Sonic, 12 Matches and an Unwritten Precedence Rule: What Sound Can Prove, and What It Cannot

The ball brushed the inside edge, or it brushed the pad. When that question reaches the third umpire, the camera frame usually has no answer. Eleven years of watching matches has taught me the pattern: wherever visual evidence thins out, the decision drifts toward human judgment, and the controversy is born precisely in that gap. In the India–New Zealand bilateral series, sound is being pushed into that gap — an audio-analysis system called Sonic, deployed in international cricket for the first time.

The surprising part is not the technology. It is the paperwork. The ICC has approved Sonic's first use for this series alone, and has published no accuracy rate, no error rate, no false-positive count. "Testing completed" is, for now, the strongest evidence on offer.

What the machine does, and where it sits

Sonic analyses the sound generated at bat-ball impact to detect contact. The rule of use is narrow: the third umpire can consider the information during a review, and the technology is not an alternative to the umpire's decision. It does not decide. It supplies the raw material of a decision.

The system was developed in New Zealand and has been integrated into the DRS stack supplied by Virtual Eye, which has held ICC recognition for ball-tracking since 2026. The BCCI assisted in the approval process. The series structure matters: five T20Is, five ODIs, two Tests. One authorisation, twelve matches.

That is the first curiosity. New umpiring technology is normally trialled in a match, an event, or a single format. Here one approval covers three formats whose acoustic error profiles are not the same.

Sound is a witness, and witnesses carry bias

In 2026 I rebuilt my home-advantage model after looking at 92 behind-closed-doors matches, and the lesson was blunt: leave environmental variables outside the model and the model lies. For a sound-based system the lesson is harsher, because the entire stadium is a variable sitting in front of the microphone. Empty or partial grounds, wind, where the broadcaster places its mics — none of it appears in the approval information. This is the single biggest technical unknown.

The error profile also changes by format. In T20I the highest-leverage reviews are powerplay edges to slip and keeper and death-over caught-behind, where crowd noise is dense and bat edges are thick. In Test cricket it inverts: bat-pad against spin, soft sound, low bounce, bat and pad in the same frame. Assuming one device performs identically in both is an assumption, not a finding.

I built my confessional model to hear what the scorecard would not admit. Sonic is that kind of input: it does not say who is out, it says where the sound came from. The question is how much weight the umpire gives that sentence, and who decides the weight.

Who gets heard first: the question nobody has answered

Here sits the story's largest risk, and it is procedural rather than technical. Sonic supplies information; it does not replace the decision. But if Sonic indicates contact while the existing edge-detection or ball-tracking input indicates none, which input prevails? No published protocol says. The same match situation will be adjudicated on different evidence sets in different series — an inconsistency built into incremental rollout.

The second gap is false positives. Sound does not only come from the bat: pad, helmet, ground, bat handle, and a full crowd all enter the microphone. No mitigation for acoustic contamination appears in the approval information. I am not calling it a confirmed flaw; I am flagging it as an open question.

The third layer is commercial and administrative. The most powerful actor here is not a board but the vendor — ICC-recognised since 2026, and the integration layer for the new input. DRS is a bundled product: ball-tracking, edge detection, audio. Each added input strengthens the incumbent's bundle and raises the wall for challengers. Series-by-series approval means fresh sign-off each time, a recurring revenue gate for vendors with established ICC relationships and a barrier for smaller entrants.

There is a quieter cost. DRS technology is procured by host boards and broadcasters, so the trial lands in the world's largest cricket market, with the host board in a facilitating role — maximum visibility, maximum exposure.

The story written before the evidence arrived

The prevailing narrative is simple: technology will finally settle bat-or-pad. That narrative is institutionally sponsored, not evidence-led. The source material itself concedes that previous technology reduced controversy without ending it. The historical conversion rate is poor, and Sonic belongs to the same history.

The number that matters most: at the time of writing, Sonic has zero matches of data. Its first use is the series that has just begun. Concluding from a zero sample that controversy will fall is model worship, not analysis. Fewer reviews would not prove better decisions either — batters might simply be reviewing less. Correlation and causation sit close together here.

I will state a falsifier before I finish: if the ICC publishes a written decision hierarchy before the Test leg — which input prevails, under what standard — my central concern is wrong.

One more thing is being looked for in the wrong place. Attention is fixed on the T20I leg because it comes first and is most visible. The real test is at the end, across two Tests, where bat-pad against spin is exactly where visual evidence is weakest and acoustic input is worth most. It is also where reputational risk concentrates, because twelve matches sit under one approval.

The question I care about has shifted. Not whether Sonic is right, but whether the governing body can integrate a new evidentiary input without two identical edges in two series producing two different outcomes. Sonic did not remove the umpire; it made the umpire's certainty doubt itself. That is good news, provided the resolution of that doubt is also written down.

The two Tests are the signal. If a bat-pad review is settled on acoustic evidence and the ICC then publishes a precedence rule, the technology is becoming an institution. If decisions change and rules do not, cricket returns to the old place where technology does not end argument, only changes its face. The simpler question remains: is cricket learning to add evidence, or only learning to look at it?