The Blank Column on the Retention Sheet: The Number Nobody Is Sending This Transfer Window

In the second week of this transfer window, a franchise sent out a note confirming a new deal with one of its raiders. One hundred and ninety words. It contained "key member of the senior side", "the spine of the dressing room" and "outstanding last season". Not a single number. I called the media officer and asked for the raider's success rate from the previous season. The answer was: "The stats team will have that, I'll get it sent across." Three days later, nothing had arrived.

In my notebook there is a page dated that same day with one line on it: the number did not come. The habit was built in 2026, during Season 5, when the league published nothing usable and I logged every raid, every cant, every empty raid by hand off the broadcast feed. By the final I had 1,100 rows. That season Patna Pirates won a third straight title and Pardeep Narwal finished with 369 raid points, a record at the time. Nobody on the desk could explain what those 369 were made of. Eight years later the same question has returned in different clothes.

The mechanics of the transfer window are simple. Retention first, then releases, then the auction. Each stage produces three documents — the retention sheet, the release list, and the wording of the contract. What none of the three contains is the performance base. A franchise will tell you how important a player is; it will not tell you how many raids he ran, how many points he took from them, how often he was out, how many minutes he stayed on the mat. The public data is incomplete too. Raid points and tackle points are available, but the count of empty raids, the conversion rate on do-or-die raids, the bonus-line attempts and their outcomes all read differently across portals. Over the last few seasons I have pulled the same player's figures from three places and found a gap of two to four every time. Where the portal and my notebook disagreed, I published the disagreement rather than the cleaner number.

The real problem is that the entire transfer-window debate runs on volume, not efficiency. 369 raid points is a sum. It tells you how many points a raider scored; it does not tell you in how many raids. In Season 5, Pardeep's raid count and success rate differed across feeds because each feed defined a "successful raid" its own way — where a raider touched a defender and was caught on the way back, one feed counted it as a successful raid, another counted nothing. Until the definition is fixed, comparing 369 with another season's total is dishonest. Two raiders with 200 and 180 total points are not the same player — one may have run 400 raids, the other 250. The auction bid does not see that gap.

The second layer is the relationship between price and output. My method is plain: divide any bid by the actual production of the previous two seasons, and see where the quotient lands. In most cases the premium is paid for highlights, not performance. A raider who produced two or three spectacular multi-point raids in a season has a clip circulating on social feeds, and that clip becomes the floor for the bidding. The player who takes two points every match and keeps his side alive has no clip. In the pages I have kept by hand since Season 5, the most space belongs to exactly those "boring" raiders who do not waste raids. In the transfer window they are priced the lowest.

The third layer is payroll, and the largest gap sits here. In 2026 the league went dark. Season 8 was postponed indefinitely, the stadiums sat empty, and I had no matches to be sent to. I took a train to Sonipat and spent five weeks through the Haryana–Punjab belt, sitting in drawing rooms with the forty-odd players whose contracts had frozen mid-cycle. Most were back on railway rosters or in fields. I logged each one on a single line: contract status, monthly income, and whether AKFI had answered a single email. In the column marked "reply", thirty rows carry a blank mark. That file remains my most useful document, because it shows what the 2026 auction changed and what it did not. Before it, the national squad was built from railway clerks and police constables; after it, the crore-rupee talk began, while job security stayed exactly where it was. When a crore figure flies around a transfer window, I hold it against that file. The ledger said 369. The season kept a different count.

An outsider will say the market already knows — whoever a team retained is good, whoever it released has fallen. That reading is easy, which is why it is wrong. The market prices reputation, not performance, and the retention sheet does the same. Remember Jakarta 2026. India lost the semifinal 18-27 to Iran, the first gold the men's side had ever missed. Framing it as a betrayal is the easiest route available. I spent eight weeks coding every Iranian tackle from their matches going back to the 2026 World Cup — Fazel Atrachali's ankle and thigh holds, the chain tackles, the wrestling-built defenders who absorb a raider instead of chasing him. Atrachali went on to top the league's tackle points, and that is a figure you can still check today. 18-27 was not betrayal, it was physics — and that Indian side was picked on reputation instead of form. The same selection logic is repeated at every auction. Jakarta gave me eighteen to twenty-seven, and a file that would not close.

One correction costs nothing and should be made. India has no official national sport — the answer to a 2026 RTI says exactly that, and printing "national game" in every preview does not make the answer untrue. In the same way, calling the league the country's biggest should always place it second, behind the IPL, and pair the claim with the fact that women's kabaddi owns four Asian Games golds and not one franchise league. That comparison comes from inside the record, not from the promotion around it.

For the lists that arrive next week, my method is already fixed. Ask for three columns against every name — how many raids, how many successful, how many minutes. A list without those three is not news, it is an advertisement. And the file that came back empty this week is also a record. It stays open, because a file is just a promise to keep looking. I do not trust a number until I have walked to its source.