The Empty Cell in the Volleyball Statistics Sheet

Fifth set. Score 13-13. The coach signalled a substitution, and a backup setter walked into position six after four months on the bench with an ankle injury. Her first rally ran eleven touches. The home side took the point, then three more to close the match. A small stand in northern Thailand erupted in a roar I still hear years later.

The next morning I opened the official tournament statistics sheet. Her row had exactly two filled cells: substitutions used and minutes played. Everything else was blank. No setting index. No distribution index. Nothing to prove that across those eleven touches she had read the opposing block three times and changed the direction of her distribution all three times.

I have kept that sheet. It taught me a working rule: a statistics sheet does not record what happened, it only records what the system managed to capture.

There are mornings like that in this trade. And there are other mornings, when I open an analysis file a colleague forwarded and find every field empty: no title, no source, an empty information list, no entities, no time-sensitivity tag, source quality unassessed. A nine-part report, neatly laid out, with not a single line of content to take apart. That report was not wrong. It was just honest to an uncomfortable degree.

Every starting line is an anonymous Sunday. To me, an empty data file is the same. It is not proof that nothing in the match mattered. It is proof that something broke between the arena and the server.

What volleyball measures, and who measures it

Volleyball carries one of the densest statistical systems of any team sport. The International Volleyball Federation runs a competition information system that consolidates data from events inside its structure, including the Volleyball Nations League, the annual commercial competition launched in 2026 and a key battleground for ranking points. At club and national level, most organisers use dedicated logging software in which every rally is coded in sequence: who served, where the ball went, who received, where the second contact travelled, who attacked, and how it ended.

A reader needs to separate three tiers of data. The first is verified official data published by the organiser, the continental confederation or the world federation, usually accompanied by a match report and a certified scorer. The second is semi-official data compiled and published by a club or broadcaster, sometimes counting only points while ignoring context. The third is community data, logged by fans. All three have value, but they do not share the same reliability, and blending them in one article without labelling the origin is the fastest way to destroy your own credibility.

In many national leagues the gap between those tiers is wide. Some matches have only a handwritten scoresheet. Some have dedicated software but not enough experienced operators. Some have their data entered after the event, meaning the error is frozen into history. Indoor volleyball has been at the Olympics since Tokyo 2026 and beach volleyball since Atlanta 2026, but the data infrastructure of the two branches is not alike: beach has two players per side and fewer variables, while indoor rotates six positions continuously and can produce hundreds of combination patterns in a single set.

In 2026 the women's world championship was hosted by Thailand and the men's by the Philippines. For anyone working in Southeast Asia, that was a rare chance to place international data infrastructure beside regional infrastructure. Based on my experience watching matches at both levels, the biggest gap is not in the software. It is in the number of people trained to sit and log long enough, patiently enough, and with enough understanding of the game to tell a good first contact from a lucky one.

Four metrics that tell a story

The first metric worth discussing is the perfect-pass rate, the share of first contacts delivered to the ideal spot, allowing the setter to run organised combinations. It is graded on a scale from poor to perfect, and only the highest grade counts. The perfect-pass rate does not measure defensive skill. It measures ball control, and those are different stories. A libero can save impossible balls and still post a low perfect-pass rate, simply because most of what she touches is difficult.

When that rate falls below a threshold, the whole attack changes shape. The setter loses the ability to run the quick middle, is forced to push the ball to the wing, and the outside hitters face a two-player block already in position. Spectators see a team attacking badly. Analysts see a second contact squeezed into a corridor thirty centimetres narrower than usual.

The second metric is attack efficiency. Many sheets report only the kill rate, points divided by attempts. But the kill rate hides errors. An outside hitter with thirty attempts, fifteen kills and nine errors is worth far less than one with thirty attempts, fourteen kills and three errors. Efficiency, points minus errors divided by attempts, shows who genuinely creates an advantage and who is merely burning possessions. It is the first number I check whenever someone tells me a hitter is in great form.

The third metric is out-of-system scoring, attacks that follow a broken first contact, when the setter has no option but to push the ball to an attacker in a bad position. These are the hardest points in modern volleyball and the ones the sheet treats most unfairly: they are counted exactly like a kill from a perfect combination, even though their tactical and psychological value is far apart. A team can win a set on three out-of-system swings and then lose the next three because it cannot repeat them.

The fourth is blocking and defence. People count stuff blocks per set while ignoring block touches that slow the opponent's tempo. A block touch is not a point, but it turns an organised rally into a scramble, and in volleyball a scramble usually favours the defending side. Dig success, the share of balls kept alive against an attack, should be read alongside the block count. Reading either in isolation leads to a wrong conclusion.

