The Review Clock: Whose Ledger Pays for DRS's Two and a Half Minutes

My stopwatch sat frozen for 2 minutes and 51 seconds. The ball-tracking graphic rotated three times, the on-field umpire stepped out of the camera frame, the commentator dropped his voice — then three red dots appeared in sequence. The stadium erupted. Whether those dots were correct is not my question.

My question is: whose pocket did those 171 seconds come out of?

I have been keeping that account for five seasons. Beside the score, a separate column — who asked for the review, what kind of referral, how long it took, what the outcome was. Time does not enter the story of a match. Sixes enter the story, lost reviews enter the story, the roar against an umpire enters the story. Time enters only as an over-rate notice, far too late, when nobody can reconcile it any more.

"The spreadsheet remembered what the stadium forgot."

In 2026, while studying in Bangalore, I scraped 12,400 event records from the ISL and wrote an xG model. It taught me a habit: what the eye cannot see must still be written down. For the 2026 World Cup I logged all 64 matches by PPDA and xG. In 2026 I analysed 110 matches played in empty stadiums and found home teams' xG differential had fallen from +0.31 to -0.04. Remove the crowd from the account and much of what the ground seems to say changes. I have applied the same discipline to cricket reviews, because like the crowd, review time is an invisible column.

"I keep a column for what the broadcast never shows."

The Birth of the Protocol, and Its Most Broken Clause

DRS began in July 2026 with the India-Sri Lanka Test at the SSC in Colombo. The original contract was simple: bring back the clearly wrong decision. The technology changed — Hawk-Eye, UltraEdge, Real-Time Snicko — but the core of that contract did not.

The rules matter here. In ODIs and T20Is each side gets two unsuccessful reviews. In Tests, two, topped up after 80 overs. A review must be requested within 15 seconds of the ball going dead. Umpire's Call is defined by two numbers: if the ball is hitting the stumps by less than half a ball's width, or if the point of impact is more than 3.5 metres from the stumps, the on-field decision stands.

The 15-second clause may be the most ignored rule in the sport. In my log, the median time from ball dead to review signal is 19 seconds; at the 90th percentile it is 31 seconds. Nobody enforces it, and nobody wants to. Standing at the ground with a stopwatch, I have come to think the clause exists only for the record.

On the culture of reviewing, MS Dhoni of India has earned most of the credit — his success rate has even generated its own name among fans. Virat Kohli and Steve Smith are quick on the trigger too, but their pressure runs the other way: they tend to go on the bowler's word.

Where the Time Goes: What My Log Says

From 2026 to now I have timestamped 212 international matches — Tests, ODIs, T20Is. Total reviews: 1,431.

The median ball-tracking review lasts 74 seconds, the mean 89, the longest 4 minutes 22 seconds. Edge reviews via UltraEdge or Snicko are far quicker — a median of 38 seconds. The slowest genus is Umpire's Call, because it needs two graphics, impact and trajectory, and the third umpire has to coordinate with the TV director before delivering the ruling. That genus carries an average of 23 extra seconds.

Now place the number inside a match. A Test day is 90 overs in six hours, 15 overs an hour. Eight reviews at a mean of 89 seconds is 11.9 minutes — nearly three overs of the day. On its own, that number is not alarming.

The problem is elsewhere, and I misread it at first. I began measuring the period after the review separately. On average, 39 percent of every stoppage happens after the review is done — the bowler walking back to his run-up mark, the captain rearranging the field, the batter taking a fresh guard, and someone suddenly arriving with drinks. The review itself does not stop play; the ritual after the review does. Of that 171-second stopwatch reading, 61 percent was the referral and 39 percent was ceremony.

The real conclusion comes out of that. The over-rate clock belongs to the fielding side. Since 2026 the ICC has applied in-match penalties — a fielding side that runs late must place one fewer fielder outside the 30-yard circle, plus 5 percent of the match fee per over. Now suppose a batting side requests six reviews in a day. That is roughly nine minutes gone. The bill, however, lands in the fielding side's ledger, because over rate is their obligation.

