The Column Nobody Totaled: An Audit of Indian Cricket's Data Claims

February 2026, Bengaluru. The IPL mega auction was underway and I was volunteering on a Hyderabad digital sports desk — twenty-two, finishing a sports science degree. The ledger lay open on the desk. Column A held the name, Column B the base price, Column C the money the franchise actually paid. All three columns were clean. Shareable.

I got stuck on the fourth column — the one nobody had built. I read Manish Pandey's line: ₹11 crore, Sunrisers Hyderabad. Then I pulled out my 2026–17 phase sheet. Two numbers, one player, and they were not speaking to each other. That night became my first real lesson. The story tends to live in the column nobody bothered to create. In cricket journalism, what you can see arrives pre-formatted — strike rate, average, auction price. What you cannot see — the denominator, the sample size, the date the rule changed — takes work to assemble. And nobody retweets work.

I decided that night that I would do the thing nobody enjoys watching: reconciliation. Before the headline, the reconciliation. Before the reconciliation, the receipts. This piece is the product of that habit, and it is an audit of three places where Indian cricket's loudest claims currently run without a denominator.

Context: the ledger, the broadcast and the gap nobody measures

Public debate in Indian cricket rests on three documents. The auction ledger. The scorecard. The broadcast contract. All three are primary documents; all three contain numbers; all three are read emotionally.

The auction ledger is the most honest of the three because it leaves no room for excuses. Under the old zonal-selector system, selection happened in a room and the reasoning was rarely stated. In an auction the reasoning is a number — purse, retention slab, bidding sequence. When Ishan Kishan goes for ₹15.25 crore (Mumbai Indians, 2026 mega auction), you can at least attempt to match that price against his phase-adjusted contribution across the preceding three seasons. That is what makes the ledger more transparent than any selection committee.

The scorecard is the second document and the most misread. A scorecard records an outcome, not a process. "71 off 47" is a fact; "71 off 47 with a strike rate of 108 until the 14th over of a chase" is a different fact. The second sentence reads badly, which is precisely why it never gets written.

The third document is the broadcast contract. IPL media rights for the 2026–27 cycle sold for ₹48,390 crore, split between Viacom18 and Disney Star. That figure makes headlines every cycle, and it is almost never accompanied by the question of how much of it reaches state-level cricket, Ranji Trophy coverage, or domestic players' match fees. Answering that requires laying three separate ledgers on one table. It does not get done.

And the fourth document, which does not exist at all: the coverage volume of domestic cricket. India is routinely described as the last country where Test crowds still turn up, and broadly that is true. What is never stated alongside it is that the Ranji Trophy survives largely on knockout-only coverage — and that column inches, stream counts and gate receipts all lag in the very country that is the sport's largest market. That gap is not an analytical error. It is a refusal to measure.

Core: three phases, one spreadsheet, and the anomaly that refused to average out

I opened the 2026 auction ledger first; the story was in the column nobody read. I am writing that sentence twice because it is the entire lesson. The ledger carried a price for every player and a projected return for none. So I built that column myself.

The method was straightforward. Take 2026 and 2026 IPL ball-by-ball data, cut it into three phases — powerplay (1–6), middle (7–15), death (16–20) — then extract the phase relevant to each batter's role: for an opener, powerplay strike rate and the middle-over decay; for a finisher, death-overs strike rate and boundary percentage; for an anchor, balls per dismissal.

Manish Pandey's case shows both the strength and the weakness of the approach. His 2026–17 middle-overs record was good — an accumulator who did not stall the run rate. But the team that paid ₹11 crore for him did not have a middle-overs problem. It had a top order that struggled to sustain pressure, and a death phase that needed acceleration rather than insurance. The ₹11 crore question was never whether the player was good. It was whether he solved the team's most expensive problem.

I wrote one line in that sidebar deliberately: this is a check on process, not a forecast of outcome. Sunrisers reached the final that season and lost to Chennai Super Kings by 8 wickets. I had two options. Concede that I was wrong because the result went against me. Or write that reaching a final and winning one are different metrics, and that a single knockout is an experience, not a data sample. I chose the second. A senior producer on that desk called me "the data girl." I replied with a four-page auction audit. The editor ran it, and a franchise data contractor offered me a junior role.

This is not autobiography. It is evidence that measuring the gap between price and value is still somebody's private habit in Indian cricket, not an institutional method.

Then comes the part I call regime-change accounting. It is the trap most analysis falls into, and the trap is invisible because the old spreadsheet still looks comparable.

The Impact Player rule arrived in the IPL in 2026. It meant twelve players, not eleven, could perform on-field roles in any given match. What changed after the rule was that teams could carry a seventh batting option and a fifth specialist bowler. The consequences: more spin in the death overs, a rewritten rationale for all-rounder pricing, and — most importantly — a shift in the benchmark for top-order run rates.

The mistake happens here. Someone averages 2026 death-overs strike rates against 2026 death-overs strike rates and announces that batting standards have risen. Perhaps they have. But the comparison has blended two things: genuine player improvement and a rule change. When a rule changes, a year-on-year comparison becomes a comparison of two different games, and it must be footnoted the way a share price is footnoted after a split.

Net run rate is the same problem. It is a metric that never explains itself. When a team qualifies on a large margin, that margin is a blend of the team's performance and the opponent's collapse on one particular day. 200 in twenty overs is capability; bowling the opposition out for 90 can be a one-day event. Both land in the same column, and then the column is treated as destiny.

