The Do-or-Die Lie: PKL's Raiders Are Getting Pricier, Not Better

Row 419 of my spreadsheet has a boy whose name I wrote in the comment column and not the name column — "third raid, 32nd minute, right corner, got away." That night was 2026, a hostel room in Delhi, a Season 6 replay on a laptop, and my cousin's Redmi in my hand. The phone was borrowed, the cut was mine, and the rumour did the rest — the 47-second vertical edit stacking 14 of Pardeep Narwal's dubki escapes from Season 5 pulled 9,000 views in three weeks and more than 200 comments, all of them inside one argument: is the dubki skill or a con job.

That argument taught me the thing my entire living still rests on — people argue with opinions, never with counts. So I started counting. In 2026, when the world was shut and empty stadiums were being replayed on loop, I rewatched roughly 180 matches from PKL Seasons 5, 6 and 7 and charted 600 do-or-die raids into a Google Sheet: raider, tackle type, outcome, minute. Lockdown made every raid do-or-die; the spreadsheet made every death a cell.

The result: do-or-die conversion climbed from 41% to 58% across three seasons, and dubki attempts fell by a third. That is the number that got me a ₹1,500-a-piece email from a Noida digital desk within 48 hours. And it is the number that the league, the broadcaster and the auction room all read wrong.

A 17-point improvement is not a raider getting better. It is the side effect of a changed match economy.

The context every preview leaves out

The PKL do-or-die rule is flat and unforgiving: after two consecutive empty raids, the third raid must score or the raider is out. The bonus line runs alongside it — one point for crossing it, but only when the defending side has six or more players on court. So a single raid opens three different risk-reward equations for the raider, and his choice is set not by his ability alone but by what his team intended in the first two raids.

What changed around Season 5 (2026) was that coaches began using the first two raids as setup raids. A bonus-line touch-and-return on the first, an empty return on the second after reading the defensive line, and then a fully rehearsed entry on the third — that became a plan, not a compulsion. When I added a column for "outcome of the previous two raids" next to the do-or-die column, the picture cleared: in Season 7 the number of cases where the do-or-die raid had been deliberately manufactured, rather than accidentally stumbled into, had gone up.

This is where the first comfortable conclusion dies. Most Hindi web copy and most commentary says raiders learned to play under pressure, that mental strength arrived. What I saw in 600 raids was not strength. It was foresight. If you know the third raid is yours, you do not play it like a do-or-die raid; you play it like a normal raid, where the first ten seconds go into reading the defensive positions and only then do you commit. A raid played with preparation converts better automatically. Skill is in there, but the biggest share belongs to the thirty-second clock that did not get spent on the first two raids.

And here the second thing attaches — a roster artefact. The three-raider system thinned out. As teams began going into matches on two raiders, the third raider's raids shrank and the star raider's raids grew. Which means the do-or-die raid now belongs to the team's number-one raider rather than to the average one. Change the sample and the average climbs by itself. That is the real statistical story sitting behind 41% to 58%.

Where the auction bought the wrong number

When I built my rupees-per-raid-point model ahead of the Season 9 auction in September 2026, the market's anchor was still Pardeep Narwal's ₹1.65 crore — UP Yoddhas bought him in the August 2026 auction for Season 8, and it was the record at the time. I had no quarrel with the number. My quarrel was that the number had set the price of the entire pool, and that price rested on volume, not on efficiency.

Here I will borrow a football-transfer comparison for two sentences and then return to kabaddi's own vocabulary. Six hundred raids taught me more about football transfers than any fax machine ever could — because in both places the market pays for what is visible, not for what wins. A centre-forward with 20 goals sells for more than one with 12 goals whose team wins nine of those twelve. In the PKL the raw raid-point total sells; the do-or-die conversion does not.

Total raid points is a count. Do-or-die conversion is a court. The auction room buys the count and the coach pays for the verdict later.

Think of it this way. A raider who attempts 200 raids in a season and brings 90 points has a strike rate of 45%. A second who attempts 140 and brings 80 has a strike rate of 57%. The auction will bid higher for the first, because his point total looks bigger. But the first raider forces the team to feed him 200 raids — which means 110 raids are burned. The second saves 60 raids, and those 60 can be handed to other hands. In tackle points the difference converts directly into the score, because every empty raid gives the opposing defence a breath.

My timestamped Season 9 experiment is useful here. Seventeen calls, eleven landed. Six bargain raiders, three overpriced, one predicted final-round steal. Of the six misses, four were cases where I underweighted injury and minutes. Of the eleven hits, the most valuable one gave me the least pleasure — a raider who went for ₹30 lakh with an ordinary point total but a do-or-die conversion above 60%. I had marked him in red three years earlier, not because he was spectacular but because he never repeated the same mistake.

The market did not see him. The market saw the highlight reel.

The part nobody counted — the dubki shrinking

The second finding in the sheet is more uncomfortable. Dubki attempts fell by a third across three seasons, in exactly the window when highlight packages, thumbnails and reels multiplied several times over.

The dubki is the escape where the raider ducks under a tackler's arm — pure physical risk, where a single frame of mistimed movement guarantees the tackle. It is kabaddi's most photogenic move and also its worst ROI, because success depends on the raider's body rather than on the defender's error.

So why did attempts fall? Three answers, all pointing the same way.

One, injury arithmetic. When a raider is playing 140 of his team's 200 raids, his body is team property, and the coach will not let it go airborne. The dubki shrank because permission shrank.

