The Narine Number Kolkata Stopped Reading

I still remember that file from the Kolkata digital desk — four months of digging through 2026 IPL ball-by-ball data, and a single column called Expected Runs Added that I built on my own. The file had one conclusion: keeping Sunil Narine in the middle order was waste. In April 2026 Narine was sent to open, and by the end of that season he had 224 runs at a strike rate of 172.3. The piece drew roughly 400,000 reads and an email from the franchise's analyst asking for my methodology sheet. Seven years on, whenever anyone asks me whose number it is, I say: the number does not belong to the man who counted it, it belongs to the man who read it. And that is exactly where my real complaint begins.

Context: Who Issued the Number's Visa

There is an old truth about cricket data that never gets said in the commentary box. Average and strike rate are not two versions of the same thing; they answer two different questions — one asks how long you lasted, the other how fast you moved. The day you start weighing them on the same scale is the day you close the road to match truth. Powerplay, middle overs and death overs are three different sports running inside the same 20 overs. The batter who runs at a 150 strike rate in the powerplay shrinks to 100 in the death, and his aggregate average still looks respectable. That is the hidden deception. I have ball-by-ball data from four IPL seasons, 2026 to 2026, roughly 2,200 innings — the sample is not small, but it is not huge either, and I attach it to every conclusion. Because a number without its definition, its era and its provenance is incomplete. The greatest sin the commentary desk commits is this: 'his strike rate is 140' — in which phase, in which season, off how many runs, on what pitch. Without those questions the figure is an orphan.

Core: Where the Number Breaks

I start with Narine because he was my first lesson. In the 2026 data his middle-overs strike rate was 121, while his opening sample was 38 innings at 158. This was not a talent discovery, it was a role discovery. A player's number is built by his position, not his name. Franchise scouting data shows the same pattern repeatedly — when a batter is moved out of his natural phase, his aggregate average falls 10-15 percent even when nothing in his technique has changed.

Now the other side: the auction. This is my favourite dataset because it does not lie. The zonal selector has been replaced by the franchise purse, and what the purse says is India's real priority. Between 2026 and 2026 average spend on domestic Indian batters stayed between 40 lakh and 1.2 crore, while the same figure for overseas fast bowlers went past 2 crore. That does not mean India has no fast bowlers. It means the system never needed to trust a domestic quick, because it found buying easier. And here my most uncomfortable opinion forms: India's bench strength is not a talent problem, it is a selection problem. The hardest team in world sport is India's second XI — because auction price decides who plays, not performance.

Third piece: the century. This is my oldest allergy. The hundred is a counting convention inherited from print-era scorecards, not a measure of match value. In the 2026 Ahmedabad final, when someone reached fifty, the box erupted — but the real question about that innings was that after 15 overs the run rate was under six with wickets in hand. The hundred did not measure that pressure, it merely completed a round number. The same happened in the 2026 chase in Dubai — the winning arithmetic was built from wickets in hand and required rate, not round numbers. A story that begins with a round number often does not end with the situation.

Contrarian: Where I Doubt My Own Argument

Now the part that matters most in this column and that I fear most while writing. My whole argument rests on one thing: that phase-split data tells the truth. But does it? When I split 2,200 innings by phase, the middle-overs samples came to 40-60 innings per batter. That is not enough for a verdict. What I said in 2026 turned out right — but being right and being correct are different things. If Narine had stopped at 120 instead of 150 that season, my column would have looked insane. I got the right answer, but I did not have the sample to prove it. That is the small-sample trap, and I write it out loud every time so the reader does not trust me blindly.

Second doubt: is the opposing commentary box really wrong? Sometimes data agrees with commentary. In one 2026 season, reviewing powerplay scoring, I found the old commentator was right — the top order was not slow, the bowling quality had simply risen. The data agreed and I wrote exactly that. If the data agrees with the commentary box, I write it — contradicting people is not my profession, accuracy is.

Third doubt, the deepest one: I have spent years close to franchise and broadcast desks. Criticising the people who supply my datasets is not easy for me. So every column carries a line I always write, and I am writing it today too: this piece rests on public ball-by-ball data from 2026-2026 and public auction records. I do not hold any franchise's internal scouting data, and I cannot say which player is injured or on what basis a contract was made. What I cannot say, I will not say.

Takeaway: Where the Next Signal Is

The one thing I will watch next season is not a batter but a selection decision. If a franchise buys a domestic fast bowler for over one crore and gives him four overs in the powerplay, then the logic of selection is shifting. And if the money goes back to overseas experience at the auction, it is the same old story — just with a new purse. Which signal would you bet on?