Who Keeps the Receipt for a Tracked Ball?

September 29, 2026. Three numbers sit side by side in my notebook — 142.3, 144.1, 140.9. All three are the speed of the same delivery bowled in the previous night's Asia Cup final in Dubai. All three were shown on broadcast. Nobody was wrong. Three systems simply measured the ball at three different points of release, and nobody showed the viewer where the measurement was taken. Timelines raged about a broken speed gun. The real story was never about the sensor; it was about the receipt. A number without a stated point of origin is not evidence — it is a claim.

I ran that first audit because the eye test had no receipts. Since leaving the print desk in October 2026, I have opened every piece with a variance line: what the scoreboard says against what the data says. In football it was scoreline versus xG. Cricket is harder, because here data does not merely describe — it rules. Out, not out, auction price, contract bonus, workload cap: all of it is now a number's decision. The print desk died the day I learned to query the match instead of narrating it.

What the machine measures, and what it does not

The Decision Review System first took the field in July 2026, during the India–Sri Lanka series. Seven years earlier, in 2026, Hawk-Eye had entered cricket through English broadcast. In the two decades since, cricket's information economy has split into clear tiers: boards own, broadcasters carry, data providers measure. Boards want play, broadcasters want drama, providers want precision. Three interests are not one, so three numbers rarely agree.

In my own work I tier sources four ways. Ball-tracking and stump mics at the ground are live primary. Broadcast graphics and live scorecards are secondary, because human hands intervene. Board press releases are institutional, and therefore interested. Press-box anecdote is tertiary at best — I never let it carry a verdict. I have watched this game for over four decades and queried it ball by ball for seventeen years. For me the eye test is a hypothesis, never a conclusion.

Where the chain of numbers breaks

Speed is the weakest link. Between two broadcasters in the same match, average pace can differ by two to four kilometres per hour. That is not a ranking error; it is a difference in sensor placement. Release point, camera frame rate, distance from the pitch — small gaps accumulate into large ones. The figure then lands on a website as 'fastest ball', and by morning social media has made it fact. Taskin Ahmed's 147 and Jasprit Bumrah's 145 are meaningless as a comparison if the two numbers came from different pipelines.

Ball-tracking is subtler and riskier. Manufacturers claim millimetre-level accuracy, but the verdict emerges only after a series of estimates: where the ball pitched, how much the surface gripped, which model bent the trajectory after bounce. Each step leaves uncertainty behind. 'Umpire's call' is not a mysterious rule — it is an admission of measurement uncertainty, an error bar dressed as law. Show the frame before and the frame after, and a good number of reviews would flip.

Strike rate is my other objection. A strike rate of 140 in the powerplay and 140 at the death are two different products: in the middle overs boundaries are available, in the last five you must run. Without dew, pitch sponge and floodlight variables, comparing two batters means placing answers to two different questions side by side. Litton Das's powerplay hitting and Shakib Al Hasan's middle-overs manipulation are both valuable, but measuring them on one scale produces bad decisions — which is exactly what auction tables do.

Auction prices and contract structures are the commercial form of that error. Valuation model inputs are not public; the output is a single number. A transfer rumour is just a row waiting for a primary key. Knowing the guaranteed-fee to performance-bonus ratio tells you whose neck carries the risk. Players from smaller boards typically sign with more bonus and less guarantee. That is not a talent gap; it is a price-structure gap.

This is where blockchain enters, and where my scepticism peaks. Fan tokens, ticketing and automated contract triggers have all been discussed inside cricket in the last two years. The logic is clean: if a contract says 'bonus after fifteen matches', and match data lands in an immutable ledger, intermediaries disappear. But a ledger does not create truth; it makes truth permanent. Wrong input becomes permanently wrong — and what cannot be erased cannot be corrected. If a board writes its own numbers into its own ledger, power does not shift. Only the record book gets louder.

Technology does not reduce error, it relocates it

June 2026 was the month the crowd became a control group. Across 92 matches behind closed doors, home win rates fell and home penalties dropped — proof that home advantage is a measurable variable, not spiritual energy. That lesson travels to cricket, and in Asia it bites harder: travel miles, rest days, temperature, schedule density. An English county calendar and an Asian circuit calendar are not the same object, so running one model across both produces noise.

The biggest misconception about technology is that error shrinks. Error moves: from the umpire's eye to the camera, from the camera to the algorithm, from the algorithm to whoever bought the model. And the story cricket loves most — the small side flattening the giant — sits on wage gaps, support staff and infrastructure. If Rashid Khan spins a match alone, that is individual skill. The system that gave him a ground as a boy is the real variable. Look at the academies and the arithmetic is blunt: fewer than one in ten players gets a genuine first-team pathway. The rest wait on domestic structures.

What I will watch next cycle

Next cycle I will watch one thing: which board first publishes a provenance standard for its own ball-tracking data — measurement point, frame rate, error margin, correction history. Because without the receipt, cricket's next generation will learn the price of things and never the truth of them.