From Earpiece to Raised Arm: A Forensics of How Reviews Are Actually Built in the Regular Season

That day I sat listening for one word on the stump mic — "wait." Then thirty seconds. In thirty seconds everything happens that cricket normally never allows: the ball-tracking loads, a path is chosen, an umpire decides inside his own ear whom to trust, and the arm either goes up or it does not. I counted those thirty seconds, not the out or not-out.

At twenty-seven, having come to cricket from a football refereeing forensics, I have learned exactly one thing: the decision is history, the process that made the decision is the future. So this piece is not a verdict on whether a particular review was right or wrong. It is a ledger — how DRS is working in international cricket this regular season, where it is breaking, and where we are asking the wrong questions.

I flipped through my notebook while writing this. In football I once coded 118 decisions in a single match; in cricket the number is more embarrassing — four to six reviews an innings, twelve to fourteen in a Test. Fewer numbers do not mean less information; each review carries more weight, because it can bend the course of a session.

---

Context: The law we know, and the process we never see

DRS was first trialled in 2026, in a Test between India and Sri Lanka; the ICC rolled it out fully from 2026. The whole system stands on three pillars: Hawk-Eye ball-tracking, UltraEdge for edge sound, and the third umpire's ruling. But what is not written in the playing conditions is "umpire's call." If the ball is predicted to hit half the stump or less, the on-field umpire's original decision stands.

That one line is the most discussed, least understood rule in cricket. It is not a technology failure — it is a deliberate error tolerance. In football terms, it is nothing like VAR's offside line; it is closer to semi-automated offside, where the system makes a call but leaves room for human judgement at the margins.

In this regular season, one pattern is clear in every review I have watched: teams that read technology as law lose reviews, and teams that read it as probability win them. If the chance of out is below one percent, the half-ball rule can flip it. Knowing this stops you burning a review for nothing.

---

Core: Inside one review there are four separate decisions

To me a review is not a package but a string of four decisions. First, the fielding side's decision to review — that is tactics. Second, the third umpire's viewing order — which feed he checks first, snicko or ball-tracking. Third, the reliability of the ball-tracking prediction. Fourth, the decision to uphold or overturn.

During my English-language commentary debut in the Bangladesh women's ODI series against India, I watched this sequence most closely. When a review goes up, an odd silence falls over the ground — the crowd knows something is being judged, but not what. And it is inside that silence that the umpire builds his call.

Layer one — where the ball pitched. Least controversial. Technology rarely errs on pitch point.

Layer two — where the ball impacted. The first crack. The height of contact on the pad, inside or outside the bat, a small spike on the snicko graph — all judgeable, but never as clean as a boundary line.

Layer three — the wicket prediction. The most disputed. How much the ball turns after pitching is a model's forecast, not a fact. That model is built on match data — and it depends on series, venue and ball condition. The same delivery can produce two different forecasts at two venues.

Layer four — umpire's call. And here is my real objection, though not to the technology, but to our expectations.

An example. Suppose the ball pitches on line, the batter is well forward, impact in line. Whether the forecast says the ball hits more than half the stump is almost a coin toss. If the on-field umpire says not out, a review will not overturn it. If he says out, a review may well uphold it. Same ball, same technology, two outcomes — purely because of the umpire's first call. Many call this a technology failure; I call it the design of human-technology coexistence.

What I see in the regular season's numbers: teams that have learned to read this design take only the reviews with high binary probability. Teams that review on emotion burn them. In a regular week, teams lose roughly three of ten reviews — a large share on that half-ball boundary.

---

Two vantage points: the third umpire's room and the umpire at the stumps

I always try to compare at least two vantage points. One, the on-field umpire at the stumps. Two, the third umpire's room, a human sitting before a screen.

The on-field umpire decides in a fraction of a second, from one angle of his body, keeping time with the ball. He has no snicko, no ball-tracking, only eyes and experience. The third umpire has every technology, but he has time — thirty seconds, sometimes a minute.

What I cannot see is this: when the third umpire watches a snicko graph he looks for a spike; when the on-field umpire gives it out he looks for a pattern. There is no bridge between these two ways of seeing. So the same ball can be read two ways in two places.

I once arranged this season's reviews in a spreadsheet — which held, which failed, which died on umpire's call. Then I realised my columns were wrong. I was counting decisions, not process. I made that mistake once in football, in 2026 in Sydney, coding 118 decisions in a match. In cricket I will not make it again.

---

Intervals: how a session break changes the review ledger

I am abnormally interested in intervals; it is an old affliction. Drinks, lunch, stumps — in cricket these are not just rest, they are a reset button. And that reset matters in review decisions.

One pattern recurs this regular season: if a side burns a review in the last fifteen minutes of a session, its confidence in the first half-hour of the next session changes. Because it has one or zero reviews left. The captain avoids reviewing for fear of going out — and exactly then a genuine out slips away.

Here I have learned caution: blaming the interval is easy, but it is a hypothesis, not proof. A pattern seen in one session becomes a rule for me only if it survives three sessions. So I keep a session-by-session ledger — when each review was taken, just before or after a break. So far reviews taken just after a break succeed slightly more often, but the sample is not enough. I am growing the sample.

One more thing must be added: the player against whom the decision goes. A lost review is not just a number — it is a batter's session of work, sometimes the fate of a whole innings. After every count, one sentence has to be written on whose shoulders that number fell.

---

Contrarian: the fight between emotion and law

Now the part where I am naturally most uncomfortable. The crowd wants the review as justice. The player wants it as tactics. The umpire wants it as process. These three expectations are not the same, and that is the biggest problem.

When a review dies on umpire's call, the crowd feels cheated. But the rule was written precisely for this situation — when technology is not certain, technology should not decide. The question is not "who was robbed," but "who decides."

My strongest objection is here: we have turned the review into a justice machine, when it is a doubt-removal machine. Two different jobs. Justice wants final truth; doubt-removal wants reasonable confidence. DRS can give the second, never the first — because no model knows the ball's future, only draws its probability.

And here comes language. I describe matches with sound, not pictures. What the umpire says on the stump mic matters to me no less than where the ball pitched. The silence before the word "stay" is the real evidence. I triangulate every audio clip with the scorecard and positional notes, because sound alone never tells the truth.

---

Takeaway: not the machine, the question must change

I think DRS controversy will not shrink in coming seasons but grow — because ball-tracking will get more precise, and precision will raise our expectations. But the moment technology becomes perfect, the umpire's call question will get louder, not weaker.

What is needed is not a change of technology but more transparency. Publishing each review's confidence level, showing ball-tracking's confidence, and putting session-based review data in the open. Then teams can review as tactics, and the crowd will be robbed less.

I count thirty seconds. At twenty-seven one rule has settled in me: I trust a pattern only when it survives three different angles. This review question has not yet survived all three. So I wait — next session, next interval, next silence.

— Root: Referee

(All facts here rest on ICC published rules, the third-umpire protocol, and publicly available session-based match data. Personal observation is the writer's own.)