Twenty-Seven Miniatures From Prague 2026: The Entertainment Counts, the Repertoire Doesn't

The instinct is old and I no longer fight it: when a new ChessBase Magazine lands, I read the annotations last. First I count. The cover of CBM #225 is given to the Chess Festival Prague 2026, and inside sits a 'Special' — 27 miniatures, which the publisher's description calls “highly entertaining.” The adjective never worries me. What worries me is the number 27 itself, because a count is only honest once you know what it is counting. So I opened the file, rebuilt each game on the board, and put four columns beside it: moves played, which side collapsed, in which phase it collapsed, and whether the losing side had any visible plan one move before the collapse.

I have kept this ledger since 2026, one sheet per issue, because a miniature is the most honest document a tournament produces. There is no room in it for pacing yourself, no room for excuses, and the losing move carries a visible timestamp. I mapped the board first, then counted every move until the shape confessed. And the shape, most issues, is not the shape the blurb is selling. The blurb sells entertainment. The ledger says that most of these games are one-sided collapses, and that rereading one-sided collapses does not build skill; it builds a highlight file.

Let me do what a count-keeper should do first and mark the limits of his own count. From this issue, the measured figures are these: a 27-miniature special, 10 opening articles, three opening videos by Jan Werle, Daniel King and Robert Ris, and an annotator list that includes Aravindh Chithambaram, Anish Giri, Gurel and David Navara. What is modelled, not measured, is the internal taxonomy of the 27 — how many came from opening traps, how many from premature attacks, how many from time trouble, how many from plain structural pressure. I am filing that taxonomy as a hypothesis for the next issue, not as data. Silence has a tax, and so does false precision; I would rather pay the first.

What the issue actually carries

ChessBase Magazine's architecture has rested on one principle for three decades: the player is the annotator. Club players and professionals receive the same package — world-class players explaining their own games, opening specialists tracking current theory, trainers supplying tactics, strategy and endgame method. In #225 that architecture is pointed at Prague, and it arrives in two delivery shapes: a direct download with the booklet as a PDF, or a printed booklet plus a download key by post. Both include the “ChessBase Book” edition for iPad, tablet and Mac.

That format pairing is not a trivial detail to me. The same material reaches you in three physical states — paper you hold, a PDF you scroll, a video you watch — and each one runs at a different speed. Paper slows you down. Video carries you along. Over the past several years I have become convinced that the club player's real problem is not a shortage of material but the splitting of it across three formats. The issue opens in the booklet and closes after the video.

Prague as the centre of this special makes structural sense. The Czech tradition — Navara, club culture, a festival sitting near the turn of the year — represents the part of Europe where chess still runs like a public institution rather than an online event. The same hall holds young Indians, Dutch analysts and Czech veterans, and the annotator list of this issue is that mix written down.

What 27 means, and what it doesn't

There is no universal definition of a miniature. Some editors stop at 20 moves, some at 25, some at 30. When a magazine announces 27 miniatures, it is announcing an editorial boundary, not a mathematical fact. That is the first audit point. On my sheet I keep a second column beside the move count: the move of decision. My experience — and mark this as a model, not a measurement — is that inside a miniature the game is settled in a corridor between move 12 and move 18. Before that is preparation; after that is paperwork. If that holds, then the lesson of 27 miniatures is not 27 wins. It is 27 decisive moves, and they should be read side by side, out of game order.

A second distinction I have recorded for years: miniatures come in two species. The trap, where the winner created nothing and the loser simply erred. And the demonstration, where the winner forced every one of the opponent's 'mistakes' in advance. The trap is far more fun to read and nearly worthless to teach; the demonstration is dry and the most valuable thing in the file. I suspect the bright majority of the 27 belong to the first species, because the selection was made on entertainment grounds rather than training value. That is arithmetic, not accusation: a collection chosen for delight will represent delight.

Third, and least written about: the loser in a miniature is often not resourceless, only already beaten. One shock, then thirteen moves that look reasonable while all pointing the same way down. So for every miniature I check whether the losing side changed plan in its last ten moves. The ones that never change plan are the ones that fall. The grid, the timestamp and the move count come before the story.

