet only records corners: 6 and goals: 0. It does not record that all six corners followed the same script, and that the opponent had read the script by the second one. To see that, someone must sit down, watch each one, and count the layers of movement. No one pays for that work at a league with 347 spectators. In October 2026, I sent that report to the head coach of the women's national team. I did not expect much. A week later, I received a reply email inviting me to collaborate on opponent analysis during the training camp. It was the first time a dataset I had counted by hand opened a professional door. Not because it was smarter than anyone else's data, but because it pointed to something concrete that could be fixed tomorrow. What I learned from that collaboration was not a new number. It was a principle: when official data and footage say two different things, trust the footage, but do not trust your own memory. In 2026, as a new staffer at SBS Sports, I was assigned women's football for the Tokyo Olympics. I re-watched the England–Japan group match and counted 17 fast counter-attacks by England in the second half. The official stat sheet recorded 3. I watched the match four more times before writing. The number 17 was not a loose interpretation of a counter. It was the count of times England won the ball in the opponent's half and moved it forward within five seconds with at least three players involved. I counted each one and logged the minute. The official sheet counted dangerous chances — a concept with no agreed definition across data providers, and in this case apparently limited to sequences ending in a shot on goal. The gap between 17 and 3 is not a technical error. It is a definitional choice. And when a definitional choice is presented as objective fact, it becomes a weapon. A team reading that sheet would believe it was threatened only three times, when on the pitch it had been threatened seventeen. My rebuttal about how the media defines a dangerous chance arbitrarily was later shared by a K League coach as reference material for his own trainees. He did not share it because it was right about women's football. He shared it because it was right about everything. This is where my stance on football in general and women's football in particular meet. Possession percentage is the most deceptive metric this sport has ever produced. A team can grind out 60% of the ball with sideways passes between two centre-backs, and the stat sheet will call it controlling the game. I have watched enough matches to know that most of that possession produces nothing. It only produces a number that looks convincing on a broadcast. In Korean women's football, the possession trap is even more dangerous, because it lets people avoid looking at what is genuinely missing: combinations rehearsed until they become reflex. A team that holds 60% of the ball and loses 0-1 to a corner will be described as controlling the match but unlucky. The truth is they controlled the ball, not the match. And the match was decided by exactly the thing they never practised. There is a question about commercial value I usually dodge on air, but here it needs saying plainly. Women's football is judged by commercial value while its competitive value is ignored. Sponsors want matches with crowds, and crowds want stars. So money flows toward famous players, not toward corner drills at 6 a.m. But if you ask me what will decide a women's national team's ranking over the next decade, I will not point to any star. I will point to whether that team has a coach dedicated to set pieces. The most expensive transfer is not written on the contract; it lives in the gap the player leaves behind. That is also why I do not trust the way the transfer market is reported. The biggest contracts are usually sold to the public as arms races of brand power. The club that pays the most for the most famous player, and the broadcast calls it ambition. But a signing's true value is not in the figure on paper. It is in the gap a player leaves when they go, and the gap a player fills when they arrive. A small club that signs the right defensive player can change an entire pressing system without a single advertising dollar. Back to the 214-match dataset. What satisfied me most was not the 23.7%. It was the fact that the number, once sent, existed in a meeting room I was not allowed into. Someone read it and decided to give more time to set-piece drills. There was no press release for that. No stat sheet recorded it. Only a few training sessions, and perhaps a few goals that will come from a corner whose taker no spectator remembers. For fans, I suggest one small change in how you watch. Next time you watch a women's match, count the layers of movement in a single corner. Do not count the crowd. Do not count the social media posts. Count how many players are actually moving with a purpose rather than out of inertia. That is what separates a coached team from a team that has merely been named. Women's football does not need more inspirational stories. It needs more people who sit down after the match, re-watch the footage many times, and count the things no one pays to count. The change is happening, but it is not happening in the stands. It is happening in the edit room, at 11 p.m., when a quiet craftsman adds one more line to his spreadsheet. And if you are wondering whether those lines mean anything, remember that every set-piece goal begins with a decision made in silence, long before the opening whistle.