From UltraEdge to Umpire's Call: Auditing Cricket's DRS Decision Tree

Frame 41 in My Notebook

I began with one bedroom, one rulebook, and a suspicion the table was lying. In 2026, in a ten-by-twelve-foot room in Mymensingh, I logged 214 officiating decisions from the 2026-17 UEFA Champions League and the 2026 Europa League final — Ajax 0-2 Manchester United, referee Damir Skomina, 34 fouls, five yellow cards. The purpose was single: measure the gap between decision and evidence. At 50 frames per second, referees' reaction time landed at 0.28 seconds, and that number became the spine of my first piece.

Apply the same method to cricket and the gap is no longer temporal. It is regulatory. A review ends in three words: out, not out, umpire's call. The first two are measurements. The third is a management decision wearing the clothes of a measurement.

My ledger has five columns per review: over, delivery type and speed, spike frame, the three ball-tracking projections (pitching, impact, wickets), and the final announcement. Where I cannot fill a cell, I leave it empty. The ICC publishes no consolidated review-level database, so this table is mine, not theirs.

The rule arrived before the machine

Ball tracking entered cricket broadcasts in 2026, on Channel 4's coverage of England matches. Hot Spot arrived in 2026, UltraEdge around 2026. The Player Review System came later still — July 23, 2026, the first Test between India and Sri Lanka at the Sinhalese Sports Club in Colombo. India initially declined; the BCCI questioned the ball-tracking element for years and fully accepted it around 2026.

The timeline matters because it shows DRS is not a child of technology. It is a child of negotiation. Who may review, how often, and when the review is returned — all of it sits in the appendices of the playing conditions. In Tests, two unsuccessful reviews per innings, topped up after 80 overs. In ODIs and T20Is, two, with no top-up.

In March 2026, at the Bengaluru Test, Steven Smith looked toward the dressing room after being given out LBW, seemingly seeking guidance on whether to review. The episode drew a boundary line: the technology was on the field, but whether decision-data could travel off it was written nowhere. The ICC clarified that external assistance is not permitted before a review. The rule was written. The suspicion stayed.

Through 2026-21, travel restrictions meant several series used local umpires instead of neutral ones — a reminder that DRS is not only about ball and stumps, but about who is judging and in what environment. My confidence on this reading is moderate; the condition is re-checking the numbers after six rounds.

Three questions, one shadow band

An LBW review's decision tree is simple: where did it pitch, where was the impact, would it have hit the stumps — Root: Referee. The first two rarely divide opinion. The third produces the noise, because "would have" is a forecast.

Here lies the geometry of umpire's call. A men's cricket ball is roughly 7.2 cm across. The rule states that if the centre of the ball is within half a ball's width of the outside edge of the stump, the projection is treated as uncertain — a band of about 3.6 cm. Inside that band, the on-field call stands and the review is returned.

Think about that: roughly three and a half centimetres. A bat is about ten and a half centimetres wide. We have authorised a half-ball-wide shadow zone inside the law, then sold it outside as accuracy.

0.23 seconds: the umpire's internal clock

Run the arithmetic. 140 km/h is 38.9 metres per second. Assume 20 metres from release to batter; the ball arrives in about 0.52 seconds. With the same kinesiology background that gave me a 0.28-second referee reaction time in football, a cricket umpire's decision window works out to roughly 0.23 seconds.

That deficit is why technology is needed. It is also why umpire's call exists. Both needs grow from the same root, and that is the system's central discomfort.

Technology does not measure; it forecasts. Ball tracking extrapolates a future path from observed speed and deviation. Damp patches, grass cover, seam orientation — all variables in a model that is not public. Since 2026, Hawk-Eye Innovations has been owned by Sony. The playing conditions are public; the spreadsheet that decides is not. Three kinds of absence deserve separate names: data never collected, data held but unpublished, and data actively withheld. The first is the sport's inevitability, the second a legal boundary, the third an ethical question.