The third instance is WPL expansion. When a league adds teams, its average score, average strike rate and youth-opportunity index all move at once — and not in the same direction. More teams means a thinner talent pool, which is why some metrics fall. That is arithmetic, not decline. The headline still reads "standards have dropped."

My first instinct was to count. My second was to count again. The second count asks one question: is this year's denominator equal to last year's? If not, any percentage increase is a new denominator, not a genuine leap.

Then came the season that taught me most. 2026, Covid, the IPL in the UAE, empty stadiums. I worked that season from Hyderabad inside a franchise's data operations. It was the season that exposed how fragile "home advantage" is as a concept.

Three cities, three grounds, no spectators, and every match at a geographically neutral venue. Under those conditions the only things you retain are dew, pitch behaviour, and prior experience. I built a neutral-venue and dew model. Dew makes batting easier in the second innings, which raises the value of the toss. I audited that season's toss data separately and wrote an internal note whose gist was this: calling the toss "luck" is convenient, but if the win percentage of toss-winning teams sits well above their performance-based projection, that is structure, not luck.

Qualifier 2 arrived — a 17-run defeat to Delhi Capitals. I wrote the post-mortem and refused to call the result bad luck. The audit found that the chase plan contained no phase-specific matchups at all. Which batter faces which bowler in which over was not pre-decided; it was reactive. Seventeen runs in T20 is one over's worth of cricket — but which over conceded those seventeen is the entire question.

The empty stadium did not lie; the attendance model did. There were no spectators, so the old justification for home advantage vanished. What remained was pitch and dew. When crowds returned the following season, plenty of people announced that home advantage was back. I have never accepted that claim. If home advantage rests on crowd noise, demonstrate it against a decibel meter. If it rests on pitch identity, demonstrate it against the curator's practice. Until then, home advantage is a statement, not a variable.

One more case. In 2026, building my phase-adjusted model at the University of Hyderabad, I stopped on Bhuvneshwar Kumar's numbers. His 17–20 economy was 7.05 — excellent for its time. Placed alongside his own team's top-order strike rate, the picture changed. A bowler's death economy can look good for two reasons: he is genuinely bowling well, or the opposition is already so far ahead that it is not taking risk. In the second case the economy flatters the bowler while the real cause is match structure. I wrote exactly that in my review of the Eliminator against Kolkata Knight Riders — that the team's core problem was top-order strike rate, not bowling, and that the strong 17–20 figure was covering for it. The headline blamed the bowling.

Contrarian: correlation is not causation, and "no adverse finding" is not "clean"

This is where I have to be most careful with my own method.

First. When a team reaches a final, we retrospectively validate its auction strategy. That is inverted logic. Reaching a final is one observation — one season, one bracket, a handful of matches that could have turned on an over. Sunrisers reaching the 2026 final does not validate every purchase at that auction, just as losing the final does not invalidate them. The correct question is: if that same auction were replayed twenty times, how often would that squad reach the final? Nobody asks it, because the answer is a count and counts carry no emotion.

Second, and this is the heart of the piece. When every column in an analysis is empty — no title, no source, no facts — the correct conclusion is not "no issues found." The correct conclusion is "the analysis could not be performed." Those two statements look alike. They print alike. The distance between them is the foundation of all reporting.

I have seen this up close. A template once arrived fully built — eight sections, a subheading under each, blank space beneath every one. Three lines in, it was obvious the underlying material had never existed. But the template looked so complete that the pressure to fill it was real. That is the actual hazard: when the frame precedes the content, the frame starts inventing content.

Cricket's daily version of this is transfer and auction rumour. "Franchise X bid ₹Y crore for player Z" travels in hours. "Player Z's phase-adjusted value across three seasons comes to ₹Y crore" takes weeks to build, if it gets built at all. At the January 2026 auction in Jeddah, Rishabh Pant went for ₹27 crore and Shreyas Iyer for ₹26.75 crore. Those two numbers made every headline. The number that never appeared beside them was the expected wickets-per-match contribution of that price inside that specific squad structure. A rumour moves in hours; a registered fee moves in weeks. I work in weeks.

Third — format. Test, ODI and T20 numbers are not interchangeable, and the point is so basic that restating it feels pointless. Yet it surfaces every season. A Test opener's average of 42 and a T20 finisher's average of 42 are different objects, and putting them in one table to crown a "batter of the year" is not an arithmetic error, it is a reasoning error. Until the format is fixed, no comparison should be attempted — because a comparison means you placed two things on one scale, and if there are two scales you have measured nothing.

Fourth — sample size. Honest declarations of small samples are rare in Indian cricket writing. "Two fifties in three matches" is a good story but not a trend. A trend needs at least a full season, preferably two. Early in my career I wrote that the sample was too small to say anything. The editor sent it back — "the reader wants an answer." I did not change the line. This article belongs to the same habit: where there is no data, I will not write a guess.

Takeaway: what to watch next

Watch three things next season, and watch them in the table rather than the headline.

First — for every major purchase, ask what the player's 16–20 strike rate or economy adds to resources the team already has. If it adds nothing, the price is a market story, not a team need.

Second — beside every year-on-year statistic, write the date of the relevant rule change. Impact Player, NRR, WPL expansion: none of these leaves an old spreadsheet comparable.

Third — count domestic cricket's coverage yourself. How many Ranji matches are streamed, how many column inches are allotted, how many spectators come through the gate. Put all three numbers on one table. India is the sport's largest market and this gap sits inside that same market — not a contradiction, but a column nobody has totaled yet.

Next season, when someone says standards have risen, ask one question: have they risen, or has the denominator changed?