The PKL Auction's Real Scoreboard: Prices Follow Reputation, Not Raid-Success Rates

Before the paddle lifts at a Pro Kabaddi League auction, a name appears on screen with a three-season raid-success rate underneath it. Last year this raider sat at 41 per cent. The man next to him in the list, whom no television graphic ever framed, sat at 58 per cent. The paddle went up for the first name and the price climbed four times past what his numbers justified. I was in a corporate studio in Ahmedabad that evening with a notebook open in front of me and two columns drawn—one for price, one for raid success. The columns did not meet.

I first saw this in a district trial. The selectors passed over a nineteen-year-old raider with a 62 per cent success rate and picked the boy whose name carried around the circuit. At the time I thought it was carelessness. Eleven years later I think it was nothing of the sort. It was a system being entirely rational on its own terms, and that same system now runs through auctions worth crores, with a few more zeroes attached.

This is not the transfer window you think it is

The most common error in reading the PKL auction is to treat it like football's transfer window. There is no transfer fee here. No club pays another club. No loan deals. No player becomes a free agent mid-season. What exists is an annual auction: a player is assigned to a franchise, the contract runs one season, and next year everything reopens. That structure makes prices unstable and makes reputation the safest asset in the room.

There is a feature of this market that auction coverage almost never touches: two economies run side by side. The first is the franchise's money—one-season contracts, bonuses, endorsements. The second is the Railways, police and Army job, which for a kabaddi player is a bigger safety net than any auction. A Railways player who draws two crores is having the best year of his career; if he draws nothing, he still has a permanent post and a sport. A player outside the institutional pipeline has only the sport.

That difference sets the temperament of the bidding. The player with a job can refuse a bid. The player without one takes the first decent offer. Two players of identical ability therefore fetch different prices, and the gap does not come from an agent—it comes from the Railway Board. The auction does not buy a player's ability so much as his capacity to walk away.

And because the contract lasts a single season, every franchise is left with one safe strategy: buy the player the audience already knows. A known player sells tickets and shirts, and if he fails, nobody has to explain the decision. The strategy is rational. That is the problem.

Two numbers, one blind market

Raid-success rate is the cleanest figure kabaddi owns: how many raids produced a point against how many ended empty or with the raider out. Next comes do-or-die conversion, the dishonest number—after the third raid every raid is do-or-die, and a failure puts the raider out and costs the team a point. A raider at 70 per cent in ordinary raids who drops to 35 per cent in do-or-die is worth less than a raider at 55 per cent flat, because the auction is not buying the first ten minutes. It is buying the last five.

This is precisely where the market goes blind. Franchises do not buy do-or-die conversion with the seriousness they apply to total raid points or Super 10s. A Super 10 is a match-level figure: it says a raider touched ten points, not when he touched them. The final three minutes and the first ten sit in the same scorebook, in the same column. The auction reads that column and spends crores on it.

I spent four months of the 2026 shutdown measuring that gap. I built a model on raid logs from Seasons 1 to 6, weighting bonus-line attempts, do-or-die success and defender matchups separately. What it shows is that a raider with a high ordinary success rate whose do-or-die numbers sag decays fast in big matches—because big matches are slow, defences settle, and the exit from a do-or-die raid narrows to a slit. A team that picks its second raider on this basis collects the dividend in the last week of the season. A team that picks on name pays for it in the playoffs.

Monu Goyat and one night with a matrix

On auction night before Season 8 I ran the same method. I built a salary-cap and team-need matrix for all twelve franchises—who had how much, who needed which position most, whose retention would come cheapest. The pattern was clean. Patna Pirates' lead raider was Sachin Tanwar, and the problem in front of him was that an opposing defence could simply surround Sachin and stop the team. Patna needed a second raider who could absorb pressure, not another quick raider who would shine on the bonus line.

Monu Goyat's previous season looked poor in the numbers. The success rate had fallen, which is why most people had him outside the retention list. I opened his raid log and found what the scorebook does not show—his do-or-die attempts per match had gone up. The coaching staff were using him as the firefighter. Send a raider only into the hard raids and his success rate will fall whether or not his level has.

A falling raid-success rate is not always falling form; sometimes it is the signature of a changed role. That is something auction graphics never show and franchises never price. I wrote on the basis of the matrix that Patna could retain Goyat, and it ran forty-eight hours before the official list. Two regional dailies picked it up. That night taught me the one auction lesson I still repeat: there are no undiscovered gems in an auction, only players whose numbers nobody has read properly.

