The Empty Column: The Decisions Indian Hockey Took on an Incomplete Record

Bhubaneswar, December 2026. The quarterfinal match sheet reached me after the game had ended — I was not in the stands, I was between a screen and a printed sheet. One column was fully filled: India had earned eight penalty corners, and scored one goal. On the same sheet, another column was empty: where the injection came from, who pushed, from which angle the trap was exited, who reached the first rebound. The result was recorded. The mechanism was not.

That night I went back to the broadcast tape and coded all eight corners by hand, because the sheet had no room for them. Three hours later I had a pattern the scoreboard never showed: six of the eight injections came through the same left channel, and the Netherlands' first runner had read that channel after the first corner. The most important thing on the pitch that night was the space nobody entered — and no column measures it, not PPDA, not possession, not shots.

This piece is about that empty column. Almost every current argument about Indian hockey — penalty-corner dependency, midfield speed, governance, the wait for an Olympic medal — stalls at the same place where the record is incomplete. And where the record is incomplete, the decision is made by habit rather than by data.

My habit goes back to 2026, when I started India's first hockey-dedicated website. In those early months all I had was the scorecard: 3-1, 2-2, 7-0. A scorecard tells you who won; it never tells you how. So I did then what I still do now. Before anyone counted, I counted by hand, and that is where the habit began. Twelve thousand articles later, my trust still sits with what I have watched being written down. I do not trust a number I have not watched being written down.

Building that habit in hockey was not easy, because hockey's public archive is thinner than cricket's. Cricket built ball-by-ball databases through the 2000s; hockey still, for many tournaments, keeps only goals and cards. You will find the semifinal score. You will not find which channel the circle entries came through. And the moment you do not know the direction of circle entries, every argument you make becomes an estimate.

In 2026, Kalinga Lancers hired me remotely from Brisbane as the Hockey India League's first dedicated data analyst. The league had no inherited database either. I built a simple penalty-corner model: teams converting above 25 percent were winning 78 percent of their matches. When I put that number in front of the coaching staff, the first reaction was that it was obvious. It was obvious; the next step was not — targeting drag-flicker Sandeep Singh at the auction. He scored eight PC goals that season and the team reached the semifinal.

I want to be careful here, because this story is usually told badly. The 25 percent figure did not say Sandeep Singh was good. It said that whichever team bought the cheapest specialist flicker would shorten its route to winning. The data did not pick a player; the data showed a price market. Without that distinction, analytics becomes publicity at club level and an excuse at national level.

At Rio 2026 I worked as a data consultant with a broadcaster. India lost the quarterfinal to Belgium 1-3. What I coded that night was not the goals — it was the distance, the passes that travelled from one team to the other without interruption. India's midfield PPDA was 15.2; Belgium's was 8.7. In plain terms: India had absorbed nearly twice as many passes before attempting a defensive action. I titled the report 'The Midfield Void' and it travelled widely in Indian hockey circles.

But PPDA is a measure I do not use with full confidence, and that is rarely written down. PPDA measures aggression without the ball; it does not measure the direction the ball was travelling. A team can record a PPDA of 15 because it is slow, or because it is deliberately holding a block and tackling late. Watching the tape, the problem was not the volume of tackles. When Belgium circulated the ball to the left, India's midfield shifted toward the ball, and the channel that opened was never filled. The statistic started the inquiry. It did not deliver the verdict.

At the 2026 World Cup in Bhubaneswar I was embedded with the Indian team as a data analyst. Across the tournament India scored 10 goals from 38 penalty corners — 26.3 percent. The 25 percent threshold was being met. The warning I gave the coaching staff was not about the threshold. It was about variation inside those 38 corners. In my coding the routine was so consistent that an opposition scout did not need a second match to identify it. In the quarterfinal India earned eight corners, converted one, and lost 1-2 to the Netherlands.

Here is the part that costs me something to write. Three matches before the world saw it, the numbers had already named the weakness — but of the 38 corners, I had fully coded only 22 from video during the tournament. For the other 16 I had logged the assist and the flick direction, not the push. When I filled the gap after the tournament, the pattern sharpened and so did an uncomfortable fact: the warning had rested on 58 percent of my own dataset. Even a correct warning can be built on an incomplete base.

Every baseline begins with someone deciding what not to count. In 2026 India failed to qualify for the Beijing Olympics for the first time. The reports that followed recorded goals and defeats; they did not record how many matches the team played in that cycle, under what conditions, or how many preparation days sat in the calendar. A national team does not decline in one match. It declines in the months nobody counted.

