Vietnamese Volleyball's Data Gap: Why Every Tactical Analysis Is Still a Guess

Second set, 14-13. I turned to the data officer sitting on my left and asked for the home team's perfect-pass rate up to that point. He flipped through an A4 sheet, looked at it for about two seconds, and shook his head. On the sheet there were only the score of each rally, the number of successful spikes by the two main attackers, and a few scrawled notes about a rally I had shouted about in the previous set. No pass rate. No out-of-system attack rate. No footwork map of the blocking pair. No record of where the libero stood in the last ten defensive rallies. I switched on the microphone and carried on with the only thing I genuinely had: my eyes, and more than forty years on the floor.

After the match, a viewer messaged my fan page: "You sound so certain in your analysis, but what can we use to verify it?" I read that message three times, then replied with one short line: "Nothing. That is exactly the problem." That answer bothered me for weeks, and it is the reason this article exists. In Vietnamese volleyball we argue passionately about tactics on a foundation of data that is close to empty — and the danger is not the emptiness itself, but that so few people notice it is empty.

We measure outcomes, not causes

What does the statistics sheet published after each match in the national championship contain? Each player's points. Successful spikes. Successful blocks. Successful aces. Four columns, all of them output data. They tell you what happened, not why it happened, and certainly not what has to change for it to happen differently.

The variables that decide those four columns are recorded by nobody. The quality of the first pass. The number of attacking options the setter genuinely has after each pass. The number of blockers the attacker faces on each rally. Where the libero and the two outside hitters stand on each defensive rally. The time between the ball leaving the blocker's hand and the ball being set for the counterattack.

At international level, the International Volleyball Federation (FIVB) and the Asian Volleyball Confederation (AVC) publish far more detailed statistics. But most of what is published is still individual leaderboards: best scorer, best blocker, best setter, best libero, best server. That is a roll call. A roll call tells you who is leading. A blueprint tells you why they are leading and what everyone else must do to catch up. We read the roll call and call it tactical analysis.

Based on my experience covering matches across eight Olympic Games, eight World Cups and more than twenty years of live commentary on volleyball finals, the biggest gap between a strong volleyball nation and a rising one is not height or training hours. It is that strong nations know precisely where they are losing, while we mostly know only that we lost.

Writing a blog on an old laptop, I never imagined I was typing experiments for fate. In 2026, at 52, after leaving my coaching chair, I started dissecting matches with a notebook and hand-drawn diagrams. My first piece ran two thousand words and drew five readers in a week. But in that process I realised something: I could draw the diagram, but I could not prove it right, because there was no data to check it against. It felt like being a watchmaker working blindfolded.

The first pass is the master variable, and we barely measure it

Start with the most basic causal chain in volleyball.

A perfect first pass delivers the ball to exactly the spot the setter needs, at exactly the right height and speed. The setter then holds the full menu: a quick middle at position 3, a quick behind the setter, high balls to the outside hitters at positions 4 and 2, back-row attacks from positions 1 and 6. With four or five options, the opposing block must guess, and when they guess they are a beat late. That beat is the entire distance between a blocked ball and a point.

A poor first pass collapses everything. The ball is pushed too far off the net or too close to it. The setter loses the quick attack and loses the back-row option. The menu shrinks from five items to one: a high ball to the outside hitter. And with only one item on the menu, the opposing block no longer has to guess — they simply stand in the right place and wait. The attacker then faces two or three blockers, hits into a sealed space, and if lucky the ball touches a block and bounces out.

What matters is that in the published statistics, both rallies above are recorded identically if the attacker does not score: one unsuccessful attack. The viewer concludes the attacker's form has dropped. A slightly sharper analyst concludes the offence is stuck. Both conclusions are correct on the surface and useless at the level of intervention, because the real cause sits in the first pass — a variable that appears in no statistics sheet.

Attack efficiency is not an independent variable. It is a consequence of first-pass quality and rotation structure. Anyone debating attacking play without those two inputs is discussing the tip of a tree without knowing where the roots are.

