No Verdict Without Evidence: Cricket's Officiating Architecture, Data Integrity and the Case for Immutable Records

On 14 July 2026, at Lord's, England needed fifteen runs off the last over of the World Cup final. Off the fourth ball, Martin Guptill's throw struck Ben Stokes' bat and ran away to the boundary rope. Umpire Kumar Dharmasena awarded England six runs — two for the running, four for the boundary. Those six runs dragged the match into a Super Over, and the Super Over took England to the trophy on boundary count-back.

After the match, Dharmasena admitted in an interview that under Law 19.8 it should have been five runs; the batsmen had not completed the second run when the throw was released. But inside the match there was no route to that conclusion. Whether four boundary runs should have been credited was not a question any review trigger could reach. The fate of a World Cup was settled by a decision with no chain of evidence behind it. That gap is the centre of this discussion.

The history of evidence management in cricket does not stop at that one match. In July 2026, a Test between India and Sri Lanka at Colombo's Sinhalese Sports Club ground introduced the Decision Review System. Since then the system has been rebuilt several times. Under ICC playing conditions, teams get two reviews each in Tests and ODIs and one in T20Is. A successful review is returned, a failed review burns away. A review is therefore itself a scarce resource, and where to spend it is a captain's tactical call — with emotion and arithmetic running side by side.

I entered this system through football, in 2026. At the Confederations Cup, referee Milorad Mažić spent four minutes on a VAR review of a penalty appeal involving Alexis Sánchez in Chile versus Cameroon. I forgot the result and instead wrote a 4,200-word timeline of those four minutes, with IFAB's 2026 VAR protocol in hand. A year later came Ivan Perišić's handball and Néstor Pitana's penalty in the World Cup final in Russia, and then a tally of 29 VAR reviews and 20 changed decisions across 64 matches — my interest drifting slowly from the question to the protocol. In 2026, sitting through empty stadiums, five substitutions and lockdown football, I coded 500 referee decisions from the 2026-20 season into a spreadsheet, clause by clause under Law 12 and Law 3. At Qatar 2026, semi-automated offside cut a check down to about 25 seconds.

That mapping does not transfer cleanly to cricket, and care is needed here. In VAR, overturning an on-field decision requires a clear and obvious error — the burden of proving the referee wrong sits on the technology. In DRS the burden runs the other way: the batsman or the fielding side seeks proof, and the umpire's on-field call stands as the default that fills the absence of proof. VAR has both an on-field review track and a coach's challenge; DRS has only a player request, which begins the process once the umpire approves it. In one system doubt is the technology's burden; in the other it is the player's.

At the third umpire's desk there is a fixed decision tree — whether the catch was completed, whether the ball was under control before touching the ground, the line of the stumps, the ball-tracking plot, audio sync, and the arithmetic of time. One question per step, two answers per question, and a defined route back to a state of insufficient proof.

Umpire's call is not a concession; it is an acknowledgement of the limits of proof. On an LBW review, even if ball-tracking shows the ball would have hit the stumps, the decision does not change unless the point of impact falls inside a defined tolerance band from the edge of the stump. The logic of that band is simple: the tracking technology carries its own measurement error, and no one's verdict can be overturned on evidence finer than that error. What the system is really saying is this — we have data, but its reliability has not reached the required threshold. That is the category of zero evidence: neither out nor not out, but no proof sufficient to rule, so the original call stands.

Which clause an incident is filed under determines which protocol applies. At IPL 2026, Ravichandran Ashwin, captaining Kings XI Punjab against Rajasthan Royals, ran out Jos Buttler at the non-striker's end before he had left his ground. In the letter of the law the act was legal, but it then sat in the unfair play chapter under Law 41 — and that classification is what dragged the incident into a moral frame. In October 2026 the ICC moved the dismissal out of that chapter and into Law 38, the run out. The words of the law did not change; the address did. And when the address changes, the protocol, the match referee's report and the code-of-conduct process all change with it.

