The Raid-Point Trap: One Mistake Shared by Kabaddi's Coverage, Selection and Auction

In the summer of 2026 I was a statistics runner at the Ahmedabad district kabaddi trials. Twelve matches, forty raiders. I imposed one rule on myself: mark every successful raid, mark every empty raid, and use a different colour for every do-or-die raid. Three days later the sheet threw up a 19-year-old with a 62 percent raid success rate whose name had not entered a single selection argument. The selectors did not know him. They knew the boy who had topped the raid-point chart at a tournament the previous season.

I gave that sheet to a local coach. He put it on the table in a committee meeting, and the boy got picked. His team won three of its next five matches. There was no cinematic twist in that. It was simply the moment I understood that the weakest data system in kabaddi is not the scoreboard but the selection committee's memory.

I learned to call the raid only after I had counted the footsteps. Even now, when a franchise sits at the auction table, I see that same sheet — only the column headers have changed.

Context: the league that made the sport legible

The Pro Kabaddi League launched in 2026. Eight teams, prime-time television, and a vocabulary that could explain the logic of the mat to a viewer sitting outside it. Raid success percentage, do-or-die conversion, Super 10, High 5, Super Tackle — this language made kabaddi India's second most data-literate sport after cricket. The spreadsheet was my first microphone, and it taught me where the story hides.

The league has since grown to twelve franchises, commentary boxes operate in five languages, and every season adds another layer of numbers. But the structure rests on an incomplete design. There is a twelve-team men's franchise market on one side and zero women's league on the other, despite India's women winning three consecutive Asian Games golds in 2026, 2026 and 2026. And on the third side sits the most checkable fact of all, the one most often repeated incorrectly: kabaddi was left out of the sports programme for the Hangzhou 2026 Asian Games.

One more verifiable point belongs here, because coverage stands or falls on it. The Government of India has never officially declared any sport the "national game"; the Ministry of Youth Affairs and Sports has said so explicitly. Where a sport's coverage is wrong at the level of its most basic fact, discussing data is pointless. My reputation skepticism aims at these structures, not at players.

Two metrics, and the holes in both

The most common error in kabaddi coverage is treating raid points as performance. Raid points describe a team's raid volume far more than a raider's quality. A team with a quick attacking phase that raids forty times will have a raider who scores twelve points at a 30 percent success rate. A team that raids twenty-five times will have a raider who stalls at eleven points on 45 percent. The scoreboard calls the first man a star, and the auction table bids for that star.

The real difference lies in the denominator. Raid points per raid and raid success percentage only mean something read together. Cricket needed three decades to learn the same distinction between batting average and strike rate. Kabaddi does not have thirty years.

The second metric is tackle efficiency, and here the High 5 is not the most reliable signal. A High 5 is a single-match milestone, produced as much by a defender's mood and a raider's impatience as by skill. The durable number is how many tackle attempts a defender makes per raid and how many he completes. Add a third, uncounted figure: the coverage tackle. When a cover defender forces a raider to change direction and the raider is dismissed three seconds later by someone else, the scoresheet credits the wrong man. The work belonged to the cover.

The do-or-die tax

Remember the rule: the third raid after two empty raids is do-or-die, and failure to score puts the raider out. This is where kabaddi's bravest and most calculated decision is taken in the same second — and where season-end numbers lie hardest.

Picture one raid. The defence is set 4-2, the six-defender condition on the bonus line is satisfied, and the raider has three routes. First, touch the bonus line and return; the risk is a cover already inside, with the return tunnel shut. Second, a dubki into the corner; the risk is an ankle-hold specialist like Fazel Atrachali waiting for exactly that, converting the whole gain into a single ankle. Third, return without a touch; safest on its face, but it burns the raid, hands the team a three-point all-out exposure and doubles the pressure on the next raid.

The raider who takes the third route is filed under "empty raid." The one who takes the first and gets out is filed under "strike rate falling." The one who takes the second and succeeds makes the headlines. All three share one truth: the raid success rate visible at season's end is not an average of ability but an average of risk management. That is why I never rate a raider on total points alone; I isolate the do-or-die subset. A raider who stays above 50 percent on the third raid is the real asset, however modest the aggregate looks.

In 2026, with live sport shut down, I spent four months building a raid-efficiency model on the first six PKL seasons, weighting bonus-line attempts, do-or-die conversion and defender matchups. It flagged the declining effectiveness of Pardeep Narwal's dubki before the general conversation caught up. But I delayed publication by two months and lost an entire content wave. The lesson holds: a perfect model is late news; an unfinished model is on-time news.