Now, rotations. Each team has six rotations, and each rotation creates a different spatial structure: how many attackers are in the front row, how many can attack from the back row, where the setter stands, where the libero stands. In some rotations a team has only two front-row attackers. In that rotation, if the team has no back-row attack option, the opposing block only has to manage two threats instead of four. That is an arithmetic advantage, and in elite volleyball arithmetic advantages are converted into points with a regularity that becomes cruel.

Every crisis on court begins with a mistimed beat that nobody wants to look at. A team concedes four straight points in rotation four of the third set, and the whole arena assumes it is about spirit. But with rotation-level data you would see that attack efficiency in that rotation ran a consistent gap below the other three rotations all season. That is not a spirit problem. It is a lineup design problem, and it can be fixed by rearranging the passers or by teaching an outside hitter to attack from the back row.

Then there is the lane. A match is won in the lane, lost in the lane, and very few people actually see that lane. The lane here is the seam between the outside blocker and the middle blocker, the gap between the block and the sideline, the patch of floor behind the block when the libero stands at the wrong angle. Those spaces are a few dozen centimetres wide and exist for about a quarter of a second. To analyse them seriously you need to know that on rally thirty-seven of the second set the block stood a certain number of centimetres off its reference position, and that the libero started early or late. Without data, every claim about the lane is selective memory.

So what is the minimum dataset? I propose six groups, and I want to be clear that this is a design proposal, not a wish list.

Group one is the first pass, graded in four levels: perfect, good, limited, error. Each pass is tagged with its court zone and the rotation in progress. This is the most important group and also the easiest to record.

Group two is attacking, but recorded along two axes at once: by rotation, and by system state — in-system or out-of-system. In-system means the rally was built from a pass good enough for the setter to run the playbook. Out-of-system means the rally had to be salvaged by individual quality after a poor pass. Merging those two into a single "attack efficiency" column is the most common error in every existing statistics sheet.

Group three is blocking, and not just blocks scored. Count block touches, count double blocks formed, and count the quality of the blocker's footwork — specifically whether the blocker closed the block in time or only got one hand on the ball.

Group four is defence, covering dig position by zone and the share of digs that turn into organised counterattacks. A libero who digs a lot of balls but whose team always counterattacks out of system is hiding a bigger problem behind his dig count.

Group five is serving, split into three indicators: direct aces, errors, and the share of serves that break the opponent's first-pass quality. The third is the most important and is almost never published.

Group six is transition, measured as the share of block or dig situations converted into points within three touches.

In terms of manpower, these six groups do not require an analytics department. Two people with two tablets, one on each side of the court, are enough — provided both use a single definition sheet agreed before the season. Vietnam's volleyball problem has never been a lack of people. The problem is that nobody has been assigned to record, and nobody has defined what a perfect first pass means in the same way.

This is the most important part, so let me be blunt: a unified definition sheet is worth more than expensive software. If a tournament organiser publishes standard definitions for the four pass levels and requires every team to record against them, we can compare data across matches, rounds and seasons. Without shared definitions, whatever we collect is just islands that never connect.

Suppose that dataset existed. Coaching decisions would change in very concrete ways.

First example: if you find your rotation four has the lowest attack efficiency and the opponent has a matching weak rotation, you adjust your serving targets to keep them in that unfavourable rotation for as long as possible. That is rotation management, and it only exists when you have rotation-level data.

Second example: if you find an outside hitter ranks among the league's best in-system but among the worst out-of-system, the right conclusion is not "he needs to spike more in training". The right conclusion is "we need to improve first-pass quality where he receives, or reduce the number of passes he has to take". The same player, two completely different conclusions, and the difference lies in whether you separated the two attack types.

Third example: if you find your libero stands about forty centimetres too deep on balls hit down the line, you can fix that in two sessions with reading drills. It is the cheapest intervention in this sport: fix the starting position, no fitness work, no technique overhaul.

What all three examples share is that none of them can be derived from the current statistics sheet. The current sheet only tells you the final outcome of a long chain of decisions. To fix the chain, you have to record every link.

The most comfortable explanation is usually the one that cannot be wrong

One explanation is repeated so often it has become the default whenever Vietnamese volleyball loses internationally: physique. Vietnamese people are short and cannot compete with the height of the big volleyball nations.

