43% Dot Balls and a 136.4 Strike Rate: Where Pybus's 'Benchmarking' Argument Holds, and Where It Breaks

The Sharjah ground was nearly empty on that October evening. Twenty thousand seats, perhaps two thousand occupied — mostly migrant workers, a few schoolchildren, and Afghanistan training under the floodlights. I was not there. I was at my desk in Delhi with an interview transcript open in front of me, in which head coach Jonathan Pybus had offered two numbers: a 43 per cent dot-ball rate, and a strike rate of 136.4. “Over the past year,” he said. That was all. Which year, how many matches, on what surfaces, against what quality of bowling, in which phase — no sample, no method, no comparison.

I keep a column for luck. It is the most honest column in the file. In recent years I have had to add another: the column for unsourced numbers. The 43 and the 136.4 are now logged there. The problem is not that these figures are wrong. The problem is that they are migrating from a quotation into a fact, and I know how that journey ends. In Karol Bagh, the par-score was a rumor repeated until it became a number.

Context: three defeats, one captaincy handover, and a series that begins tomorrow

Read Afghanistan's T20I record off a scorecard and you see a team that flickers and then contracts in big tournaments. Read it as a system and a different picture appears: one of the fastest-rising spin-first bowling units in the world, a batting order whose depth looks adequate on paper and uneven on grass, and above both a leadership arrangement that has, over the past few months, moved from Rashid Khan to Ibrahim Zadran. That third element is the real story, and it is the element the interview spends the most space on — far more than the technical cricket.

The 0-3 sweep by India is done. Zimbabwe arrive in Sharjah from 27 October — the same ground where 160 is a first-innings par and where evening dew makes 165 a chaseable target. Pybus has two jobs right now: solve the dot-ball problem, and demonstrate that the captaincy change has not split the room. In this interview the second job is done far more loudly than the first, which itself tells you where the pressure sits.

Core: 43 and 136.4 are two numbers that talk to each other, if you join them

Start with arithmetic, because arithmetic is cheaper than rumor. If a side's dot-ball rate is 43 per cent and its strike rate is 136.4, then across 120 balls that produces roughly 51.6 dot balls, 68.4 scoring balls, and about 164 runs. Now look at the part nobody looks at — the run rate on scoring balls. 164 runs divided by 68.4 balls gives 2.39 runs per scoring ball, a scoring-ball strike rate of 239.

That is a startling figure. In my file, top-ten men's T20I sides over recent seasons sit at a dot-ball rate of 36 to 41 per cent, a strike rate of 142 to 150, and a scoring-ball run rate between 225 and 240. Which means: if both Afghanistan numbers come from the same sample, their hitting on scoring balls is at the top of the benchmark range — the power is not the problem. The problem sits entirely at the door where the ball goes dot.

That yields the central claim of this piece: Afghanistan's problem is not a strike-rate problem, it is a distribution problem — six or nothing. In coaching language this gets called a lack of intent. In data language it is a failure of rotation. And the fix is not more sixes, because the sixes are already at benchmark. The fix is converting ten of those 51 dots into singles. Eleven runs, about nine points of strike rate, with no additional risk taken. It is the cheapest trade in T20 and the least executed.

There is an honesty condition, and I will not hide it: the entire calculation rests on the assumption that both figures come from the same sample, the same format and the same quality of opposition. The interview does not satisfy that condition. If the 136.4 includes West Indies and Uganda while the 43 includes India and Australia, my conclusion inverts. I hold this number inside a band of plus or minus six points, and the width of that band is the truth.

Where the dots actually happen — a question the interview never asked

Dot balls are not distributed evenly in T20. In the powerplay they come mostly from movement on the new ball and are tolerable, because boundaries come with them. The real damage happens in the middle overs, 7 to 15, where spinners turn the ball, batters consume deliveries trying to get set, and the scoring rate slides from 8.5 an over to 7.2. In my experience, sides that eat dots at above 40 per cent through the middle almost always finish under 155 — even at a small ground like Sharjah, where square boundaries run 60 to 65 metres and 155 is not a defendable score.

There is a further layer absent from Pybus's accounting: in Sharjah, dew settles the result of most night matches regardless of who bats better. In 2026, when the IPL moved to the UAE and the stands were empty, I built a private database of four thousand T20 matches. In that regression, second-innings variance in Dubai and Abu Dhabi was explained more by toss and evening dew than by team strength. Twenty-two thousand empty seats, and home advantage quietly rounded down to zero. Sharjah belongs to the same cluster. Bat first there and your 43 per cent dot problem becomes twice as expensive, because you need 175, not 160.

Captaincy and batting: the question the interview cannot answer, only the sample size can

Ibrahim Zadran's defence occupies the longest stretch of the interview. Pybus argued that the captaincy is costing him runs at the crease, and that this is a baptism of fire. I want to test that claim with arithmetic rather than sentiment.

