Umpire's Call: Where the Decision Boundary Still Hangs in the DRS Era

LITON SHEIKH, MUMBAI — March 2026. One of eleven candidates in India's first VAR certification course. The instructor handed us tape of an ISL semi-final, Bengaluru FC versus Mohun Bagan, and gave us seventy hours to separate out every marginal call. I logged fourteen marginal offsides and three penalty-area incidents, and filed a twenty-eight-page decision audit that FIFA's officiating department later requested for its Asian development archive.

What I learned in those seventy hours was not about football. I learned that two kinds of truth live on either side of a line — what happened, and what we were able to see. Technology does not close the gap between them; it makes it sharper.

I have carried that question into cricket ever since. When ball tracking arrives on screen with three projected deliveries, one of them labelled 'Umpire's Call', the stadium screen lights up and thirty thousand people roar at once. Two truths in the same instant: a human being's judgement, and the outer edge of an algorithm's competence.

In my own ledger, twenty-three of the sixty-one international reviews I logged last season returned to Umpire's Call. Technology intervened and could not decide. This piece is about those twenty-three, and why the number rises in Asian cricket.

DRS was first used on 23 July 2026 at the SSC in Colombo, India versus Sri Lanka. It was optional then, series by series. India avoided it for years. From October 2026 the ICC made it mandatory across Tests and ODIs: two reviews per innings in Tests, topped up after eighty overs; one each in ODIs and T20Is. A successful review is retained; a failed one is burned.

The least discussed element is the decision threshold itself. It is arithmetic. If the ball is predicted to hit more than half the stump, the decision flips. If less than half, or the edge of the line, the on-field call stands. Many assume this 'half-ball' rule is a technological limitation. It is not. It is a deliberate legal choice: where the machine is uncertain, the human gets priority.

The ball is roughly 72 millimetres wide. Across that width the system must predict bounce, seam angle, drift and pitch unevenness — especially on subcontinental surfaces where the ball can lose pace in a divot. Add those uncertainties and the error envelope often exceeds half a ball. Yet the screen shows no error envelope. It shows a clean, coloured, confident path. My objection is not to the result. It is to the presentation.

Acoustic evidence carries its own trap. Broadcast feeds are compressed, transmitted and decompressed; frames and milliseconds disappear. On UltraEdge the gap between picture and sound can flip a verdict. At the 2026 Silent Whistle Study I ran on empty-stadium decision-making, I found umpires grew more confident in sound-based calls but more review-dependent overall — silence sharpens the ear and, with it, doubt. But I never declare certainty from audio alone. Channel count, compression, stump-mic placement: without those, 'I heard it' is a breach of VAR discipline. Certification is not the same as certainty — I learned that in 2026.

At Lord's on 14 July 2026, the overthrow off Ben Stokes' bat gave England six runs, a tie, and a boundary-count title. The umpires applied the law correctly. The law itself was written for an era in which nobody imagined a ball could hit a bat and produce eight runs. That is a jurisdiction failure, not a technology failure. A legally correct decision can still be a morally unjust outcome; conflating the two is how analysis gets lazy.

At Lord's in July 2026, Mitchell Starc's catch of Ben Duckett was ruled invalid because replays showed the ball touching the ground. The soft signal is gone; the third umpire now decides alone. I support that, with a caveat: ground contact is judged in two dimensions for a three-dimensional event. In that category the variance is higher than for spin bowling. The remedy is not more technology but a written rule that gives the batter the benefit of doubt.

Review usage is an economic decision. A captain spends a scarce resource in eight seconds, off a broadcast replay, inside a wall of noise. Successful Test sides do not win more reviews; they waste fewer. In Asia, front-foot LBW reviews are the most expensive kind, because the ball-tracking envelope does the most work on spin-friendly pitches. Review discipline is trainable. Boards simply do not train it.

Neutral umpiring arrived in 2026; COVID forced home officials back in England in 2026. The Elite Panel holds roughly a dozen people against a hundred-plus international matches a year. Fatigue erodes the speed of decisions before it erodes their accuracy, and slower decisions breed more reviews. The relationship is not linear, and I have not verified it directly.

Technology does not reduce controversy. It relocates it. Umpire's Call does two jobs at once — concealing the machine's uncertainty and granting the on-field umpire legal shelter. Those jobs should be separated. If the ball clips the stump, the graphic should read 'technological uncertainty', not 'Umpire's Call'. Change the label and you change how people understand the decision.

And the whole apparatus is also a broadcast product. More cameras, better tracking, higher frame rates — all of it follows the broadcaster's investment. The final of a major series and a routine bilateral do not carry the same technology. That inequality produces inconsistency of decision, and the ICC should regulate a minimum technological standard as strictly as it regulates the pitch.

Looking forward: change the language of Umpire's Call; formalise review education with decision trees and post-match audits; make minimum technology a mandatory condition. The real question is whether the sport wants to be fair or to be provably correct. Those are not the same thing. Fairness needs consistency; proof needs precision. The ICC has been sprinting at the second while talking about the first.