The Empty Data Sheet and the Trap of Belief in Volleyball Analytics

The Empty Data Sheet and the Trap of Belief in Volleyball Analytics

At 2:17 in the morning in Nagoya, I opened the statistics file for a men's V.League volleyball match that had finished a few hours earlier. The header row was perfect: both team names, sets, points, spike success rate, blocks, perfect-pass rate, service aces. The body of the file was blank. Not a single line of data. I stared at the screen for five minutes, then did something my fourteen-years-younger self would never have done: I shut the laptop and went to sleep.

The next morning my editor asked whether I could still file. I said yes, but I would write it differently. That empty file could easily have been filled in with professional imagination: recall the feeling in the arena, the sound of the ball hitting the wooden floor, a dig that made the whole gym hold its breath, then attach a few plausible-sounding estimated metrics. In this trade, people do exactly that, sometimes by accident, sometimes on purpose.

Data does not lie, but the person who selects the data knows very well how to lie. An empty file is at least more honest than a file full of real numbers filtered to tell a story that was written in advance.

Volleyball has more data than ever before

When I started writing about volleyball, a match report fit on half a page: who scored how many, how many blocks, how many service errors. Now a single international match generates hundreds of rows of data. Volleyball World's systems log every rally at major events such as the Volleyball Nations League and the World Championship, producing metrics my predecessors never had: perfect-pass rate, out-of-system attack share, attack efficiency split by position, point distribution by court zone, average rally duration.

In Japan, where I have lived and worked for a decade, the V.League publishes post-match statistics almost instantly. A viewer at home can see which outside hitter was most efficient in set three, which team passed worse in the deciding frame. In Vietnam, the national championship and youth competitions now keep their own data sheets, though the level of detail and consistency still trails the major leagues. That gap is narrowing fast.

The problem is not volume. The problem is that the more metrics there are, the more ways there are to misread the game, if readers do not know how those metrics were produced.

The first misread metric: perfect-pass rate

Perfect-pass rate is the most quoted number in modern volleyball discussion. It measures the share of first contacts delivered to an ideal position, allowing the setter to run the full attacking menu. It sounds unambiguous. But the definition of an ideal position is not standardized across data providers. Some count any ball landing inside the three-metre zone. Some require the ball to reach the setter with the body already turned to the net. Some deduct credit if the ball arcs too high even when it lands in the right spot.

One rally, three stat sheets, three different results. And nobody notes which definition they used.

This is the biggest blind spot in modern volleyball analysis. When an article reports that Team A passed perfectly 62 percent of the time while Team B managed only 48 percent, readers have no way of knowing whether the two rates were measured with the same ruler. In fourteen years of watching professional matches, I have seen analyses conclude that a team was weak in reception simply because the data sheet they used was stricter than the opponent's.

Blocks, serves, and what the stat sheet leaves out

Blocks per set is another number that creates illusions. The most common counting method includes touches that slow the ball but still send it back to the opponent. A light touch that disrupts the opposing attack and lets the back-row defenders scramble is credited the same as a clean kill block. Two players with identical blocks-per-set figures can contribute completely different value to their teams.

Serving is harder still. The ratio of aces to service errors is used to judge effectiveness, but it ignores the most important part of serving tactics: breaking down the first pass. A player who misses four serves while scoring only two aces may be doing exactly what the coach asked, if those four misses came with six poor opponent receptions and two blocks that followed. The stat sheet records three errors. In reality he was the biggest source of pressure in the match.

When I watch Yuji Nishida serve, I do not need a data sheet to know how heavy that ball is. The crowd sees the opposing defenders retreat, sees the setter leave position, sees an entire team lose structure in a fraction of a second. No metric captures that moment. That is precisely the gap professional data is trying to close, and it has not closed it yet.

Small samples, large errors

A three-set volleyball match contains only about seventy-five points. An outside hitter may take fifteen to twenty attacking swings across the whole match. With samples that small, one lucky deflection or one refereeing mistake can move a success rate by several percentage points. Publishing a post-match analysis that draws conclusions about an individual's form is a professionally risky move.

To judge properly, you need at least ten to fifteen matches, comparisons against opponents of similar level, and a separation of in-system attacks from out-of-system attacks. That separation matters enormously, because an attacker's success rate depends heavily on pass quality. With a good pass, the hitter finds the gap. With a bad pass, the hitter faces a double block. Two completely different situations, recorded in the same cell.

From Belgrade to Nagoya, I have grown used to noting down the rallies that the data sheet cannot describe. Players such as Tran Thi Thanh Thuy or Nguyen Thi Bich Tuyen on the Vietnam women's national team routinely attack off imperfect passes. Their success rates therefore look lower than their real ability, while a hitter on a team with an elite setter can post prettier numbers without necessarily being better.

The trap of adding more metrics

The default response of the analytics industry when it hits a problem is to add a new metric. Cannot measure service pressure, so create a break-down metric. Cannot measure blocking influence, so create a net-touch metric. Cannot measure pass quality, so create a ball-difficulty metric. Every new metric promises a sharper picture while increasing the number of unverified variables.

The core problem is not the number of metrics, but the consistency of their definitions. One tightly defined metric, with its measurement method published openly and held constant across seasons, is worth more than ten cleverly named metrics that each provider measures differently.

Sports analytics has a persistent temptation: turning data into authority. When someone says the numbers show this, the argument usually stops. But numbers do not generate meaning by themselves. They record what a person, at a particular moment, decided was worth recording. That decision carries assumptions, biases, and the technical limits of the measuring tool.

In Vietnam, most volleyball statistics available to fans are translated from foreign sources or hand-recorded by volunteers in the arena. That work deserves respect, but errors compound: the recorder is not formally trained, the definitions get mistranslated, and then the figure enters an article and becomes something everyone repeats. Mistakes do not correct themselves; they spread.

Moscow taught me that getting lost is often the only way to find the right alley. In 2026, while working in Russia, I took the wrong metro line and missed a press conference. Because of that, I spent two hours in a cafe near the stadium rereading the tournament's stat sheets, and I realized the metrics I quoted daily were being measured in at least four different ways. Since then, every time I cite a figure, I ask who recorded it.

Listening to an empty stadium, I understood that noise was never the audience. The months without spectators during the pandemic taught me the same lesson about data: the noise of hundreds of metrics is not evidence of understanding. A dense spreadsheet can be nothing more than the echo of choices nobody ever checked.

A football player runs more than twelve kilometres in a match, and that distance has never been what makes a stadium rise to its feet. In volleyball, each rally lasts a few seconds, and what people remember is the instant a player launches off the floor. The longest distance is always the one from a player's feet to a spectator's heart, and it has no unit of measurement.

What I want to keep

That night, I did not write the piece. I sent my editor a short message: the data has not arrived, I will file later. He replied with a single line: that is fine.

I tell this story not to advertise my own honesty. I tell it because I believe Vietnamese volleyball fans, after years of following domestic and international competitions, are sharp enough to tell when an analysis genuinely has data behind it and when the writer is only colouring in a feeling. The only thing that sustains trust over the long run is honesty about what you know and what you do not.

Sarina Koga once said at a press conference that she does not read the post-match stat sheet, because she knows exactly which rallies she wasted. I think that is a healthy attitude. Good data helps people see what the naked eye misses, but it cannot replace looking. Volleyball analysis, in the end, remains a human act between humans, merely supported by measurements that are never complete.

That empty file is still in my archive folder. I keep it as a reminder: if there is no data, write about the absence of data. Being honest about the gap is the first step of any decent analysis.