The structural discount on defenders

Defenders get treated strangely at auction. Tackle points are a team act. In a chain tackle the credit splits across several defenders, but on the auction sheet it becomes a single line. The defender who puts in the first hand on every tackle does not get the point; the man who arrives third does. That is not corruption, it is the limit of counting.

Tackle-based valuation therefore systematically underprices defenders. Fazel Atrachali is the exception that proves the rule. His price does not come from tackle counts; it comes from the fact that his presence on the mat changes the opposing raider's decision tree. A raider who knows the ankle hold is sitting on his left changes his very first step, and that changed step never appears in a scorebook. A defender's price is set not by tackles but by the fear he plants in the opposing raider—and fear has no column.

The overseas slot: the price of scarcity

The PKL caps the number of foreign players. That cap manufactures scarcity, and markets sell manufactured scarcity dearer than talent. Indian raiders are abundant—dozens enter every auction, so raiding is cheap. Foreign defenders are scarce, so they are expensive. This is not a measure of quality. It is an import duty.

The Iranian defensive model comes out of wrestling—waist pressure, the ankle grip, pulling a body off balance and down. That model proved devastating on the PKL's standard-style mat, and the auction priced it accordingly. But a slice of that price is simply the passport. Lift the cap tomorrow and the same defenders get cheaper without their skill changing by a rupee.

Women's kabaddi: the market that does not exist

The Indian women's team has won four Asian Games golds—2026, 2026, 2026 and 2026. Four golds and not one franchise league. That does not merely mean women players are unpaid. It means there is no raid log. No do-or-die conversion. No auction matrix. A sport without a scorebook has no market, and a sport without a market builds no reputation that anyone can sell.

This is also where the "national game" claim collapses. India has no officially declared national game; it is a checkable fact, and it gets repeated before every major event. Call a sport national while half its players have no league, and the word becomes sentiment rather than policy.

The counter-intuitive turn: data did not fix the auction

Here I have to stand against my own position. For years I have written that the auction does not read data. The truth is less comfortable: data did not fix the auction, it built a second reputation market.

There used to be one kind of star—the one whose name a commentator had already said. Now there are two: the name star, and the "numbers" star, the player a coach ran out in the minutes where he could produce countable acts. The second star is more dangerous, because he launders the old selection logic into the language of efficiency. A coach who wants his favourite can now point at a number on camera.

Who decides the minutes? The coach. Who picks the coach? The franchise, and sometimes the federation. So the raid-success rate on your screen does not measure ability alone—it measures opportunity. A player sent into the third raid every match will always look poor. A player fed the bonus line in the first ten minutes will always look good. The auction sees both in the same column.

None of this makes franchises stupid. It means the thing they most want to measure—whether a boy will panic in the third do-or-die raid of a playoff—is not measurable by any spreadsheet. So they price the thing that is measurable: the name. Reputation pricing is not an error. It is a rational answer to missing information.

The ankle hold the model could not see

In Jakarta in 2026 I sat in the front row of a major tournament for the first time. India lost the semifinal to Iran 18-27. I spent three weeks dissecting that one match—Iran's wrestling-built defence, Fazel's ankle holds, the chain tackles, and the way Iran pushed Indian raiders away from their preferred direction.

One tackle in that match should not have happened according to my model. The raider had already committed his weight forward, the ankle was open, and the angle of the hold looked weak on paper. The hold landed. I went back frame by frame and understood that the Iranian defender had not stopped the raider, he had slowed him—and once he slowed, the other two arrived. The model had measured the angle of the tackle. It had not measured the timing.

I build models so that I can hear the question, not so that the match goes quiet. Jakarta taught me that atmosphere is data with a pulse. The day I believe my matrix can measure the nerves of a third do-or-die raid is the day I stop being a commentator and become a calculator.

What to watch next

Watch three things at the next auction, and none of them depends on a player's name. All three depend on structure.

First, whether any franchise publicly names do-or-die conversion as the basis of a buy. The day that happens, the market will have made its first real reform—because teams will finally be pricing the number that shows up in the last minute of a match.

Second, whether a women's league gets a date. This is not only an equity question. It is a data question. Until women players have raid logs, their efficiency stays invisible—and invisible efficiency is priced at zero, even when it arrives with four Asian Games golds.

Third, whether Railways and police recruitment changes. Transparency there would increase the number of players who can turn down a good bid, and prices would become a little less reputation-driven.

None of these shifts will land next season. But the real question at an auction was never who costs the most. It is who was never counted. Until the scorebook grows a column for that player, the paddle will not rise for him.