The Do-or-Die Math: Why PKL's Raid-Success Rate Is Its Most Expensive Vanity Metric

One raider on my tracking sheet last domestic season finished with a 68 percent raid-success rate — top five in the league — and his team missed the playoffs by three wins. The second column of that same sheet said something else: his do-or-die conversion was 41 percent, below the league average. The broadcast showed the first number. Nobody showed the second.

I found the format at 38; the voice arrived later, carrying receipts. On 22 October 2026, in the Ahmedabad press box, the scoreboard read India 38-29 Iran and the senior beside me had already typed his guaranteed-gold headline. I was stuck in the second-half tackle column. The column my editor spiked that night — 'too negative' — became my first script. Nine years later I am putting the same question to the league's regular season, because the biggest lie of this season is running in the brightest font: raid-success rate does not win you matches; do-or-die net does.

The league's statistics ecosystem has thickened over the last few seasons. Raid points, tackle points, super raids, super tackles, bonus points and raid-success rates run across every screen. Fantasy teams are built on those numbers, and auction bidding rises on them. At the 2026 auction, Pardeep Narwal drew a bid of roughly ₹1.65 crore, and that bid was built on a mountain of raid points, not on a tackle column. The market sent a clear message: the man who scores points is expensive, the man who stops points is cheap.

Remember the rule structure. After three consecutive empty raids, the third raid is a do-or-die raid — no points and the raider is out. On the other side is the super tackle: three points if a raider is stopped with three or fewer defenders left. Those two rules changed the real economics of a match, yet the graphics still put the old number on top. I fill a sheet for every match, in three columns: raid success, do-or-die net, and last-ten-minute tackle conversion. Where my numbers collide with official league figures, the official figures win and I will explain the gap in a correction video.

To understand the real accounting you have to open up a failed do-or-die raid. The raider is out, so the team loses a point; the opposition gets a revival, so one of their men returns to the mat; and the turn changes, so the next raid goes to them. A failed do-or-die raid is really a two-point swing, and the tempo loss sits on top of that. Raid-success rate dissolves all of it into a single fraction, where a bonus taken on a safe first raid counts exactly as much as a raid played for the team's life in the 30th minute.

India's raiding temperament has changed too. The dubki died because defences learned to tackle low — that is a coaching decision, not a tragedy. The bonus line and the hand-touch replaced it, and both inflate raid-success rate. The raider piles up safe points and stays in the headlines while his team sinks after the third empty raid. In the matches from last season that turned in the final five minutes, the losing side's do-or-die conversion was below 40 percent almost every time.

The international receipts say the same thing. That final on 22 October 2026 belongs to India at 38-29, but the story of the match was the tackle points Iran kept accumulating in the second half — nobody lost the first half, India's raiding structure was squeezed in the second. Russia paid for the flights; Jakarta paid for the credibility: on 23 August 2026 India lost the Asian Games semifinal 18-27 to Iran and settled for bronze; the next day the women's final went 24-27 to Iran. Eighteen points means under forty percent scoring across forty minutes — that loss belonged not to raiding but to the pressure that pushed Indian raiders onto do-or-die raids again and again.

The strongest domestic evidence is the final of 14 October 2026. Patna Pirates blew Gujarat Fortunegiants away 55-38 and everyone remembers Pardeep Narwal's 369 raid points. I remember the 55 — Gujarat conceded 55, which means their third tackle collapsed. In a match where a side gives up 55, the raider's record did not win it; the defensive structure lost it. The column I filed that night was 'defensive collapse, not a dynasty', and that is still my yardstick.

So when I watch a live match I count three things and stop counting the scoreboard. First, who won the do-or-die raids in the last ten minutes — that is the real scoreboard of a final. Second, the tackle efficiency of the cover pair: if two covers are not creating a single assist on the third raid, the raider is getting free points. Third, raider substitution — the team that pulls its main raider off before the third raid is avoiding do-or-die, which means it is playing to avoid defeat, not to win. A coach who does not keep his best raider on the mat for the third raid is brave in the data and a coward on the scoreboard.

Here is the weak link in my own argument, because a receipts-first record means the whole record, not the edited cut. The super tackle flips my arithmetic: if the defence has only three men, taking the do-or-die risk becomes rational, because failing costs three points — the net equation changes. The second problem is sample size: a team gets few do-or-die situations in a season, and one eliminator's result can be pure noise. In an eliminator last season I leaned on do-or-die net to call a team favourite and they went out conceding three points on a super tackle — that is my bruise and it is on my sheet. The third problem: the league does not publish do-or-die splits in one place, in one format; my ledger is private and its sampling is not disclosed.

That is why I am writing the bet down instead of commentating. Whenever India next meet Iran, my pre-registered call is this: the team that finishes above 55 percent do-or-die raid conversion wins the match — and if India win with a do-or-die conversion below 50, I will retire this formula in public. The open question is how long the league keeps that first number glued to its graphics while matches are decided by the second.