When the Spreadsheet Lies: The Discipline of Cricket Analysis in the Data Age

It is 1:40 a.m. In a small Dhaka flat, the laptop throws blue light on the wall. Fourteen clips of a single match sit on the screen, and beside them an open spreadsheet — rows of green and red cells. The numbers are so clean that the mind wants to accept them as truth. That night I stopped. A tidy diagram can feel like proof, yet it is not proof — it is only a guess, arranged beautifully.

In Dhaka, I built a lab to hear what the crowd cannot. In that lab I learned something that still keeps me awake: the greatest enemy of analysis is not a lack of data, but too much faith in it.

Cricket is drowning in data. Every delivery's speed, line, length and turn is measured. Wagon wheels, pitch maps, fielding positions — all of it now lives on screen. Broadcasts pour out numbers every over, and every franchise keeps a team of analysts behind the scenes. Data is abundant, and inside that abundance hides a new danger.

To grasp the problem, you first have to separate the formats. Test, ODI and T20 are, in truth, three different games. In T20 the powerplay is the first six overs and the death overs are 16 to 20. In ODIs the powerplay rules differ and the field spreads through the middle overs. In Tests the meaning of the first hour with the new ball is something else entirely. An analyst who ignores these distinctions and drops one format's number into another may produce something elegant, but it will be wrong. Without format context, any cricket number is decoration, not analysis.

Sample size is another trap. One match, one innings, one over proves little. A batter makes eighty in one chase and zero in the next — both numbers are true, and neither is his real value. Analysis means recognising the risk hidden inside a small sample. Who demands a larger sample, and who trusts a small one? Skip that question and analysis walks into its own trap.

The most neglected ingredient is luck. The toss is a pure chance element. Dew makes the ball harder to grip in the second innings. Rain brings the Duckworth-Lewis-Stern (DLS) method and rewrites the target. Strip these away, and a claim that "this team lost for this reason" tells only half the story. Unless you separate the toss, the dew and DLS, you will inevitably fuse outcome and skill into one.

This is where my lab works. I watch the game six times, then cut it into clips. On each clip I ask: did this ball's outcome come from the bowler's plan, from the batter's error, or from plain fortune? A dropped catch is not technique, it is an event. An edge through the slip gap is not a plan, it is luck. The analyst's job is to hold that line.

Watching matches over many years, I have learned one pattern: collapses do not simply arrive — their countdown begins earlier, and we fail to read it in time. Belgium 3-2 Japan was not a collapse; it was a countdown we misread. The same holds in cricket's death overs. When a side loses six wickets in the last five overs, the crowd says it "could not handle the pressure." The analyst sees that the countdown began in the sixteenth over, when a set batter could not rotate strike.

Outside the ground, my favourite laboratory was the empty stadium. The empty stadium taught me that silence has a formation. With a crowd, every error is covered. In silence you can see who is truly communicating and who is merely guessing. Cricket is the same — the noise of numbers buries reasoning, and a quiet, patient read reveals the real structure.

Still, I must admit a contradiction. Data analysts have now walked into the dressing room, and their conclusions often detach from the actual rhythm of the match. A dressing room is not a spreadsheet; inside it work a tired seamer's knee, a newcomer's fear, a set captain's instinct. No software captures those.

The franchise auction obeys the same logic. A transfer is not a purchase; it is a system asking a question. An expensive player is a big name, but the right player is the small name that fits the system. Huge auction prices are often a brand race, and real value is added at smaller teams, where a bowler slots neatly into one specific role.

Take one concrete case. In November 2026 in Kolkata, Rohit Sharma scored 264 in an ODI — still the highest individual score in ODI cricket. The number is astonishing, but place it on a Test or T20 scale and it becomes meaningless. The number is true, but its language is ODI only. The first T20 international was played in 2026 between England and Australia; how far the game has travelled since is itself a subject for analysis. Virat Kohli, Steve Smith and Kane Williamson — each built his best years in a different format.

One layer usually escapes the eye: the gap between expectation and reality. The media builds a story, and that story slowly becomes an expectation. When expectation does not match the underlying facts, disappointment is inevitable — and yet that is a failure of narrative, not of the game.

Now my central point, which I repeat in every piece: an analyst's job is not always to answer; sometimes it is to say honestly, "we do not know." An empty result, a refusal, a "the data is not enough" — these are not failures, they are professionalism. An analyst who fills every blank with a guess is not an analyst; he is a storyteller.

This is why I object to the hot-take culture that brands every defeat a "mental collapse" or a "failure of character." Such framing is comfortable, because it requires no thought. The truth is that an innings breaks slowly — one poor shot, one bad over, one changed field. What we call a "collapse" is often a timeline we failed to read in time.

I am a former coach. As a coach I learned that a player's true value can be measured two ways: in the sample, and in the context. One averages fifty at home on soft pitches; another averages forty away on hard ones — who is greater cannot be settled in a single line. Holding that nuance is the value of analysis. In today's cricket that value is being buried under a flood of numbers.

So what is the discipline? My answer is plain. Fix the format first, then check the sample, then strip out luck — and state clearly what you do not know. Follow those steps and analysis becomes slower, but honest. And honesty is what lasts.

In the next match I will therefore watch for this: in which format, on what sample, and after removing which luck factor does this story still stand? The answer may not come today, but three matches later. The question, though, must be asked now — because amid the crowd of numbers, the truth is the quietest voice of all.