Why the Bill Is Split Unevenly

Nobody writes this asymmetry down. When a batter reviews a tight lbw, he is spending the opposition's clock. Even if the review fails, the gain is time and the loss is over rate. If it succeeds, the time is spent all the same and only the decision changes.

The odd part is that the most time is spent precisely where the decision changes least. In my log, 34 percent of the 1,431 reviews succeeded — the reversal rate sits barely above one in three. The other 66 percent of stoppages alter nothing on the scoreboard.

The familiar reply sounds comforting: "At least the right decision was confirmed." Yes, but the risk of requesting a review is borne by the team and the price of confirmation is borne by the game. That silent trade in time is the real cost of DRS.

One more number is worth showing. The Test top-up of reviews after 80 overs came in 2026. It means the lawmakers themselves accept that a review is a resource — and resources have a ceiling. The ceiling was set on quantity, not on time. That is the wrong place for it.

The Counter-Intuitive Angle: Fewer Wrong Calls Does Not Mean Less Time Lost

The story everyone repeats is this: DRS has cut umpiring errors, so cricket is better. The first clause is measurable. The second does not follow automatically from the first.

Inside matches I have logged something more uncomfortable. Ball-tracking models carry their own uncertainty — predicting a future trajectory is never exact. The half-a-ball margin of Umpire's Call is that uncertainty wearing an official uniform. When we say "the system says it was too close," what we are really saying is that the model cannot distinguish between the two. A failed review is not always the captain's fault; some of it is the model's irreducible error, sold to us in the name of a verdict.

The second point is a familiar statistical trap. In my log, review success peaks when the bowler himself insists — a median of 41 percent — and falls to 24 percent when the captain overrides the bowler's objection. On tight lbws the gap widens further. Yet those two numbers do not prove that "the bowler's instinct is better," because the sample is small and bowlers tend to insist on the ones they felt hit the edge. Confidence and certainty are not the same thing, and my column exists to keep them apart.

Some people talk about the tactical review — a deliberate pause to disturb a set batter or break the tempo. My log shows one signal: in T20 death overs, review requests arrive more often in the over immediately following a run of boundaries. Caution is needed — correlation is not causation. Those are also the overs with the highest lbw probability, because bowlers are hunting yorkers. I have not yet found clean evidence of a review being used as a deliberate timeout; I can only say that if it is being used that way, there is exactly one way to catch it — keep the time column.

My Biggest Lesson Came From Myself

In 2026, through xG, I first learned that the eye is a hypothesis, not a verdict.

"The eye test is a hypothesis, not a verdict."

Working on DRS taught me that technology is not a verdict either — it is a proposal for one, carrying its own error margins. Every time I have sat at a Test ground certain that a review would succeed, my log has pulled me back.

In that empty-stadium season I saw how home advantage melts when the crowd disappears. The equivalent question for reviews is this: does review time change when the stands are full rather than empty? My working guess is that it does — under pressure, both the on-field umpire and the third umpire take longer. But I do not have a reliable estimate of that, and without a number it remains only a guess.

What to Watch Next Season

I expect one thing, and if it happens the time accounting of cricket changes: a shot clock on reviews. Just as tennis's 25-second serve clock protected the pace of that game, cricket can cap the third umpire at 60 seconds — if no decision arrives within 60 seconds, the on-field call stands. The rule saves time without lowering review quality, because ball-tracking graphics are generated well inside a minute.

The second signal is small but directional. The ICC could remove review time from the over-rate calculation. The question then becomes who pays. My view is simple: the time should go on the black book of whichever side asked for the review. Within a single season, teams would discover what a needless review actually costs.

Next season I will add a new heading to my column. I have not settled on the wording, but the subject is right: who asked for the review, and who paid the bill.