Zero Is Also Data: The Honest Scoreboard of Cricket's Abandoned Matches

13 June 2026, Trent Bridge, Nottingham. The page is still in my notebook — two columns, "shot type" and "pressure index", and the date at the top. By three in the afternoon both columns were empty. India were playing New Zealand in a World Cup group game. Tickets had sold out, the commentary box was ready, the graphics team had built both playing XI cards, millions of fantasy teams had locked. Not a single ball was bowled. The covers came off, and the board showed two words: NO RESULT.

Kane Williamson and Virat Kohli, two captains, entered that day's record without scoring a run or bowling an over. It is the rare moment when the sport formally announces its own absence. An event with no data ends up shaping three weeks of the points table.

What I understood that afternoon is not written in any analytics tutorial: in cricket, "no result" is not an empty space. It is an official settlement. Points are distributed, net run rate shifts, semi-final probability moves, and the audit trail gains a line nobody bothers to read.

Reading the empty column

My habits came out of a football lab. Across June and July 2026, in my first year of a broadcasting degree, I hand-coded all 64 matches of the Russia World Cup from open event data — 1,214 shots, each tagged for body part, assist type and defensive pressure — and built an xG model in R on a laptop in a Koramangala flat with a failing fan. It put France's open-play xG at 1.02 per match, mid-table. Their PPDA of 16.9 was the highest of the four semi-finalists. France won the final 4-2. The lesson transfers directly to cricket: the hardest part of a model is not the cells filled with numbers, it is the cells with nothing in them.

Statistics classifies missing data in three grades — missing completely at random, missing at random, and missing not at random. Cricket's rain data belongs to the third. It is not random; it is bound to geography and calendar. Bristol in June, the Caribbean in August, Colombo in December. The match that produced no data is often not played where most cricket is played.

Something else landed on my desk this week. A feed returned empty — no title, no source, not a single information point. The draft report was about to say "no notable risk found." Empty data and zero risk are not the same thing. One means we do not know; the other means we know there is nothing. A desk that cannot tell these apart makes the same error as a captain who reads rain as parity.

Four grades of absence

Cricket has four distinct categories of not-happening, and each has a different analytical status. Pooling them is the most common mistake in the sport's numbers.

The first is abandonment without a ball bowled, as at Trent Bridge in 2026. Zero observations. Counting it in a team's form series, or folding it into a batter's average, is straightforward arithmetic misconduct. Zero observations do not mean zero performance.

The second is a rain-shortened match settled by DLS. Partial observation, and a result produced by a model rather than by play.

The third is the drawn Test — the strangest category, because the data is complete and the outcome is nil. Five days, more than four hundred overs, two innings, four different pitch conditions, and the final line reads "match drawn." People who call every draw a lack of excitement are really saying they prefer data to outcomes.

The fourth is administrative cancellation. In September 2026 the fifth India-England Test at Old Trafford was called off hours before the start; it was later played at Edgbaston and the series finished 2-2. The cause there was not weather. It was a decision. Rain is a climate; cancellation is a management choice. They belong in separate tables.

Where a model stands in for a match

DLS is the one place in cricket where a match that did not happen is assigned a number. Frank Duckworth and Tony Lewis built the model in the late 1990s; after Steven Stern's revision was adopted in 2026 it became DLS. The core idea is simple: a side's remaining resource is the product of two things — overs left and wickets in hand. The model converts that resource into a percentage and derives a target.

It measures resource, not conditions. A shortened game is not the game it simulates. The batting order changes, because the all-rounder at five suddenly sees ten overs left and plays a different shot. Bowling rotation changes, because a spinner's fourth over evaporates and he is under-used next match. Dew becomes a variable with no seat in the model. DLS is an honest lie: it claims precision when it only offers consistency. And cricket needed consistency more than precision, because the same rule has to bind both sides — however far that rule sits from the actual conditions.

The arithmetic of the points table

The 2026 World Cup had no reserve days in the group stage; that provision existed only for the semi-finals and final. The consequence was structural: rain introduced inequality into the table. Some teams played fewer matches, and net run rate comparisons began between sides whose denominators were different. Put two fractions with different denominators side by side and the table stops being arithmetic. It becomes a scheduling coincidence.

That produced the administrative lesson still visible in the World Test Championship table. When Covid cancelled series and no two teams played the same number of Tests, the ICC adopted a percentage-of-points system, and later cycles kept it. This was not a technical tweak. It was official acknowledgement that cricket's schedule was never comparable; the pandemic merely stripped the disguise off an older inequality. When data goes missing, the cleanest answer is to give every match an equal share. But that is a model too, carrying its own bargain: less exposure for the team that plays less, more opportunity for the team that plays more. The percentage table did not remove unfairness. It changed its shape.