Two, the defence read it. The dubki's entire business model depends on the tackler committing. In Seasons 5 and 6 defenders lunged with arms wide; by Season 7 they had learned to hold distance with the leg forward. Attempts did not fall because raiders got scared. They fell because the return on attempting fell.

Three, and most uncomfortable, analytics made risk invisible. Once a successful-raid-percentage column exists on a desk, the reward for a risky escape is zero. The raider who completes seven of twelve raids safely looks good. The one who tries the dubki seven times out of twelve, succeeds four and is out three times looks bad — even though in that third case he dragged the defence forward and opened the bonus line, which no cell records.

This is where I begin to doubt my own sheet. A spreadsheet can hold six hundred do-or-die raids and still miss the point, because the things it does not record — fear, fatigue, the effect on the next raid — are often what decide the match.

Where I could be wrong

My strongest counterargument is this, and I am putting it first: 600 raids come from 180 matches, roughly 3.3 do-or-die raids per match. The sample does not cover Seasons 8 through 11 — the four seasons of the league's biggest broadcast expansion and heaviest auction inflation. If conversion rose above 58% in those four seasons, my plateau claim collapses.

The second objection is about charting method. I logged each raid's outcome from the broadcast angle. When the camera sits behind the raider, an escape looks bigger and a tackle looks smaller. Some of my "got away" calls will certainly be cases where the tackle actually landed and the raider escaped past the line, or where the umpire gave the raider the benefit of the doubt. My 58% is not a clean number.

The third objection is the heaviest. Saying the conversion rise is "only" the rule and the roster insults the raiders who genuinely changed their technique. Pardeep Narwal did not throw the same dubki in Season 3 and Season 7. In later seasons he jumped less and played more touch-and-return — a deliberate decision, not an accident. If I decide the improvement is purely systemic, I repeat exactly the mistake I made after Jakarta in 2026 — I mistook being noticed for being right, and the thread kept receipts.

The receipt is still with me. August 2026, the Jakarta semi-final — India lost 18-27 to Iran, having already fallen 23-24 to South Korea in the group. India failed to win Asian Games kabaddi gold for the first time. At one in the morning I posted a twelve-tweet thread arguing the defeat was selection by reputation, that India had no answer for Iran's wrestling-derived thigh holds. Four hundred quote-tweets came, and two follows. Those two follows taught me to stop writing "we lost" and start writing "this is the number that lost it."

That same lesson applies against this article. My 41-to-58 curve covers three seasons. A trend shows in three seasons; it does not prove. I know Fazel Atrachali became the PKL's all-time leading tackle-points defender — that too is a story built on continuity, not on three seasons. If I trust defence by my own standard, I must also accept that a conversion plateau could be a defensive victory rather than a raider's decline.

What nobody is counting: women, and an empty auction room

A splinter comes out here that bothers me most.

India's women's kabaddi team owns four Asian Games golds — Guangzhou 2026, Incheon 2026, Jakarta 2026, Hangzhou 2026. The men's team has golds too, but the one men's gold India missed (2026) is still invoked in every preview, while four straight women's golds never reached an auction room, because the road there was never built. There is no franchise league.

My do-or-die sheet does not contain a single women's raid. That is my carelessness, and it is the ecosystem's carelessness. We are building careers on data from the half of the sport whose data nobody collected. If you do not know the do-or-die conversion of the raiders who won four Asian Games golds, you hold only half the answer to this argument.

So what should an auction buy?

My clear suggestion, which I repeat every season and which nobody hears: price a raider on three columns — total raid points, do-or-die conversion, and seconds spent per raid. The third column is the most important and the most absent. A raider who ends a raid in 20 seconds gives his team ten seconds that go into the next raid or keep the defence from resetting. It has no single number, so it has no price.

I know this model will not win an auction room. The auction room plays counter-strike — whichever player a team stood up for, that team will take at any price. I accept that. My job is to keep remembering that being sold is not proof of being right. I learned that in Class 11 on a borrowed phone, on the day the dubki cut travelled faster than the truth while I was still holding the phone.

What comes next

My public, timestamped prediction for Season 12, and I will tweet the miss exactly the way I did in Season 9: the league's do-or-die conversion will plateau at 58% or fall over the next two seasons, and the fall will be sharpest in teams without two or three raiders of depth.

The reason is simple. At the same time raiders were converting the first two raids into setup, defenders were learning to read the setup. A defence that knows the third raid is the real raid can hold its commitment through the first two — and in the Season 7 footage that is what I see, defenders lunging less in the first two raids. That adjustment is now complete. The do-or-die raid will only be worth as much as the preparation behind it, and once every coach prepares, the preparation edge disappears.

Before that day arrives, watch three things, and I will stake the claim that they will show before the do-or-die numbers do: a raider with conversion above 55% is bid to parity with a total-points raider for the first time in an auction; a broadcast desk starts using the phrase "strike rate"; and a major franchise publicly demands seconds-per-raid data. If none of the three happens, my model is wrong and the auction room is right.

When I borrowed my cousin's Redmi in 2026 I had one clip, one number and one question. Today I have a 600-raid sheet and the same question. Everyone knows how many raiders the league turned into crorepatis. Nobody asked which number earned those crores — and which one just looked good on television. At the next auction, when a raider's name is called, look once at how many of his do-or-die raids are still alive in the sheet. The rest is a highlight.