The Truth Between Two Frames: DRS, Umpire's Call and Cricket's Decision Ledger

November 6, 2026, Arun Jaitley Stadium, Delhi. I was three rows up in the West Stand, my notebook open, my right hand pressed flat on the page — old habit, otherwise the wind lifts the sheets. First innings of Sri Lanka against Bangladesh, the middle overs. Angelo Mathews was walking out to bat when his helmet strap snapped. He left the field, replaced the strap, came back. The two minutes were gone. Shakib Al Hasan appealed, the TV umpire looked at the clock, and for the first time in the history of international cricket a batter was timed out.

That evening in the stands, and in the press the next morning, almost the entire argument was conducted in moral vocabulary — law against spirit, courtesy against hunger for victory. Some said this was not cricket; some said a rule is a rule. My ledger had no column for what happened, because there was no frame to enter. The truth was not in a frame. It was in a clock, and the number on that clock was not visible to anyone standing on the field at every given moment.

Rewind the tape; the truth usually hides between two frames. In the timed-out case it hid outside both frames — inside a time limit, which produces no image. Cricket's most disputed decisions are born precisely there: where there is no camera, but there is a clock.

Modern cricket's decision architecture stands on three tiers. On the first tier are the two on-field umpires, armed with one pair of eyes and a fraction of a second. On the second tier is the TV umpire, with a bank of replays, UltraEdge, ball-tracking and feeds from dozens of cameras. On the third tier is the match referee, with over rates, the code of conduct and behavioural reports in his book. Three tiers give verdicts in three different languages, and that is exactly where the widest gaps open.

Ball-tracking first entered an international Test in July 2026, at the Sinhalese Sports Club ground in Colombo, during the India–Sri Lanka series. There was fear at the time that technology would swallow the umpire's authority. It has not. Rather the opposite: technology has drawn a tolerance band around the umpire's call. That band is called umpire's call — the technology cannot say conclusively that the on-field decision was wrong, so the decision stands.

Before 2026 there was a convention called the soft signal for low catches. The on-field umpire would give a soft 'out', and the TV umpire would watch the replays and either uphold or overturn it. The convention was dropped in 2026. The reason was simple: a soft verdict tugs at the TV umpire's mind, and the job of technology quietly shifts from making a call to assembling support for one.

Over rates belong to the same family of legal instruments. A side short of overs can lose five per cent of match fees per over, the captain is penalised separately, and repeated offences in Tests open the path to suspension. Since 2026 the punishment in ODIs and T20Is has moved onto the field itself: if the final over is not started on time, an extra fielder must come inside the circle — something that can directly change the result of a match.

I tried to count this whole apparatus by hand long ago. After the 2026 domestic football title race I built a forty-column decision ledger, recording beside every disputed call the referee's body orientation, his sightline, and whether his own movement created the blind spot. The federation returned it unread, labelling it 'not an official document'. Those column definitions are the ones I later carried across to cricket, because the technique changes, the questions do not.

Ball-tracking is a projection, not a recording

Start with the obvious point. A ball-tracking camera does not show the ball's future; it shows a projection. The ball's position is measured at a handful of points after release, and physics equations then estimate the rest of the journey — friction, bounce, spin, wind speed, even ground moisture. If the system cannot capture the exact instant the ball pitches, every subsequent estimate carries the same uncertainty.

Then add frame rates. If a camera runs at fifty frames per second, the gap between two frames is twenty milliseconds. A ball travelling at 140 kilometres per hour covers nearly seventy-eight centimetres in that gap. The moment of bounce, the point of contact with the pad, the touch of the bat's edge — the precise location of any of these can fall inside that seventy-eight centimetres. The umpire's call band is largely an acknowledgement of that uncertainty. When the ball is hitting the stumps by less than half its width, the technology says, in effect: I am not certain.

There is another layer: when the point of impact is roughly three metres or more from the stumps, ball-tracking is treated as less reliable, and the threshold for overturning the on-field decision tightens further. That rule was not written for drama. It was written because the ball has travelled further, swung more, and the probability of a projection error rises.