Serving deserves its own section. The ratio between aces and service errors carries growing weight in the modern game, as the jump serve has become a weapon for breaking the reception system. A team with twenty aces across a tournament but sixty service errors is gifting the opponent more points than it earns. I have watched sides take the first two sets on aggressive serving and lose the next three as the error rate spiked with fatigue.

The situations the sheet forgets

Two tactical situations almost never appear on a statistics sheet, even though they decide many matches.

The first is the 2-for-3 substitution. A coach spends one substitution to bring on a backup setter and an opposite, preserving three attacking options in the front row. Technically it is a complex positional swap that requires the backup setter to understand the defensive system step by step. Statistically it is nearly invisible. The backup setter plays two rotations, touches the ball four times, scores nothing, commits no obvious error, and the sheet records exactly two cells, as in the case I opened with.

The second is the stuck rotation, when a team repeatedly fails to score while the opponent runs a streak, usually because an outside hitter and a middle blocker land in unfavourable positions at the same time. In those minutes a coach must choose between using a timeout to break the run or letting the team work it out. The final sheet will show a team losing a set by five points. It will not show that the team lost six straight points in the fourth rotation.

Something else goes missing too: the international transfer certificate. Every cross-border move by a player requires that document, confirmed by the federation of origin and accepted by the destination federation. It appears in no statistical table, yet it decides whether a player can take the court in the next round. A transfer does not buy a player, a club buys a stretch of lost time. And that stretch only starts counting once the paperwork closes.

In Vietnamese volleyball these stories are unfolding on home courts right now. The current generation includes names such as Tran Thi Thanh Thuy and Nguyen Thi Bich Tuyen, alongside a group of young players promoted to the national team earlier than usual. When a leading hitter moves abroad, her data at home and abroad do not share a frame of reference. Points scored in a league with full logging cannot be compared directly with points scored in a league that keeps only a paper scoresheet. The writer has to rebuild the frame, and that is the part of the work nobody sees.

An empty cell is not a zero

Back to the blank analysis file from the start. The strongest temptation when holding such a file is to fill the gaps. A writer can reconstruct a lineup from memory, assign a plausible perfect-pass rate, write a sentence about form based on feel. The resulting article reads smoothly. It is wrong in exactly one respect: the entire data section was invented by the writer, and the reader has no way to know.

In this trade I distinguish two kinds of blank. The first is blank because there was nothing to record: a postponed match, a cancelled event, a field that does not apply to the branch under review. The second is blank because collection failed: the record was never fetched, the data was never parsed, the format was never recognised. The two look identical on screen but mean opposite things. The first is information. The second is an incident. Confusing the two is the most serious mistake a data writer can make.

One small detail in that file deserves a pause. The source article type was flagged as unclassified. In operational practice that label usually accompanies a failed fetch or parse rather than describing a genuinely content-free article. And if the source was in fact a data-rich piece, losing the entire input means every downstream conclusion loses its footing. In that case, stopping and demanding a re-run of the collection stage is the professional act, not an evasion.

Another blind spot is rarely mentioned: the branch. An analysis labelled simply volleyball, without stating indoor or beach, cannot apply the right competition system, cannot apply the right scoring conventions, and cannot compare against the right opponents. Beach volleyball has two players per side, no dedicated setter, no libero, and a completely different convention for counting attack points. Misrouting the branch produces an analysis that reads beautifully while describing a different sport.

Controlled scepticism means aiming doubt at the right target. Doubt the data, doubt the source, doubt the logging process. Do not doubt the people without evidence. In the case of that blank file, the target for questioning is the data pipeline, the ingestion log, the format check. Players, coaches and the people logging late at night in an arena still deserve the kindness they have always received.

The tears on the Tokyo track are the only thing a stopwatch cannot measure. That holds for volleyball in its own way. A setter's four months on the bench appear in no cell. Neither does the first touch after an injury. The only thing recorded is minutes played, a dry number incapable of retelling anything.

Reading volleyball without letting data lead you

My method is simple. Before a match I set three specific tactical questions: can the home side hold its perfect-pass rate high enough to run the middle, which hitter will be forced to carry the out-of-system load, and are the middle blockers quick enough to touch the ball before it crosses the net. After the match I read the sheet to answer exactly those three questions, and only then read the rest.

This keeps me from reading statistics top-down, starting with the summary row and then hunting for data to justify a conclusion I already held. It also helps me spot the places where the sheet answers nothing at all. Three questions, three gaps, and that is precisely where direct observation by eye is required.

People go to a stadium to see who wins, then realise they are watching who becomes. A statistics sheet can only answer the first half. The second half belongs to the one who stays after the roar dies, folds the notebook, and wonders why none of those eleven touches had a cell to mark.

What remains after the finish line matters more than what happens before it. For anyone working with sports data, that remainder begins with a very modest question: is this file blank because nothing happened on court, or because we did not look carefully enough?