Shami's 131 in 35: A Selected Sample, Not a Career Ledger

"131 wickets in 35 Tests, average under 25."

That was the first line of the ICC report on Mohammed Shami's Test retirement — no date, no injury history, no home-away split. I copied it onto paper and separated three things. First: 131 divided by 35 is 3.74 wickets per Test. Second: 131 is not a career total; it is a selected slice of 35 Tests across three countries — India, South Africa and the West Indies. Third: the full ledger of the career being declared closed does not exist in one public place.

This is not a tribute. It is an audit of the account on which the next two years of India's fast bowling will be narrated.

Context: how I read numbers

In October 2026 I talked my way onto a Delhi digital desk's Ranji rota by promising to log every ball of Delhi's home season at the Feroz Shah Kotla by hand. Five matches, 1,842 deliveries — batter, bowler, line, length, shot type, false shot. From that sheet came a false-shot rate of 21.8% for Navdeep Saini against right-handers. Nobody had asked for it. Delhi won the Ranji Trophy that season, and my sheet was the only record in the office of how the seamers had actually bowled. I stopped writing "bowled well" the day I started writing "induced 14 false shots in 42 balls." I kept the ledger by hand until the numbers started talking back.

Then June 2026 arrived. My desk handed me the Russia World Cup data blog; I watched all 64 matches with a notepad and published a daily xG table. My model had France at 10.4 tournament xG against 14 goals scored. Finishing is noise, chance creation is signal. In September, during the Asia Cup in Dubai, I applied the same logic to cricket. On 25 September Afghanistan tied India, and I realised my model had no term for pressure. Since then every number I publish carries the words "against" — or it does not get published at all.

Shami's file gets tested on exactly that rule. 131 wickets against what? An average under 25 against which era of fast bowlers? And most importantly — which 35 Tests? The report never asks.

Football, cricket, esports — the same ledger, different ink.

Core: what 3.74 wickets a Test actually means

Start with the dumbest available arithmetic. 131 wickets, 35 Tests, 3.74 per match. If the average is under 25, total runs conceded across those 131 wickets must sit below 3,275 — roughly 93 to 94 runs per Test. This is not a volatile profile. It is a flat, disciplined one. Which is the first contradiction: the bowler the country remembers as a spell-bowler produces a career sample that is steady rather than spiky, while the memories are all spikes.

Now the comparison. Zaheer Khan took 311 Test wickets at roughly 32.9. Ishant Sharma took 311 at roughly 32.1. James Anderson took 704 at roughly 26.4. These are external benchmarks, they are not in the ICC report, and they should be re-verified against ESPNcricinfo or official scorecards. But the direction is clear: a sub-25 average belongs in Anderson territory, roughly seven runs per wicket below the Indian fast-bowling norm. Seven runs a wicket is not a small difference. It is a category change.

And that is exactly where my first suspicion begins, because the 35-Test sample is drawn from three countries. India, South Africa, the West Indies. Which means England, Australia and New Zealand are outside the count. Shami's 2026 England tour, where the ball seamed less and he averaged above 30, is not in that window. If that tour is missing, the story's opening line is not a description of a career — it is a selection from one.

The number can be correct and still misleading. A number is only as honest as the person who typed it, and only as innocent as the frame around its sample.

Core: Johannesburg 2026, five wickets and a small target

Now the match that will appear in every tribute. Johannesburg, 2026, second innings, 5/28.

We remember that spell by its wickets. The correct way is to remember it by the match state. India had given South Africa a modest target — around 240 — on a Wanderers pitch where the ball sometimes stopped and sometimes climbed to chest height. Defending 240 on that surface is technically harder than defending 400, because you have no runs; you only have overs and time.

In such a defence the bowler's job is not to bowl a good ball repeatedly. It is to take wickets in the specific window when the batter is trying to settle. Read 5/28 that way — situation first, milestone as residue. I was watching that match with a scorecard on the laptop and a notepad in hand, and my notepad that evening held over numbers, not a wicket list. That habit has stayed with me for every fast bowler I have assessed since.

India won that Test. But a caution belongs here: winning and causing the win are different claims. The report tells us India won both highlighted Tests after Shami's major contributions. It does not tell us how much of the win came from his bowling and how much from everything else.

