Perfect Pass, Spike Efficiency and the Blank Cells in a Volleyball Stat Sheet

Last season I kept a habit of photographing the technical stat sheet after every set at the Italian top-flight championship. One match made me sit longer than usual: the team that won 3-1 was the team with the lower spike success rate. The person next to me pointed at the screen and asked exactly one question — "Is the data wrong?" The data was not wrong. The error was in the reading: we were comparing two metrics that share a near-identical name but are calculated in completely different ways.

Since then, whenever I pick up a volleyball stat sheet, I do one thing first: establish how that metric is defined, who published it, and how large the sample is. Data never lies; only readers hurry. The rest of this piece is how I work through the four foundational metrics of modern volleyball, and the places where a stat sheet looks solid but is actually hollow.

Four metrics, four tiers of source

Volleyball has a quirk that makes reading numbers harder than in football: every rally ends in a point, so any metric can be polished by choosing the denominator. The four metrics I use most are perfect pass rate (Perfect Pass%), spike efficiency, blocks per set, and ace-to-error ratio.

Perfect Pass% measures the share of first passes delivered to the ideal position, enough for the setter to run the full tactical menu. It is the root metric of the whole defence-to-counterattack system. Spike efficiency is (spike points − errors − blocked) divided by total attempts, unlike success rate, which only counts points over total attempts. Blocks per set is a normalising metric that allows comparison between matches of different length. Ace-to-error ratio measures the trade-off between value and risk in an aggressive serve.

Alongside the four metrics sit four source tiers: the world federation and the body operating international competitions, national league organisers, the press, and self-published sites. The source tier sets the maximum confidence of everything downstream. A stat table taken from a self-published site may be correct, but it does not permit me to conclude at a high level.

There is a technical problem few people notice: in volleyball, most detailed stat tables are designed as graphics rather than text tables. When data is collected automatically, those graphic blocks are often skipped entirely. The result is a stat table that looks complete but is in fact empty — and that is the life-or-death difference between "blank cell" and "zero".

Transfer news uses a different frame of reference entirely: the International Transfer Certificate (ITC) and the foreign-player quota decide whether a player can take the court at all, before any metric becomes meaningful. Reading a transfer item while ignoring those two facts makes every metric behind it worthless.

Reading each metric: what to trust, what not to

Start with Perfect Pass%. A team can post a very high "successful" first-pass rate and still never run a quick attack, because the pass landed half a metre from the ideal spot. Many stat sheets carry a single first-pass column that merges all three quality levels. When I see that column, I always ask: does the publisher separate the "perfect" tier? If not, the metric only says the ball did not hit the floor; it says nothing about how many options the setter had. A setter like Simone Giannelli can run almost the entire tactical menu from a perfect pass; from a pass half a metre off, that menu shrinks to two options.

Next comes the most confusing pair: spike success rate and spike efficiency. An attacker takes 20 swings, scores 9 points, commits 5 errors and is blocked 3 times. Success rate is 45%; efficiency is 5%. Same player, two readings, two opposite conclusions. At team level, this explains why a side can win with a lower success rate: they swung fewer times but barely erred, while the opponent swung a lot and lost points in rallies nobody records in the "points" column.

Blocks per set fixes a very common comparison error: adding raw block counts across a match. A five-set match always yields more blocks than a three-set match. Normalising by set is mandatory; without it, every cross-match comparison is meaningless.

Ace-to-error ratio is my favourite metric because it exposes the trade-off most clearly. A team with 8 aces and 12 service errors is not a better serving team than one with 5 aces and 4 errors. But look only at the ace column and the first team looks far stronger. This metric is also where data gets distorted most, because service errors are often recorded in a different table from direct service points.

Then there is the thing a stat sheet never shows: rotation sticking. That is when a team repeatedly fails to side-out within one rotation while the opponent piles up points — usually in a rotation with only two attackers in the front row. No column records "this team lost five straight points in rotation four". To see it, an analyst has to read the video alongside the score sheet. This is where my first-hand watching experience is used most: I do not argue with emotion; I argue with sample size. And one stuck rotation in one match is not yet a sample.

The standard fix for a stuck rotation is the 2-for-3 substitution: bringing the backup setter and an opposite in for the middle blocker and setter in the front row, preserving three front-row attacking options. It is a decision built purely on rotation data, but it is only right if the reader knows exactly which rotation the team is stuck in and why.

The counter-intuitive angle: missing data is not bad data

There is a subtler mistake than any metric error: reading the absence of data as evidence. A team with no recorded blocks is not a team that cannot block. It is entirely possible that the block section for that match was never loaded, or is trapped inside a graphic file. The principle I keep: absence of evidence is not evidence of absence. A blank cell and a zero are two entirely different things, and merging them is the most common error in the volleyball analysis I read.

The second consequence: correlation is not causation. A team with a high Perfect Pass% usually wins a lot, but not necessarily because of it. They may simply be the team with the better middle-blocking unit, forcing opponents to serve more safely, and that is what keeps their first pass clean. When two metrics move together, I always ask which one is the cause and which is merely the consequence of a third variable.

World Cup 2026 taught me a lesson: a model does not need to be big, it needs to be right. That lesson applies to volleyball exactly the same way. A model built on four precisely defined metrics is worth more than a model with twenty metrics, each drawn from a different, unverifiable source. Error is not the enemy; it is the silent teacher of every model. But error only teaches something when you know where it came from.

The signal for the next round

For the rest of the season, what I track is not which team currently has the highest spike efficiency, but the Perfect Pass% gap between the first set and the last set of the same team. That gap is usually a fitness signal, or a signal that the opponent has read the serve direction — and it shows up before the standings change. If you only have time for one column each week, watch that one.