Suppose Ibrahim has played 20 to 25 innings as captain. Per-innings variance for a top-order batter is large — a standard deviation of 25 to 30 runs is normal — so across 22 innings the standard error of the average sits at roughly 5.5 to 6.5 runs. The implication is direct: if his average has fallen eight runs, the fall is inside the noise. Proving a captaincy effect would need at least 40 to 50 innings, and even then you would have to control for opposition quality, batting position and toss.

I am not claiming the burden is imaginary. I am claiming that the evidence available cannot carry the claim, and that Pybus has substituted intangible leadership qualities for measurable output in his own defence. That is a protective move, and a comprehensible one. But a data desk logs it the way it logs anyone else's: as a hypothesis, not a finding.

And there is a crack in the piece that I found genuinely interesting. The same report has its author writing that Rahmat Shah impressed while Ibrahim “did not impress with his batting returns.” The reporter's own evaluative line runs directly against the coach's defence. In an access-driven interview that is rare — and perhaps accidental — but it puts in front of the reader the fact that Pybus's defence rests on things that cannot be measured rather than things that can.

Captaincy is a pricing problem, not a character question

I have always read captaincy as a pricing problem. You have limited overs, limited matches, and one player with limited mental bandwidth. If the captaincy consumes 20 per cent of that bandwidth and returns only toss calls and field settings, you must ask whether that 20 per cent is worth more deployed elsewhere. Rashid Khan carries experience from the world's best T20 league, output across three formats, and a career that is already a large part of the team's brand. Ibrahim carries youth, format expansion and institutional investment. These are two different assets and pricing them on the same scale is a category error.

That is why the passage dismissing speculation about Rashid's unhappiness matters more to me than the technical section. Pybus called him a model professional who holds nothing back, and stressed the closeness of the two men's relationship. That density of superlatives does one job in any press conference: reputational repair. It is also necessary, because a captaincy handover that looks natural from outside depends on the consent of the team's biggest name inside.

Contrarian: “benchmarking” is defeat management, and it is also necessary

Now the disagreement, with the caveat that disagreement is not the same as dismissing the room.

Pybus called the 0-3 defeat to India a benchmarking exercise — “a better gift I could not have received as a coach.” In cricket language that sentence is elegant. It is also dangerous, because benchmarking means you measured something. And if you measured, you owe the listener the measure. The interview offers nothing beyond “strengths and weaknesses.” I have seen this shape before. I went to Kazan to watch a model fail in public. It was instructive — but only because I wrote the failure down and published it. A model that does not publish its misses becomes a rumor within a season.

Here, though, I have to price my own bias. The room is treating benchmarking as a strategy, and the room may be seeing something the data misses: an emerging side does not get three matches against India every year. Losing a series is a cost; playing it is an investment. If the return arrives inside two years, today's defeat was a price, not an accident. That argument is valid. It is also true that the word benchmarking gets used most when the underlying gap is frightening — and losing 3-0 to India is frightening. I hold both: the message can be strategic and the gap can still be real.

My second objection is structural. Almost every piece of information in the interview arrives through the coach's mouth. Thirteen direct quotes, no Zimbabwe perspective, no independent analyst, no direct comment from Rashid Khan. And the unhappiness speculation is a construction raised only to be extinguished in the same paragraph. That is an old device: introduce the rumor so the subject can deny it, leaving the reader with the denial and not the question.

Third, and this goes to the top of my column: do not read the correlation between captaincy and batting form as causation. It is equally possible that Ibrahim's form dipped first and the captaincy followed, or that a third factor produced both — a run of turf surfaces, a shift in batting position. The interview raises the question and answers it only through the coach's denial. I do not have the answer. I only have the observation that the answer has not yet been written.

Takeaway: three things in the Zimbabwe series that will settle this argument

I will not forecast, because since 2026 I have stopped publishing single-number predictions. I will write down the conditions the thesis rests on and the conditions that would kill it.

First: does Afghanistan's dot-ball rate drop below 40 per cent on Sharjah nights — and if it does, does their scoring-ball run rate fall below 2.39? If yes, the problem is risk management rather than rotation, and this analysis needs rewriting from the top.

Second: how many balls does Ibrahim consume in the middle overs? If he is striking below 100 across 15 to 20 balls, the captaincy-burden debate is moot — the issue is role, not office.

Third: the toss. In my file, the second-innings win share in Sharjah sits around 60 to 65 per cent, tied more to dew than to team strength. If Afghanistan bat first three times and lose three times, the 43 per cent argument is not worth a close read — the result was priced before a ball was bowled.

A monk does not argue with the market. He re-runs the regression. I will do exactly that on the evening of 27 October, and if my band is wrong, the error goes into the same column as luck.