The crowd left, and the truth showed

My most useful study was not the one with the most data. It was the one where a variable disappeared. On 16 May 2026 the Bundesliga returned behind closed doors. Over six weeks I compared the 83 closed-door fixtures of that season with the 223 crowd-backed fixtures from the same season. Home wins fell from 43 per cent to 33 per cent; home goals dropped from 1.74 to 1.43 per game. I then ran the same test on IPL 2026 — 60 matches in the UAE, empty stands — and found chasing sides won 34 of them.

I first called it "the effect of no crowds." That name was wrong. The crowd coefficient was never a number; it was a question. When the crowd vanished, so did every variable that travels with it — umpire pressure, bowler energy, captain's courage, and the steady confidence of a television cut that looks into a foreign player's nerves. None of these can be separated out. What can be done is to accept that absence is itself an intervention: when a variable is removed, the remaining data reports its weight. The same lesson applies to abandonment. Rain took nothing away. It exposed something.

The fifth column of a pitch report

Pitch reports usually carry four items: grass, cracks, bounce, spin. There should be a fifth — abandonment rate, the share of scheduled matches at that venue lost to weather. That number matters to teams as much as the colour of the surface, because it tells you what probability your points are staked on. A side heading twice in a season to a high-abandonment ground is taking on two risks over which it has no control.

No one publishes this metric officially. The broadcast booth hears noise; the notebook hears patterns. Both are watching the same match, but one sees clouds and the other sees a pattern that keeps repeating.

Umpire's call: cricket's other unfinished verdict

Beyond abandonment, there is one more place where the game formally remains unresolved — umpire's call. The ball is hitting the stumps, ball-tracking has an estimate, and the decision still stands because the uncertainty sits outside a fixed margin. This is not a technical defect; it is a deliberately built moat. The system is admitting that some things are measured too thinly to justify overturning anything.

That logic is not applied to abandoned matches. We hand a washed-out game a full row in the points table without asking what it is measuring. With umpire's call we accept uncertainty; with abandonment we convert it into an outcome. Both are rules of the same sport, resting on different philosophical ground.

The other side of the picture

The most attractive story is that rain is the weak team's friend. The reasoning is elegant: in a group stage, one point is worth more to the side with a lower win probability. To prove it, you need the pre-match win probability for every abandoned fixture, and that data does not sit in any public log. Without it you can only show that after some washouts some teams moved up the table — correlation, not cause. A single match can lie; a league table rarely does, but an abandoned match is less than even a single match.

The second misreading is treating "no result" as neutral. It is not. Two teams taking a point each is a settlement, and its value is asymmetric, because positions in a table are built relative to each other, exactly as net run rate is. This is cricket's rare moment where not happening is still an event, and events send someone a bill.

That bill has an economic side the match report never carries. An abandoned fixture is hours of broadcast inventory already sold and never delivered; a refund process for the host; a day's wages for ground staff already committed; and for the fantasy player, an entry that neither won nor lost. No model counts these costs, because they never appear on the scorecard. But money spent this way is part of the system, and behind every abandoned match sits a spreadsheet nobody wants to read — the same spreadsheet that sits behind every transfer fee.

The third error happens at the desk, and it is the most expensive. An empty feed, an empty report, and a reader left believing a check was done and nothing was found. That is the identical mistake committed with abandoned matches: we read absence as a clean bill of health. In Bangalore the data desk is another kind of stadium, and the hardest work there is not producing a number. It is stating which number does not exist.

What to watch next

This argument returns within months. The T20 World Cup in India and Sri Lanka in February-March 2026 will be played in India's dry season. The rain variable shrinks — and dew, heat and schedule fatigue take the seat the clouds used to occupy. Anyone who reads "lower rain risk" as "lower uncertainty" will repeat the error the empty feed made. Uncertainty never disappears. It changes address.

The page is still in my notebook, both columns empty. I have not torn it out. It records that a match did happen — tickets sold, captains on the field, commentators at the mic — and that no data came of it. Cricket's most honest line is the one with nothing written on it.

The question is not who benefited from the rain. The question is whether a match that could not record itself in data should still receive a full row in the table. And if the answer is yes, whether we are quietly admitting that fairness in cricket sometimes only means both sides were denied equally.