There is a subtle lesson buried here. The rule often stays the same and only its residence moves. A decision that once carried the label of being against the spirit of the game becomes, once relocated, a merely procedural dismissal. Moral outrage does not disappear; its address does — from the player's character to the language of the regulation.

In June 2026 the ICC removed the soft signal, effective from the World Test Championship final. Previously, on a low catch, the on-field umpire gave a preliminary verdict, and the third umpire carried the burden of clear evidence to overturn it. With the soft signal in place, the field call was the default; with it gone, the default became not out where proof is absent. Note what did not happen: accuracy did not increase. What changed is whose shoulders carry the doubt.

Cricket's system also contains a human valve. At Trent Bridge in 2026, after Ian Bell was run out, India's captain MS Dhoni withdrew the appeal with the umpires' consent. That valve has a name in the law book, but its use rests entirely on the decision-maker's judgement. In a protocol-driven system that space is small, but it is not zero — and it is precisely the space that generates the most argument.

Where evidence is never recorded, technology can do nothing at all. Everything we know about Lord's 2026 we know from television replays and a later interview. The review room's inputs — which angle, which frame rate, which audio sync, which line of reasoning produced which decision — are not written into any verifiable, timestamped ledger. So the same incident stays disputed years later, because reconstruction means assembling memory, replay and interview.

Blockchain is one possible answer here, but its limits need stating. In 2026 the ICC launched digital collectibles called Crictos with FanCraze; in football, club fan tokens on the Socios-Chiliz model and FIFA's FIFA+ Collect sit at the level of fan economy. Cricket's real deficit is not commercial but officiating: an immutable, timestamped record in which every review's inputs and outputs are appended as hashes.

What blockchain proves is limited — it proves the record was not altered afterwards, not that it was correct when written. A hash chain can show that nobody changed the data later; it cannot show that the data was right at the moment of capture. If someone enters the wrong fact at the entry layer, the blockchain immortalises the error. The real obstacle in cricket is therefore not technology but the discipline of data entry — which incident, at what time, coded under which definition.

That discipline matters even more for betting integrity. The ICC's anti-corruption unit looks for suspicious betting patterns, unexpected momentum shifts and pre-match contact. A tamper-resistant log helps in that work, but only when the rules of capture are written down and universal in advance.

The same rule governs my own work. If a source yields no core facts — no headline, no date, no account of the event — the honest answer is one: there is not enough information. An analytical frame can be filled in, but a filled frame is not a filled verdict. What happens when an umpire guesses happens precisely when an analyst builds a verdict out of zero information — except that no review exists for the analyst.

The crowd wants a verdict; the protocol wants reliability — and the collision between the two is where anti-VAR and anti-DRS anger is born. VAR did not reduce controversy; it moved controversy off the pitch and into the review room and the grey zones of the rulebook. Coding 500 decisions in 2026, I saw that most of the argument was not about the decision itself but about why this decision was not reviewed, why that angle was not shown, why the impact fell inside the band. The more visible the evidence, the more refined the grievance.

There is another uncomfortable truth. The wider the tolerance band of proof, the less the umpire's role shrinks — it grows. Every decision outside the band defaults to the field call. Technology has not removed the umpire; it has seated him at the centre of an invisible tolerance, where his call can no longer be verified and yet remains final.

One question remains inside this architecture: who gets the benefit of doubt? In DRS it is the batsman; under umpire's call it is the fielding side; after the removal of the soft signal, on low catches it is the batsman again. That balance has never been openly debated — and yet it decides which side wins which match.

The next reform is not another high-speed camera; it is evidence governance. Publishing the dimensions of the tolerance band, documenting the decision tree of every review, and holding that record in a verifiable ledger will not reduce argument — but it will give the argument an address. Would those six runs at Lord's in 2026 have been written into an immutable record? They would have — if the rules of writing had been settled beforehand. The question, then, is not one of technology. It is one of will.