The analyst's chair has moved east

The biggest structural news in this issue is not a result but an order of names. Aravindh Chithambaram annotating games from Prague is one of the more solid achievements of the last eight years of Indian chess, because becoming an annotator is harder than winning a title. Titles you take. The annotator's chair you are given, when an editor decides your eye sees something the others missed. The most valuable thing in an annotation is never the explanation of the win; it is an honest count of the opponent's alternatives.

Anish Giri carries a different meaning here. His elite reputation rests on the density of his draws, and his annotation runs on the same habit: he explains how equality was maintained — which move, at what moment, on whose instruction. For a club player that is directly useful, because most players cannot hold their patience in an equal position and break their own structure chasing imaginary activity. Giri's annotation is a cure for that failure, not a celebration of a win.

David Navara is this festival's local pillar, and his analysis usually shows the depth of preparation — which line appeared for the first time on which day, whether it was home work, and whether the opponent knew it. That is the kind of writing that pulls a magazine away from news and towards curriculum. Gurel is the least familiar name on the list, and for me the most interesting entry in the issue. When an unfamiliar name explains his own game, you see the thing big names often annotate out of sight: solid defence with no reputation attached. I read that chapter twice alongside the famous ones, because there is less shine in it and more structure.

Thirteen inputs, one repertoire

The second half of the issue exposes what I call the opening economy. Ten opening articles, three opening videos — Werle, King, Ris. The count is thirteen. Thirteen inputs is a lot for one week of study and far too many for one repertoire. Years ago I set myself a three-check rule and I apply it to every new theoretical idea. One: does the line survive to move 20, or die at 14. Two: to understand it, do I need to learn three new moves or thirty. Three: does the opponent have a clear equalising route that I must also know. Only after all three does a line enter my file.

Judged by that rule, perhaps three or four of the thirteen will survive. That is not a weakness of the issue; it is the nature of opening material. A magazine gives you the weekend's excitement. A repertoire requires Monday's routine. The two run at different speeds. And here the video-paper gap becomes decisive. King and Ris have an advantage print cannot reproduce: you watch them reject candidate moves, and the rejection is the actual lesson. But video has a tempo that overwhelms the slowness of a booklet. In my experience the club player watches the video and never opens the booklet; the professional opens the booklet and uses the video only for the lines where his understanding snags.

The entertainment ledger is not the training ledger

The most counter-intuitive thing about this issue is that its most attractive section is probably its least valuable for training. The reason is selection bias. A game becomes a miniature when one side collapses early. A collection therefore shows you the winner's moves and never the loser's thirteen perfectly reasonable moves that made the collapse possible. A player who memorises 27 miniatures arrives at a tournament setting traps until move 12 and standing exposed at move 20, because he studied the attack and not the defence. I use the collection exactly one way: from the loser's side, I write three moves that should have been played and were not. The rest is entertainment, and I am not stingy about calling entertainment entertainment. The danger is only that someone mistakes the entertainment ledger for the training ledger.

The second blind spot belongs to the annotator, and it is a limit of the genre rather than of the medium. When a player explains his own win, you are reading a ledger written by one party to the transaction. He will tidy his own moves, under-report his risks, and be less than generous in counting the opponent's chances. That is not dishonesty; it is human bookkeeping. The researcher's third option is the engine, which counts both sides and explains neither. So I keep all three together: the player's feeling, the engine's number, the scoresheet's structure. Together they make a complete ledger. Alone, none of them does.

The third blind spot no editor asks about: time. The catalogue says how much material there is; it never says how much time the reader has. A club player has forty minutes a week. Booklet, PDF, three videos and ten articles do not fit into forty minutes. What fits is the brightest object in the box — the 27 miniatures. Editorial selection plus the reader's time budget closes into a perfect circle.

Until the next issue

I am not letting this issue go quietly. By 17 January 2026 I will have written down three things, publicly. First, I will replay three of the 27 on my own board and record after each whether they still hold past move 12. Second, whichever of the ten opening articles pass my three checks go into a single file, and that file will also record how many lines I dropped. Third, when CBM #226 arrives I will publish an audit note on this issue, whether my forecast was right or wrong.

My real question here is not editorial. It is structural. If 27 miniatures leave you with only three moves worth repeating in your own games, what were the other twenty-four — training, or merely attraction? And the next time a magazine promises 27 entertaining games, will we count the moves first, or applaud first?