The Immutable Ledger: Auditing Powerplay Data at the T20 World Cup

The Immutable Ledger: Auditing Powerplay Data at the T20 World Cup

Hook: the number nobody could verify

Six overs gone, the graphic flared gold — a powerplay strike rate of 214. From the next desk someone said, "That's the best opener in the tournament." I turned and asked three questions. One: how many legal deliveries were actually in the over block? Two: what were the release speeds and angles of the two main bowlers, and how much bounce did the surface offer? Three, and the one that matters most: who recorded the number — the on-site scorer, the broadcast data operator, or a third-party feed — and when?

Not one of the three answers existed anywhere in the room. The number had already travelled. My objection was never to the number; it was to its incomplete register. The 2026 T20 World Cup is running across India and Sri Lanka — opened February 7, final on March 8 at the Narendra Modi Stadium in Ahmedabad, twenty teams — and every day it manufactures a dozen such numbers, and every day nobody is answerable for them.

I work with ledgers. In cricket the ledger is the ball-by-ball record: which over, which bowler, which batter, which line, what release speed, which field, and who typed it in. Whether it is an xG-PPDA matrix in football or a powerplay strike rate in cricket, every event is a block and the innings is a chain. The founding idea of a blockchain is far older than the technology: each entry chained to the last, so that changing one entry puts the whole chain under suspicion. I am only asking cricket to restore that old discipline.

Context: four football lessons, one cricket rule

In late 2026 I was building xG-PPDA matrices for Premier League midfielders at a Manchester agency. Ross Barkley sat at 0.12 xG per 90 and 8.7 pressures per 90, and I advised against a £15m bid. The agency went ahead. Barkley made two league starts in his first half-season. From that day every memo I wrote began with data provenance and error bars.

The secondment to a broadcast data desk at Russia 2026 came because of that memo. In the final I tracked N'Golo Kanté's 55th-minute substitution and measured France's block, not individuals — Luka Modrić's 694 minutes, 2.3 key passes per 90, 88 percent pass completion, 10.2 km covered per match. France won on collective defensive structure. In 2026, when the Bundesliga returned to empty stadiums, home win percentage fell from 43.3 to 33.3 over the first five rounds — 45 matches. Clubs asked me to model crowd effects. I refused. The empty stadiums taught me one lesson: bring more sample or bring silence.

In 2026, after Qatar, the Enzo Fernández file landed on my desk. Seven matches, 8.2 progressive passes per 90, 2.8 tackles per 90 — and a sample of seven. I recommended against triggering the full £106.8m release clause, proposing add-ons instead, and wrote separately that tournament samples and club-form samples must never be pooled. The club ignored me. That is the football half of this essay. Now back to cricket.

In T20, the powerplay is the first six overs, thirty balls — a quarter of the deliveries, thirty percent of the overs, with fielding restrictions and only two fielders outside the circle. A disproportionate share of boundaries is manufactured there. In the 2026 edition the group stage alone produces dozens of powerplays, and yet an individual batter faces perhaps six to ten of those overs himself — 36 to 60 balls. That is my binding numerical constraint, and everything below follows from it.

Core: three files, three findings

File one is the strike rate. A side can be 62 without loss after six overs, or 62 for three. Same run rate, entirely different match state. Likewise 55 for one and 48 for two look comparable, when in fact both top orders have been broken. A powerplay run rate can never stand alone as the basis of an opinion; it must carry the wicket column, the quality of the opposing attack, and the target beside it. Strip those three away and 214 is decoration. I re-ran the old file against the 2026 edition: Arshdeep Singh and Fazalhaq Farooqi jointly topped the wicket chart with 17 each, and Jasprit Bumrah took 15 wickets at 4.17 an over to win Player of the Tournament. Bumrah's number survives an audit because its denominator is public — eight matches, roughly thirty overs, every ball on video.