Sound-based edge detection is a different instrument running at a different sampling rate. In many Test matches we see what looks like bat on ball with no spike on screen, and supporters conclude the technology was asleep. In truth a microphone can pick up bat brushing pad, strap movement, even the umpire's footsteps. No spike does not mean no edge; it means the specific moment of sound could not be isolated. The eye and the ear are two separate technologies with two separate limits, but the broadcast presents them as a single truth.

A review is a scarce resource, and scarce resources carry a price

My undergraduate degree in economics earns its keep here. A team receives two reviews per innings, refreshed after eighty overs in Tests. A review is therefore a limited asset. Every use carries an opportunity cost — if the review fails, the asset is gone forever, and the team stands naked at the most important moment of the match.

The question is not moral but quantitative. What is the expected return on a review? At which phase of a match is it highest? Is the captain trusting a bowler's emotion or information from the dressing room? The quality of that decision is tied directly to a board's budget, analyst headcount and video library. Wealthier sides now keep a dedicated review coach who watches the television feed and signals within seconds; smaller boards wait for the same signal. The question is not why a review was squandered. The question is why some people know and others do not.

This is where cricket's DRS diverges from football's VAR in an interesting way. In football the referee goes and looks himself; in cricket he does not. Information asymmetry in cricket is therefore wider, because information must travel to the TV umpire and then return to the field. How much of that path is visible to any given team depends on how much the broadcaster chooses to show.

What the crowd sees is a cartoon of the decision, not the decision

In every major series I note that different broadcasters render ball-tracking differently. Some show a three-dimensional path, some a flat line. Some show the impact point frame by frame, some simply deliver the verdict. Camera count, camera angle, the wealth of the host broadcaster — all of this feeds into a decision that the audience assumes is neutral.

I readily acknowledge the limits of my own vantage point. I count frames from a monitor in Delhi; I have no seat in the ICC's technical committee, and raw Hawk-Eye data never reaches my desk. What I can do is keep the pattern ledger — and keep a clear wall between pattern and proof.

When I was the only woman sitting in the press box at Ambedkar Stadium, I learned one thing: the decision nobody can see is the decision most loudly debated. Silence in a press box is not consent, but silence is often the translation of helplessness. That is precisely where cricket's decision architecture fails — thirty thousand spectators accept a verdict after watching a three-dimensional graphic, while the measurement of whether the ball was hitting the stumps is never published.

The ledger of the spirit

On July 2, 2026, at Lord's, Alex Carey stumped Jonny Bairstow while Bairstow believed the over had ended. Marylebone Cricket Club later stated that the decision was lawful but against the spirit of cricket. That sentence deserves attention — law and spirit are two separate ledgers, and neither one's measurement criteria are public.

This is where I dissent most loudly. The unwritten ledger called the spirit of cricket has no columns, no definitions, no recurrence rate. It appears only when a lawful outcome offends the majority. The Mathews timed-out, the Bairstow stumping, the low-catch debates — in every case the same demand rose: this should not have happened. But nobody has yet been able to offer a pre-agreed measurement to support it.

The second demand now heard loudly is to abolish umpire's call or automate everything. It is humanly understandable but technically counterproductive. Abolishing umpire's call does not remove uncertainty; it makes uncertainty invisible. Forcing a verdict where the projection carries error does not reduce mistakes — it only makes them sound more authoritative.

Third, it is worth avoiding the trap of reading bias into pattern. One can observe how decisions shift against left-handed batters, or how quickly a third umpire responds to a major board's appeal. But the null hypothesis has to be written first: perhaps there is no bias, only ordinary base rates. What my ledger keeps returning to is not bias but the shape of the instrument — where cameras are placed, who controls the feed, and who has an analyst on staff.

The next reform is about disclosure, not about the law

Cricket's next great reform does not lie in abolishing umpire's call. It lies in the open: displaying ball-tracking confidence in real time, making the clock visible on the field, and mandating a minimum camera count for every broadcaster. If the information stays private, decisions must be accepted on faith — and faith is a poor instrument for settling the fate of a World Cup in front of thirty thousand people. So the question is blunt: if a decision cannot be shown to the crowd in real time, should it be a decision at all?