No Evidence, No Ruling: How Cricket's Review System Decides on 'Insufficient Evidence'

The file was empty. Title: N/A. Source: N/A. Information points: zero. The column marked 'entities involved' was blank as well.

That is not an unfamiliar position for a rules desk. It happens on the field every other day. A frame arrives on the third umpire's monitor — grainy, half-cut, ball and bat sharing a single pixel. Fifty thousand people in the ground start shouting. Someone in the commentary box declares it obviously out. And the on-field decision stands, because the playing conditions say a call may only be overturned on conclusive evidence.

The question was never what the replay showed. The question is who writes the ruling when the replay settles nothing — and who wrote that clause in the first place.

When I covered my first Test in 2026, the third umpire had two cameras and nobody counted them. Today there are thirty cameras, UltraEdge and ball-tracking, and the arguments are exactly as loud. Technology produces evidence; it does not assign weight to it. The weight is fixed by a clause drafted in a committee room.

So today's case file is not the empty one. It is the doctrine that entire review system rests on: insufficient evidence.

Context: what the review system is actually built on

DRS is usually described as technology. It is really a procedural code with three pillars. The first is the player challenge — two per innings in Tests, ODIs and T20Is, retained if the review ends in an umpire's call. The second is the umpire referral: run outs and obstruction go upstairs on the umpire's own initiative, and players cannot challenge them. The third is the third umpire, who holds all the evidence and one standard of proof: conclusive.

Two places in that structure admit, on the record, that measurement runs out. The first is umpire's call: if more than half the ball is not inside the stump line, ball-tracking returns 'umpire's call' and the field decision survives. The second was the soft signal, in force until 2026 — the on-field umpire's provisional opinion, offered before the review, which coloured how the third umpire read a tight catch.

In 2026 the ICC's Men's Cricket Committee recommended abolishing the soft signal, and it went. It is the cleanest reform of the decade, and in this desk's ledger it is filed under 'the official got it right' — no qualifiers. The reasoning was blunt: when the evidence is already thin, adding your own opinion to it is not verification, it is contamination. The ICC conceded that the on-field signal was not evidence but prior bias.

— Root: Laws Desk experience + Referee

Core: the 2026 overthrow and the lie that outsold the truth

Now the case that has generated more words than any other in the last six years, at least half of them wrong.

14 July 2026, Lord's. England needed nine off three. Ben Stokes hit towards deep midwicket, two runs were completed, Martin Guptill's throw deflected off Stokes's bat and ran to the boundary. The umpires awarded six — two run, four for the boundary. The equation changed, the match tied, the Super Over followed, England won the World Cup.

By the next morning a single sentence was circulating on every channel: it should have been five, not six. Panellists repeated it. Twitter threads repeated it.

My ledger records that claim as false. The MCC clarified that Law 19.8 — overthrow or wilful act of the fielder — was applied correctly: two runs had been completed and the boundary four was added to them. The third run, in progress, would only have counted had the batsmen crossed at the moment of the throw. They had not. So six, not seven. And plain arithmetic says six as well.

None of which means the 2026 final was free of a rules crisis. It just was not the overthrow. The crisis was that the result of a World Cup was decided by a provision no one could review — boundary countback. The Super Over tied, and the trophy went to the side with more boundaries. That was the rule, not the sport. It could not be sent upstairs, because the DRS menu has no line item called 'the provision that decides the result'.

In 2026 the ICC removed boundary countback; a tied Super Over now means another Super Over. That reform came from no umpire and no camera angle. It came from a committee room. Which is the real lesson of this beat: cricket's biggest errors are not made during reviews, they are made when the list of reviewable things is drafted.

And that is where the live gap sits. DRS covers dismissals only — caught, bowled, LBW, stumped, hit wicket. The count of runs, the application of the overthrow law, the award of a boundary, a dead-ball call: none of these can be challenged by a player. A side whose wicket was wrongly reprieved gets a review. A side whose match turned on two wrongly awarded runs gets nothing.

In a referee's reading the asymmetry is obvious: protection is granted where the controversy is most televisual, not where the damage is greatest. An lbw frame makes drama. A run count does not.

Third point, almost never discussed on air: the third umpire does not see with his own eyes. He sees the feed the host broadcaster sends, and the host broadcaster is a commercial party to the same tournament. Which angle runs first, which frame gets magnified, how fast the review is paced — those are production decisions, and no cricket law accounts for them. The ICC does not publish full third-umpire audio. The court we are auditing never releases its transcript, only the cut chosen for broadcast.

— Root: Referee

How the clauses divide: who wrote them, who gains

Now the question that belongs at the end of every rules argument. Who is umpire's call for?

The superficial answer is: to protect the umpire. The ledger shows something else. Umpire's call bites hardest on retained reviews. A review that ends in umpire's call is not lost, so the uncertainty is not charged to the team. Conversely, the side with reviews in hand can gamble in the final overs.

Put plainly: umpire's call is not a shield of neutrality, it is a strategic resource — and it is distributed as unevenly as a batting order. A team with three reviews left will challenge fearlessly on the fourth morning. A team with one will hoard it until the last hour. Same wrong decision, two different prices.

The soft signal carried a similar one-way benefit. Against a team whose umpire had said not out, the fielding side needed conclusive evidence of a catch. For the side that had been given out, doubt alone was enough. One resolution, two burdens. That asymmetry ended in 2026, which is why the change should be called a correction and not a softening.

A glance at the sanction tariff confirms the pattern. Slow-over-rate fines, code-of-conduct levels, and the rates that apply across the IPL, domestic competition and international cricket do not match. When the same lapse costs two different pockets different amounts, inconsistency of punishment damages the game more than the original offence. A rule's authority is built not from its text but from the uniformity of its application.

Contrarian: outrage is not evidence

Before we debate intent, let the record show: the most damaging claim in this whole affair was not made by an umpire. It was made by a panel — 'it should have been five'. It spread so widely that millions still hold it as fact, while the MCC's clarification, the text of Law 19.8 and simple arithmetic all return the same answer.

That is this desk's sharpest warning. The most dangerous thing in cricket is not a wrong decision but a consensus built around one, because consensus removes the need to check.

Then there is the easiest accusation available: scrap umpire's call, let the technology decide everything. It sounds revolutionary. Its outcome would be that a millimetre of ball on stump becomes a wicket, and the argument simply migrates from the decision to the measurement. The problem is not that umpire's call exists. The problem is that umpire's call is the one door being demanded shut, in a building where several doors were never built at all. Runs scored, overthrow applications, dead ball, fielding obstruction — no door exists for any of them.

And the least heard point: where the official got it right deserves the same scrutiny. Abolishing the soft signal in 2026, removing boundary countback in 2026, retaining the review on an umpire's call — all three belong in the record without applause and without a trend. This desk's commitment is symmetrical: the same coldness for the errors, the same space for the corrections.

— Root: Referee

Takeaway

The next question is not whether umpire's call survives. It is whether the review menu ever expands — whether a captain can one day say 'that was five runs, not six, and we are sending it up'.

Until then, one thing is worth remembering. Refusing to rule on an empty file is itself a ruling, and in this sport it remains the most honest one available.