I do not deny that height is a real constraint. It caps the ceiling of a volleyball nation. But a ceiling constraint does not explain the distribution of points within a specific match. Within a match, points are decided by how often you control the first pass, how often you force the opposing setter to scramble, and how often your block closes the lane. None of that is about height.

Moreover, the physique explanation has a very convenient property: it cannot be wrong. If the team loses, it is correct. If the team wins, people say the team overcame its physical limits, and it is still correct. An explanation that can never be wrong is not a scientific explanation, it is a mantra. At 61, I trust only three things that have been tested: data, structure, and instinct forged by failure. Height is none of the three.

A testable version of the physique argument would sound different: with two teams of equal height and equal training hours, the team with the more stable first-pass system will win more than sixty per cent of transition rallies. That statement can be measured, can be falsified, and can be used to design drills. The standard physique argument leads to no drill at all.

There is a second blind spot that I consider more serious and far less discussed: the risk of publishing data.

If you publish each player's perfect-pass rate, coaches will start teaching players to pass safely rather than effectively. Players will favour passes that are reliably graded "good" over risky passes aimed at position 3 to open the quick attack. Their numbers improve while the team's offence gets worse. This is a general law of measurement systems: people optimise what is measured, not what is wanted.

The fix is not to withhold data but to publish the right indicator. A first pass should only count as perfect if it opens at least three different attacking options for the setter. That definition fuses pass quality with tactical purpose, and it cannot be gamed with safe, square, harmless passes.

The third blind spot sits on the media side, and I include myself in that. When an attacker misses repeatedly, the headline is a drop in form. When a team loses three sets in a row, the headline is a mental crisis. Both headlines are easy to write, easy to read, easy to argue about — and both ask the wrong question. Instead of asking whether a player is in form, the right question is how many options the team's first pass is giving the setter per rally.

Here I want to be honest about one of my own limits. In 2026, during the semi-final between France and Belgium, I said on air that France would win 1-0 through a lightning counterattack, and the result was exactly that. The clip spread to more than a million views. But there is a vast distance between a correct judgement and a repeatable method. I was right once, based on seeing Matuidi drop deep to lock down De Bruyne and Griezmann operating as a second number ten. That was observation by eye, not a dataset that others can verify, re-test and apply to another match. I was lucky. And in an industry where everything can be recorded, relying on luck is a poor strategy.

That is why I now open each of my pieces with a specific situation rather than a general claim, and try to close with a verifiable conclusion. But I admit: most of those conclusions of mine have never been checked systematically, because there is no baseline data to check against. We are doing criticism in a room where someone has taken away all the rulers.

A test for the next two seasons

I want to offer a falsifiable prediction, tied to a deadline, so that four years from now I can come back and see whether I was right.

My prediction is this: if for two consecutive seasons the national volleyball championship publishes two indicators — perfect-pass rate by rotation, and out-of-system attack rate by player — the ranking of the top four will shift in a determinate direction: the team with the highest perfect-pass rate in season one will finish in the top four in season two, even if that team does not own the most highly rated attacker in the league.

This is a statement that can be wrong. If at the end of season two the leader in perfect-pass rate is still in the bottom half of the table, my hypothesis is refuted, and I will have to write another piece explaining why I was wrong. I have no intention of dodging that. Treating a wrong prediction as an experimental result is the only way to keep this analytical work meaningful.

There are three checkpoints to watch. The first is whether publication of the data happens at all. The second is whether teams actually use the data to change their reception alignments and rotation structure — this is observable through shifts in where outside hitters stand between two seasons. The third is whether a hitter who once had a low out-of-system rate improves markedly after his team improves the first-pass quality around him.

I know someone will say that producing data is the federation's job, not a commentator's. But volleyball is a sport in which every tactical debate, whether on television or in a coffee shop, rests on the implicit assumption that we are watching the same match. We are not watching the same match. We are only watching the same score.

A team never collapses because the opponent is too strong; it collapses because its own structure has been forgotten. And a volleyball nation does not collapse for lack of data — it simply stands still in exactly the same place, season after season, while other volleyball nations have taken a few more steps down the same road. The question for next season is simply this: will anyone sit down, pick up a tablet, and record the first pass.