Core: Centurion and the repeatability test

Next tour, Centurion, first Test, eight wickets.

This is the more important of the two, because a single match is an event and two matches across different tours, different squads and different conditions are the first indication of a pattern. The biggest lie in cricket analysis is treating two data points as proof when they are only two data points. They suggest; they do not confirm — and only if we know the pitch, the toss and the age of the ball. We do not.

What we do have is real. Johannesburg and Centurion are both high-pace, high-bounce venues, and at both Shami's output was match-shifting. For an Indian fast bowler, that is not unheard of, but the repeatability arriving away from home conditions is not nothing.

My noise file — crowd, dew, travel, toss — is empty on this case because the source supplies none of those variables. When a number surprises me, my instinct is to hunt the variable I forgot rather than the story I want, and here I have to assume that the bounce at both venues suited Shami's seam position. How much of that was skill cannot be settled without innings-level data.

And that gap is the real one: there is no public index of which fast bowler generated pressure in which over-window. Seam position, ball-by-ball line, the reverse-swing window — the broadcast truck knows this; the scorecard does not.

Core: what is actually lost

The easy answer is: a bowler with 131 wickets. But 131 never goes into a team's account. Overs do. Shami's utility was in overs, not wickets.

India's fast-bowling success of recent years has been built on splitting three jobs: the first new-ball spell, the middle overs that let spin breathe, and the old-ball window that has to manufacture reverse. A large share of the first and third sat with Shami. Remove one bowler from that window and the damage does not show in a wicket count — it shows in another bowler's over-load.

This is where the blind spot sits. We read retirement as a wicket swap when it is a workload swap. India had and has Bumrah, Siraj, Umesh and Ishant — but a queue is not a division of labour. A fast bowler's greatest contribution is often the burden he takes off a colleague, and that burden has no public metric.

Which ties into congestion. Two matches in a week, three formats in a month — on this calendar, injury is not an accident but a structural output. No medical team can fix load from outside, because the medical team's job is recovery and the calendar's job is load.

Contrarian: "India's pace attack is weaker now" is an unfinished claim

My brand is counter-intuition, and brands calcify into tics, so I stop here. The easy story is that Shami's exit weakens India's overseas pace attack. I am filing that as unresolved, not confirmed.

Three things are missing. First, the away Test record of whoever replaces him — absent from the report. Second, the calendar of India's coming overseas tours, so we know where the exam is. Third, the pitch profile of those tours. Without them, "weaker" is a feeling, not an assessment.

Second, and less said: retirement is not evidence of decline. The report gives no recent form, no injury history, no age, no selection context, no stated reason. We assume players leave when they fall. Sometimes true, often not. Until the board speaks, the inference should wait.

Third — and this is the most useful finding — the weakest evidence in the file is not the retirement but the frame of the coverage itself. What the truck knows about Indian fast bowling never reaches the commentary box in that form. What reaches it is the top of the scorecard. So retirement coverage is dominated by exactly what the truck had least of: statement.

And I will publish the boring finding, because the boring finding is true: Shami was among India's most effective fast bowlers abroad, and South Africa suited him. 5/28 and the eight wicket haul at Centurion support that. I will not invert a correct consensus. I am only questioning its scope — it is a truth about a sample, not a verdict on a career.

Takeaway

Over the next 18 months, do not compare records. Watch who bowls India's fourth and fifth spells on the next overseas tour, and what the run rate does in that window.

Three triggers are logged. One: if first-spell economy on an away series goes above five and the third and fourth spells hold there too, the story becomes about the division of overs, not wickets. Two: if senior fast bowlers become unavailable in the same block, the effect will exceed Shami's individual share — no model answers that; only the calendar does. Three: if Shami joins the squad abroad in any capacity, his real transfer will not be wickets but a reading of the old-ball window that no public dataset indexes.

I did not find the pattern; it waited for me to finish counting. The counting is not finished. The 131-wicket account is closed, but its other rows — home Tests, the England tour, the injury months — are still not on any public table. Until they are, we are writing a full verdict on half a ledger, and that is not done to any bowler — Shami or his successor.