Dhananjaya Steps Down: What 18 Tests of Data Never Told the Scoreboard

Sylhet, March 2026. Dhananjaya de Silva is captaining Sri Lanka in only his second Test in charge. He makes 102 in the first innings and 108 in the second — twin centuries in a single Test — as Sri Lanka win by 328 runs. Nobody forgets a start like that. But a scorecard never tells you one thing: that dazzling beginning was far tidier than the chapter that followed.

In 2026, at 35, he stepped down as Test captain 'with immediate effect.' He conveyed the decision in writing to Sri Lanka Cricket's (SLC) selection committee and made clear he remains available as a player. Years of watching the game teach one lesson — the scoreline never panics; fans do. But I do not stop at the scoreline. When I built my xG model in 2026, the entire premise was this: what the scoreline refuses to say needs a separate structure to be heard.

Test captaincy is a distinct profession. In a five-day match a draw is a live outcome, so leadership is not just the toss and the declaration — it is bowling workload, session-by-session planning, and timing. When Dimuth Karunaratne retired from Tests in 2026, Dhananjaya stepped into that vacancy, which makes this a chain of succession rather than a sudden coup. In 2026 he said the mantle should pass to the next generation. SLC CEO Ashley de Silva's statement was warm and forward-looking — no pressure, no bitterness, no whiff of a sacking. In South Asian cricket administration, a clean handover is rare, and that is the most significant signal here.

A word on data hygiene first. Of the twelve information points I am working from, eleven carry no named source — age, match counts, win-loss record, series results and batting average are all unattributed. I am treating them as data pending verification against ESPNcricinfo or Cricbuzz records. My habit is not to make a claim until the model is audited.

Sri Lanka is currently a mid-tier Test side: strong at home, historically vulnerable away. That structure matters when reading Dhananjaya's ledger. Seven wins, eight losses, three draws across 18 Tests — a win/loss ratio of 0.875, roughly par for a mid-tier side, neither a collapse nor a revelation. And only three draws in 18 Tests (about 17%) suggests a captain who pushed for results rather than protecting a draw. That draw rate is low for Test cricket, and it is a subtle fingerprint of declaration timing and pitch preparation.

The geography of the series wins matters more. Of four named series victories, four came at home — Afghanistan (home), Bangladesh (home), New Zealand (home) — with the only overseas success coming in Bangladesh. The record is home-weighted from the top, and that is the biggest methodological signal. At the 2026 World Cup in Russia I tracked PPDA for every France match and saw how deliberate a deep block could be. The same logic applies in cricket: spin-friendly home pitches, and a struggle against seam, swing and bounce abroad. The eight defeats almost certainly cluster on away tours, because there is no evidence of an overseas series win beyond Bangladesh.

Now the batter. His batting average as captain is 36.45 — below the 40-plus expected of a specialist top-order Test batter, but acceptable for an all-rounder. Three centuries in 18 Tests — the Sylhet twin tons plus two more — is a healthy conversion rate. The issue is distribution: one dream start, then a mixed run. The captaincy burden suppressing a batting average is well documented, so separating individual skill from role pressure in that 36.45 is not easy. This is where I stay cautious: both the career baseline and home/away splits are missing, and without them, neither an 'underperformance' nor an 'overperformance' verdict is defensible.

At 35, this is a rational late-career de-risking. Batters typically peak between 27 and 33; 35 is the start of the descending slope. Stepping down while remaining available as a player is the classic play of dropping the leadership load to extend the batting years. History suggests batter-captains often lift their output in the 12–18 months after relinquishing the armband. But treating that uptick as certain contradicts my method; it is a testable hypothesis, not a proven fact.

Here is the counter-intuitive part. What everyone is reading as a relief is actually a clean transition, not a crisis resolved. The link between a captaincy average and the burden of captaincy is correlation, not causation. How much of the suppressed average is role and how much is age cannot be separated from this dataset, especially with no home/away split. The second trap is Sylhet: reading those twin centuries as proof of captaincy genius is wrong — it is one match inside an 18-Test, mid-tier tenure, and it is the media's prime hype anchor. Third, the 'next generation' framing is double-edged: if SLC installs a genuine young face it holds; if it installs a 30-plus stopgap veteran, the whole narrative rings hollow. Most important, Sri Lanka's overseas weakness is structural; a change of captain will not fix it, because it is a question of seam-bowling resources and tour planning, not personal leadership. In my 2026 empty-stadium study I learned to treat context as a variable, not noise — and here, the tour context is the real variable.

So what comes next? The identity and age of the successor will decide whether the story holds — if SLC names a young captain within weeks, the 'planned handover' reading survives; a long delay or a stopgap move makes the succession vacuum the main risk. Dhananjaya's post-captaincy batting average becomes an indirect witness — up and it is read as the burden lifted, down and it is read as age catching up. The real test, though, comes on the next overseas tour, where a home century will be no alibi. The question is not simple — does changing the leader change the team?