Governance works the same way. The road from the IHF's de-recognition to the birth of Hockey India is not a story of sentiment; it is a story of paper — meeting minutes, sponsorship agreements, calendar decisions. Those who read it as a clash of personalities are not reading the column with dates in it. And the dates show that the 2026 crisis did not arrive in a day. It accumulated in the years when the domestic calendar was shrinking under its own administrators' decisions.

Odisha belongs in this ledger, because it is the least-written part of the story. The 2026 World Cup, the 2026 World Cup, Kalinga Lancers, and the return of the Hockey India League in 2026-25, when the Birsa Munda Hockey Stadium in Rourkela filled. For years Odisha performed the role that belonged to the federation on paper and to nobody in practice. Calling that generosity is easy; calling it a system is harder, because a system means the sport does not ride one government's budget cycle.

One correction on the league is necessary. The HIL stopped after the 2026 season and returned in 2026-25. Seven seasons. Through them Indian hockey had no continuous high-level domestic data. When the league stopped, the read was that demand was absent. In 2026-25, the ticket line in Rourkela challenged that read directly. The question was never demand; it was broadcast, calendar and cost. When a series stops, its data stops too — and the absence of data is then mistaken for evidence of absence. That is the oldest error in statistics and the most common habit in sports administration.

There is another empty column, and it is gendered. At Tokyo 2026 the Indian women's team lost its semifinal 1-2 to Argentina, then lost the bronze match 3-4 to Great Britain. That tournament received less analysis for its semifinal than the men's defeats received for theirs. This is not a complaint; it is a measurement. Coverage is also a dataset, and women's hockey still holds a small column in it. A sport with a thin record forces its next generation to rediscover its own history.

Now the part where I argue with my own method.

The three examples above — 26.3 percent in 2026, 15.2 PPDA in 2026, 78 percent in 2026 — set the same trap. These are correlations, not causes. Teams converting above 25 percent won 78 percent of matches; that does not mean conversion creates wins. The reverse is equally true: good teams create more corners, and leading teams are awarded more corners, so their conversion looks better. India stood at 26.3 percent in 2026 and went out in the quarterfinal. The number was right; the inference drawn from it was wrong. The anomaly usually arrives on time; it is the explanation that runs late.

Second, an empty column is not evidence. Saying the record was incomplete does not mean the decision taken was wrong. The staff saw things that night my tape did not hold — a player's breathing, a dressing room's mood, an injury. What is missing from data is a gap in data, not a gap in the world. My claim is narrower: decisions taken without a record cannot be audited afterwards. That is an administrative failure, not a technical one.

Third, and this is aimed at my own community. The 'eight Olympic golds' is a measurement imposed on players who never watched that team play. The recorded score of the 2026 Berlin final is 8-1; the scorer breakdown is where the record thins, and folklore begins there. Balbir Singh Sr.'s three golds and his five goals in the 2026 Helsinki final deserve the same documentary weight as 2026. The 2026 Rome silver and the 2026 Kuala Lumpur World Cup are line items in the same ledger, not commentary. The 41 years between the 2026 Moscow gold and the 2026 Tokyo bronze is a number to be read through dates, not emotion.

Fourth, the least comfortable point: geography. I am writing this from Brisbane. I did not stand in that stadium and I did not hear that crowd. My base is the feed, the official sheet and my own notebook. Distance has an advantage — I am spared the sentimentality that blames 'the system' after every defeat and forgets it after every win. It has a cost — what I cannot see never enters my ledger. So I depend on reporters on the ground, and naming them is my habit.

So what should be recorded next? These are administrative demands, not technical ones. First, a full record of every penalty corner: injection point, pusher, exit angle, first rebound. Second, PPDA for every match with its context — who was leading, who chose to block. Third, a continuous record of the domestic calendar, so that the next time a league stops, the demand question is settled by data rather than by guesswork. Fourth, equal depth of data for the women's and men's teams, because a team with a small record will always have a small story.

And the next time a team takes eight corners and scores one, the question should not be what the conversion rate was. It should be how many of the eight came from the same routine. A team can hide behind one routine for years; the data needs one night to show it — provided that night was recorded.

The column that was empty on that sheet in December 2026 is still empty on many sheets. The tape remembers what the scoreboard forgives. The question is not who won. The question is who accepted the job of filling the column in.