77 Teams, Five Continents, Zero PGNs: Auditing the Prison Chess Championship's Data Blackout

May 13, 2026. It was not a date on chess's calendar. That was the day the first Continental Chess Championships for Prisoners began, running through May 23. Seventy-seven teams. Five continents. Three sections — Open, Women's and Youth.

Then I did what I do after every major event: I asked for the scoresheets. Board-by-board results, time control, team line-ups, one player's name, the prize fund, the chief arbiter.

Nothing came back.

A 77-team tournament, and not a single PGN in the public domain. That is not a complaint. That is a fact, and it is the most interesting part of this story.

In 2026 I started testing video hot takes from ChessBase India's makeshift studio at the National Premier in Delhi — I have tested hot takes on a chess stream in 2026, and the numbers never lied again. Twenty-three minutes of dead air, a three-minute clip, 40,000 views. The lesson was singular: an event with no metric has no evaluation. You can only be sentimental about it.

That is exactly what has happened to the first edition of prison chess's continental championship.

Context: one initiative, two names, three sections

Chess for Freedom is FIDE's programme that takes chess inside correctional facilities, working with prison administrations, national federations and local organisations. The logic is straightforward: chess is cheap, teachable, and demands the traits — patience, calculation, impulse control — that prison systems repeatedly link to recidivism.

FIDE called the May 13–23, 2026 event a milestone under that initiative. The headline says Intercontinental Online. The summary says First Continental. That is not a trivial naming slip. One name points to an online platform, the other to a continental championship — and when organisers cannot agree on what the event was, I have no reason to believe they can agree on how well it went.

The second thing that jumps out: 77 teams, five continents, three sections. Three sections means structure. Open, plus Women's and Youth. That is not decoration, it is architecture. In any other 77-team tournament, that figure would tell me how many women's teams entered, how many teenagers played, how participation split across sections.

Here, I do not even know that.

Core: four instruments, all four useless

I want to be precise, because this is not a hit piece. It is an audit.

I have four instruments for measuring any chess event: average field rating, opening variety and engine match rate, execution stability under the time control, and performance against a player's age curve. All four are useless here — no ratings recorded, no opening record, no declared time control, no age or federation data. I do not even know whether this was classical, rapid or blitz, and without the time control the first condition of technical evaluation collapses.

'High-level, organised tournaments' is the framing on record. That is advertising, not assessment. Seventy-seven teams and five continents establish scale, not strength. I am confident the field was wildly uneven in strength — that is not a criticism, it is the ordinary truth of international chess. But measuring it requires data, and the data does not exist.

Country names appear: Zimbabwe, El Salvador, Serbia, India. Their teams are said to have produced 'standout performances.' The source for that claim is not stated. Scores, opponents, board results, a single player's name? None. A country's name is not evidence of playing strength; it is a signal of institutional participation, and even that signal is unattributed here.

In any other sport this would be impossible. Think of any recent 77-team tournament — a cricket league, a continental kabaddi round. Scorecards, top scorers, best elevens: all public. The reason is not only transparency, it is marketing. You cannot sell what you have not measured.

Now the uncomfortable part. The available information mentions no webcast, no scoreboard, no arbiter list, no broadcast arrangement. And when questions were put, no accountable person came forward publicly.

In 2026 I sat in a half-empty hall at the inaugural Tata Steel Chess India in Kolkata. The broadcast treated that silence as reverence, when it was product failure. I wrote at the time: a city of 14 million and this event has no street poster — that is not a chess problem, it is a marketing crime. Kolkata did not shake hands, because nobody extended one.

The same motion is visible here, at larger scale. When organisers withhold the numbers, broadcasters cannot build a story, writers cannot file a headline, and the part of chess that actually deserved attention quietly drains away.

Return to the three sections. Open, Women's, Youth — this inclusion architecture is the most solid element of the whole event, and the least discussed. For India it gets more interesting: the country's chess base — academies, online rated events, state circuits — is dense enough that scaling a prison chess programme there should be comparatively easy. That is inference, not established fact. But if true, an Indian team being present in the first edition signals real pipeline strength — and we still do not know a single player's name on it. 'Youth' implies an age category, but eligibility rules were never published, so I cannot even say whether it meant juniors or young adults.

Contrarian: I may be the one who is wrong

Now I have to attack my own argument, because every hot take that opens with three points and never audits itself becomes propaganda.

Possibility one: the data was withheld deliberately, and perhaps that is the right call. Publishing prisoners' identities, status and case details is an ethical hazard. Prison security may bar releasing PGNs — board-level data for a team can indirectly reveal where a given player is held. If participant protection was the priority, then my demand for scoresheets is reckless. I accept that possibility, on one condition: that it be declared as policy, not practised as silence.

Possibility two, and the one that bothers me more: I am measuring with the wrong instruments. Elo, ACPL, engine match rate — those are not the metrics of an event whose purpose is not chess excellence but reconnection. The real measures for prison chess programmes are retention in the programme, the number who keep playing after a tournament, and change in recidivism. Chess media has no apparatus for any of the three. I am demanding PGNs from an event that should be judged by human outcomes.

But here the coin flips. If that is true, then FIDE has no proof of 'global impact' either. When organisers say 77 teams and five continents, they are claiming reach, not result. Reach and impact are different things. The same metric I am using to flay the event cannot be used to certify it. Both sides of the ledger are dark.

The real target of my criticism is not the missing data. It is that this event is making two incompatible claims at once — 'we stay silent to protect participants' and 'our impact is global.' Those cannot coexist. Either stay silent and claim nothing, or measure and publish.

Takeaway: one prediction, testable later

By the next edition, one of two things should appear. First, a minimum public dataset — team list, participation counts by section, time control, and board results under anonymised player codes. Second, a programme-outcome report recording at least how many participants kept playing chess after the tournament.

If neither is public by the third edition, 'milestone' is branding, not progress.

Prison chess may be the cheapest social intervention in the world. I am not arguing to shut it down — I am arguing to count it. Because what is not counted is the first line cut from a budget. I have watched three eras of sports media, and the whisperers lost every single time. Prison chess has to stop whispering.