The Kabaddi Ledger: 41,000 Raids, Three Columns, and the Question Nobody Asked

In May 2026, in the summer after my twelfth-grade exams, I posted a three-column spreadsheet on a free Blogspot page: raider, previous season's raid points, auction price. Divide the third column by the second and you get the only question a franchise should care about — what exactly is it buying? The top name on that sheet was Pardeep Narwal. In October 2026, Patna Pirates won a third straight title and Narwal finished with 369 raid points. The sheet had not predicted a price. It had predicted price per raid point, and it was right.

The spreadsheet was right in 2026. It was not asked the question until 2026 — by which point every broadcast had graphics and it had become mandatory to repeat that the Pro Kabaddi League is India's most data-literate sport after cricket. The claim is true. Testing it is still outstanding work.

Kabaddi's rule architecture is an economics problem in a jersey. A thirty-second raid, the continuous “kabaddi-kabaddi” cant, and the do-or-die raid after two empty ones force the raider into a risk he would never take voluntarily. The bonus line pays for depth, the all out returns an entire dismissed team to the mat with two points attached, and the super tackle pays two points to a defence down to three men or fewer. The result is a scoreline that swings fast and a single raid that can flip a match — precisely why kabaddi fits a short prime-time window better than cricket does.

The Pro Kabaddi League built a market on top of that structure: an auction, graded base prices, retentions, a separate overseas pool. The transfer window currently running is loudest with agents and rumours. The real story sits elsewhere — in release-clause structure, in the wage bill, and in the question of what output a franchise believes it is buying when it writes a crore-sized cheque.

Nine years of watching this sport have convinced me the clearest change is not in raiding but in tackling. Defenders no longer chase a raider; they hold a fixed angle at his wrist, his ankle, his waist. It is the shift track and field saw when athletes stopped changing their stride and started changing their starting block — the result shows up not in one season but in three.

In March 2026 the Tokyo Games were postponed, the league was shelved, and my campus shut. I spent five months coding more than 41,000 raids from the first seven PKL seasons into one column structure. Forty-one thousand raids is a census, and once you have counted them, the counting does not stop. The fastest finding was that the dubki — the airborne duck — fell from 9.4% of all raids in 2026 to under 3% by 2026.

The dubki's decline was a coaching decision, not a tragedy. The day defences made the ankle hold a discipline rather than a gamble, the raider in the air became the easiest target on the mat. I went looking for a streak and found a ledger instead — dates, counts, and the specific row where consensus breaks.

Iran's defensive build was sitting in that same ledger years earlier. The wrestling-derived waist and ankle grip, the corner defenders' angular discipline, the tackle-point totals of players like Fazel Atrachali — all of it was public two seasons before Jakarta. In September 2026 I watched the men's semifinal at four in the morning from Delhi: a 23-24 group-stage loss to South Korea, then 18-27 to Iran. Bronze — the first Asian Games since 2026 where India did not win gold.

Plot that bronze against eight years and it reads as a data point rather than a defeat. I did not file a lament that night. I wrote three thousand words on Iran's ankle holds and India's selection by reputation, with Atrachali's tackle numbers as the spine. That file deleted “guaranteed gold” from my vocabulary permanently.

The auction is the same argument from the other side. Weigh a raider by raid points per crore and you learn that the market prices hope while output is accounted for later. Retention is the hardest decision in the sport because it shows you a price and hides a value — and value is the last column anyone fills in. Every transfer has a price. Almost none of them have a value.

The most revealing number in the sport belongs to the women's game. India's women have medalled at every Asian Games where kabaddi has been staged, so the international product demonstrably exists. Domestically there is no franchise league for that product. Kabaddi's largest market-design failure is that its most successful international product runs without a home market. That is the attention economy upside down: where the performance is, the capital is not; where the capital is, the pressure to perform is not.

The institutional picture is dimmer. Friction between the International Kabaddi Federation (IKF) and the Amateur Kabaddi Federation of India (AKFI), and two failed Olympic bids — Tokyo and Paris — stalled on the same two questions: participation spread and the standing of the women's game. Until those columns are filled, kabaddi will stand at the Olympic door without walking through it.

This is where my own method is weakest. PKL data is clean, public and complete, and that is exactly why it quietly becomes the definition of the sport. Circle-style kabaddi has no comparable ledger. Women's internationals have no season-by-season raid log. Pre-2026 domestic play has no database at all. The day I publish a conclusion without naming those three blind spots is the day my own method convicts me.

And the least-said thing: more data did not make kabaddi coverage better, it made it more confident. The “next Pardeep” frame runs exactly as it ran in 2026, only now it arrives with a graphic. A young raider's second season is harder than his first because the defence has now seen him — that line is in my database, and it does not appear in any breakout story.

A model has to be allowed to be wrong. If the dubki's share crosses 5% over the next three seasons — through a coaching shift, a rule relaxation, or a new generation of raiders — my conclusion is falsified, and I will write that. A prediction that does not state its own falsifying condition is not a prediction; it is a position.

The most useful thing anyone can do in this transfer window is pre-register one number — raid points per crore, tackle success rate, do-or-die conversion — and set its threshold before the auction, not after the result. When the next crore lands on a raider, the real question will not be whether he was expensive. It will be: did anyone open the column with his value in it, or did everyone read the one with his price?