A review is a currency

Returning reviews on umpire's call has created a small economy inside the match. Two reviews mean limited capital. An umpire's call review is not counted as failed, because the on-field verdict is the valid one. The consequence is strange: the blame for an on-field error and the return of a review end up merged in the same event.

In fan culture, MS Dhoni's so-called "review system" rests on exactly this capital management. Tail-enders' reviews are often banked; a senior batter spending one on personal conviction is a cost to the team. Failed reviews never appear in the scorebook, and rarely in the explanation either — while in the financial ledgers I have followed for years, every clause is answerable.

One ball, three rulebooks

Now to comparative governance. Same pitch, same ball, same projection — but bilateral series, ICC events and franchise leagues do not share identical review appendices. Who may review, and what may be challenged, differs by competition.

For a fan watching from a room in Mymensingh, the result is that cricket has not one rulebook but a shelf of them. Ask why a review was returned in one tournament and not another, and no single, current, public document answers. Multiple versions mean multiple standards; multiple standards mean multiple escape routes from accountability.

When the stadiums emptied, the numbers finally spoke without the crowd

In 2026, when sport stopped, I went back into data, using the Russia ledger as my template. The Russia ledger taught me that memory is a spreadsheet with redactions: 64 matches, 455 VAR checks, 20 on-field reviews, 17 overturned decisions, a record 29 penalties, every call checked against the 2026-19 IFAB Laws.

Across 306 post-lockdown matches in the Bundesliga, Premier League and La Liga, I coded 1,842 fouls, 73 penalties and 1,106 yellow cards. Average fouls fell from 26.3 to 22.8 per match; the home-win rate fell from 43.2 percent to 38.1 percent.

The data is football's, but the inference transfers. An on-field umpire's call is environment-dependent. Home sides, crowd pressure, series consequences — all of it changes the air in which decisions form. Umpire's call is a stored copy of that environment, carried inside the machine.

The law's cleverest move is this: the projection shadow zone is not merely the machine's limit, it is a legal instrument that preserves an environment-shaped judgement inside a machine-verified process. The technology no longer makes the call so much as guards it.

Contrarian note: not proof, but the styling of proof

Convention holds that DRS reduced controversy. My ledger says it relocated the argument. The question used to be the umpire's eye; now it is the machine's margin. And margin debates are less useful than eye debates, because behind the margin stands a company, whereas behind the call stood a person.

Second discomfort: we audit the ball, not the process. Who sits in the review discussion, which frame they see, which frame nobody asks for — none of it is recorded. Where data is absent, numbers do not stand, estimates do.

Third, and most awkward: in the pursuit of accuracy, the on-field umpire has become a provisional authority. The announcement is now conditional, not final. The authority question that followed football's VAR has not spared cricket; it is simply heard less.

A batter's career now rests on a three-and-a-half-centimetre forecast. Kinesiology taught me that decisions born under time pressure are probabilistic, not certain. A machine can shrink that probability; it cannot zero it — yet the law keeps speaking the language of zero.

The table nobody has published

First task: a central, public, review-level dataset — every review, every frame, every announcement, with source and date attached. Without records there is no correction, and without correction there is no trust.

Second: harmonise the versions. Bilateral series, ICC events and franchise leagues should share review appendices. The same ball deserves the same reading, otherwise we are distributing identity cards, not rules.

Third, a proposal for what comes next: a distributed, tamper-evident public ledger recording each review's frames, model version and verdict immutably. The technology exists. The obstacle is not technical but proprietary. Blockchain is no miracle here — it is an accounting principle whose value depends entirely on who runs the nodes.

Line decisions — no-balls, height, possibly foot faults — will likely move to machines, while the LBW verdict stays human. Which changes the question: do we trust the machine, or the machine's owner?

I do not count the points until I have audited the cells beneath them. How many reviews stalled on umpire's call, which appendix bites hardest in which format, at which frame the spike arrived late — those cells are still empty, and no decision written above an empty cell is a decision to me. It is only an announcement.