Jakarta 2026: when the system beat the talent

In August 2026 the men's semifinal was played in Jakarta. India stopped at 18; Iran reached 27. I was there as a volunteer reporter for campus radio, and three weeks later wrote a 4,000-word breakdown — not an emotional reaction but an anatomy of Iran's wrestling-built defensive structure. The Jakarta semifinal taught me that atmosphere is data with a pulse.

Iran was not an upset that evening. They had reached the 2026 World Cup final and had been co-favourites since. Their edge sat not in the raid but in the shape of the tackle: the ankle hold below the knee, the chain tackle in which the first defender does not stop the raider but forces him to change direction, and the second defender is already waiting in that changed direction. Indian raiders went out in several different ways that evening, but the structure that dismissed them was one.

My target here is not a player. It is selection logic. When selection rewards reputation and aggregate points from a previous tournament, a squad fills up with raiders who, against a good defence, take the same route they took against a weak one. India did not lose talent in Jakarta; they met a defensive architecture for which their raid plan held no alternative. 2026 was not an exception, it was physics.

Every transfer window is a raid with contracts

In auction season the loudest noise is retention and release news, and the least work goes into pricing output. I read every bid as a question: how far does this price run ahead of expected output over the next two seasons? I call it the premium test.

Every transfer window is a raid with contracts: timing, gaps and exposed flanks. A franchise has to weigh three things together — the raider's efficiency base rather than his total points, his matchup history against specific defenders, and his availability record. The third is the least measured and the most expensive. A raider who has been available for one and a half of three seasons shows that absence six months later, when the team has no backup.

When rules change, the market's arithmetic changes with them. The bonus point is awarded only when six or more defenders are on the mat — a condition, not an open door. A Super Tackle exists when three or fewer defenders remain, meaning a team near an all-out can turn defence into attack. In a league where these rules keep shifting, last season's scoresheet cannot set next season's price.

For the 2026 auction I built a salary-cap and team-need matrix across twelve franchises, and since then editors ask for "the matrix" before they ask for a transfer story. That is not a compliment; it is a warning. When the market starts asking you for a model, you owe it the map of where the model is blind.

The pipeline: railways, police, army

India's real selection machine is not the franchise. It is the job. Railway, police and army recruitment give a player economic security that survives an injury-hit season. This system has a side effect rarely discussed: it rewards durability, not peak capability. The player who accumulates steadily across a long domestic circuit is safest inside this structure, even if his efficiency is lower than that of a man who plays fewer matches but takes the hard raids.

The same logic repeats in the awards system, where career totals and years of continuity are counted. My objection is not to those players. It is that a system which measures availability and not risk gradually filters out the players who take match-winning risks. That is an argument about selection logic, not about anyone's intent.

The women's game: three golds, zero leagues

India's women won Asian Games kabaddi gold in 2026, 2026 and 2026. Captains like Payel Chowdhury are the clearest record of that era. What did the sport return for three golds? Not a single franchise league. No bids, no salary cap, no auction table, none of the data scaffolding on which the men's market stands.

I read this as a mispriced market rather than a sentimental grievance. Where a sport holds three continental golds and no league of its own, the biggest pipeline risk is not a shortage of talent but the absence of a market for it. A sixteen-year-old raiding at 60 percent success at district level has no career graph of the kind a male player has — only a job application form and a wait for a trial.

Counter-angle: the efficiency turn will overcorrect

My worry is not that kabaddi has too little data. It is that the next data cycle will work backwards. The moment franchises price raids on success percentage and do-or-die conversion, the market will tilt toward raiders who keep their percentage high by taking fewer risks. The man who takes the dubki on the third raid and occasionally gets out will get cheaper. An efficiency metric punishes bravery — that is not a feature of the model but its limit.

Then there is what I regard as the darkest side of the sport's datafication: the same live feed reaches betting companies' servers first and audiences later. Digitisation enriches coverage and simultaneously feeds a market with no relationship to the sport, only to its numbers. Underdog stories behave the same way — the district-trials kid, the unknown raider of a first season — consumed, discarded, and never followed by any redistribution of resources.

And one thing sits outside every model. In Jakarta that evening an ankle hold landed that should not have landed: the raider's speed was right, his direction was right, his angle was right. The defender shifted his body weight the wrong way, and the raider got caught in that error. My model could not have named that hold in advance. I build models to hear the question, not to silence the match.

Takeaway

The next auction table needs three columns, and all three are currently either missing or in the wrong place: the isolated do-or-die subset, raid points per raid, and a two-season availability record. An editor who demands all three is the one actually separating reporting from rumour.

From eleven years of watching matches, one thing is worth recording: kabaddi's best data sensor is not a selector but a coach who has already counted fifty raids. The question for next season is whether we reach that coach — or go back to buying the boy with the headline aggregate and repeat the same old mistake.