File two is matchup granularity. A spinner's overall economy is one figure, but it is the average of four: against right-hand top order, against left-handers, in the powerplay, in the middle overs. Wanindu Hasaranga's googly, Kuldeep Yadav's left-arm wrist spin, Rishad Hossain's flat quick slider — you can place all three in one column, but that column cannot drive a tactical decision. To make a claim about one batter against one bowler I need 600 balls of face-off. Six hundred balls is roughly ten combined domestic and franchise events. No single World Cup can generate it.

File three is the sample threshold, and this is my least popular position. In football I would not write a talent claim under 900 minutes. Cricket's equivalents are: 300 balls for a batter-level verdict, 600 balls for a bowler-versus-hand matchup, and at least ten matches across varied opposition and surfaces before commenting on a team's powerplay strategy. In a twenty-team World Cup most batters face under 120 balls. This tournament cannot deliver individual verdicts. It can only raise flags. A flag is not a ruling, and anyone who confuses the two is not using data — he is staging it.

That is why every calculation of mine stays open. Which delivery, which provider, what timestamp. Before you trust the xG or the PPDA, ask who recorded the input and when. The rule is identical in cricket: if the camera coordinates, the release point and the bat contact point cannot be traced, the strike rate is a rumour dressed in a tall inferential structure.

Contrarian: winning the powerplay is not winning the match

Start with the easy correlation, because that is where everyone stops. Teams that score more in the powerplay generally lead the match. The relationship is real; the causation is not. Powerplay success is largely the product of three external factors: the toss, the pitch, and the quality of the opposing attack. Openers in form produce good powerplays and good positions — meaning good teams win more powerplays, and teams that win more powerplays are good teams. The only exit from that circle is to measure and subtract the opposition attack's quality. Skip that step and we are selling team quality as individual skill.

The second counter-intuitive angle comes from the empty-stadium natural experiment. I refused to publish the 2026 Bundesliga drop from 43.3 to 33.3 percent home wins because 45 matches is a signal, not a test. Cricket made the same mistake behind closed doors: two rounds of data produced loud claims about umpiring and reviews. I said then what I say now — bring more sample or bring silence. Home advantage is a four-way tangle of team quality, travel fatigue, pitch age and crowd presence, and no model separates them without error.

The third angle returns to the recorder. Control percentage, false shot, pressure points — these are not physical quantities but human coding decisions. Two agencies can log different numbers for the same delivery, because one weighs bat edge with seam position and the other only the outer contact point. That is why I want approved control-data distribution in the press box, with bowler, over number and timestamp attached.

One more habit of long standing. Boards and clubs write injury ledgers too, and those ledgers are usually written at the PR desk, not in the physio room. The "minor niggle" announced before a semi-final is curated information. I believe an injury report only when the training footage confirms it. And the most convenient metric for hiding a fitness deficit is the effort family: distance covered, sprints completed. Pointless running also produces pretty numbers. Cricket's equivalent is intent and dot-ball pressure — a batter who cannot rotate strike against good deliveries still collects a high intent score, and nobody asks how good the ball was.

Takeaway: what I will watch in the next round

For the remaining group matches I am pre-registering three things, so that nobody later claims the forecast was written after the fact. One: I will only read powerplay strike-rate tables in wicket-weighted form, with strike rate multiplied by the cost of losses. Two: I will look at a spinner with sub-seven economy only if the split against left-handers is published. Three: before any knockout report on home advantage or umpire effects, I will demand sample size and timestamp — otherwise my answer is the same as six years ago: bring more sample or bring silence.

Whoever lifts the trophy in Ahmedabad on March 8, my ledger stays open. The highlight reel of the last over will be gone in eleven seconds; the ball-by-ball chain will sit there to be read for years. At sixty-three, I still trust the ledger more than the highlight reel. The question is not for today but for three months and five years from now: the number that shaped your opinion this evening — did you ever open the block behind it?

— The Data Monk