ChessBase Magazine #225: Prague 2026's 27 Miniatures and the Slow Arithmetic of Elo-to-Odds

After the Prague International Chess Festival 2026 I filled exactly one column in my notebook, and I have been filling it for years: short games. ChessBase Magazine has now promoted that column to a flagship: 27 miniatures, games decided inside thirty moves, presented as the most entertaining material in issue #225. Thirty years ago a miniature was filed under 'accident' and dismissed in a paragraph. Today it is curated. That is the real news of this issue — not any single rating graph, but the fact that entertainment is now being counted in moves, and the shorter the game, the higher its price. The day entertainment becomes measurable, someone builds a market around it.

The contents are dry and legible. Prague 2026 is annotated by the men who played it — Aravindh Chithambaram, Anish Giri, Ediz Gurel, David Navara and others. Ten opening articles offer repertoire ideas; the opening videos come from Werle, King and Ris. Delivery is either a direct download with the booklet as a PDF, or a printed booklet with a download key by post. The issue also ships as a 'ChessBase Book' for iPad, tablet and Mac. Reading that list, an old habit returns: I weigh the material before the subject. Four annotators, ten articles, three video presenters, twenty-seven miniatures. This is a training package, not a newspaper.

The background matters. Prague was played in February–March 2026, and its Masters has become the densest invitational in Central Europe — not only in strength but in structure. Three generations sit at the same table: players who appeared in the 1990s, the top of the 2000s, and teenagers born after 2026. Navara is to that city what the Tal Chess Club lineage is to Chennai — institutional memory. Gurel is the newest data point in the same table. Giri is the control group for consistency. And Chithambaram is the answer to a question I wrote into my ledger in 2026: can a player out of the Tamil Nadu pipeline cross 2700 and actually win an elite invitational?

My method is old and I do not hide it. In 2026, at thirty-seven, I was handed the residual book at a Mumbai odds supplier — the sports too thin for the main desk, chess among them. I pulled 1,400 classical games from 2026 to 2026 and fitted a draw-rate curve to Elo gaps. The result: fifty points of separation produced a 62% score expectation, not the 57% the desk assumed. I kept the long tail in a notebook nobody asked to see, and that notebook still says what it said then.

The curve is now twenty-two years old, and I attach a warning every time I cite it: Elo is a relative, zero-sum, slow-decaying number; it is not an absolute measure of strength. Pool effects, rating inflation and inactivity decay all sit inside it. When I say the Prague 2026 field averaged around 2680, I am not claiming every man in it was as strong as a 2680 of the 1990s. I am claiming only that the gaps between them that week were the gaps my table describes. Those gaps are the thing the market misprices.

Aravindh's story earns its place because it sits in my ledger as three separate entries. Around 2026 he was near 2600 and described as a strong grandmaster, not an invitational threat. By 2026 his classical score had turned upward; by early 2026 he had crossed 2700, and he left Prague with the tournament. The data here is not that he won; the data is when the market agreed to price him. My notebook had his projected score against that field at 52–54%, a number that looks insufficient for victory and was therefore ignored. After the win, the same number was retroactively renamed obvious talent. The number had not changed. Its price had. I ran the residual again. It had not changed its mind.

Giri is a different kind of entry. His work is not the fastest climb but the smallest fall. The value of a player of his type over a season is set by how little he concedes to those rated below him. In my register, this is where elite chess and betting share a vocabulary: a draw against a weaker opponent is, on any list, the most expensive mistake available. Giri makes it rarely. That is why his annotations are worth reading — he writes about his losses with the same granularity as his wins, and few players do.

On Gurel I will stay cautious. The Turkish teenager has posted one of the fastest rating gains of the past two years, but his elite classical sample is under thirty games. Below thirty observations I call a result an anecdote, not evidence. That he annotates his own game in the issue is itself a signal — the editors are willing to put him in front of an audience — but the distance between signal and evidence is twelve to eighteen months, and it has to be walked.

Navara's value is of another order. Czech chess carries its institutional memory in him, and his contribution answers the question every home organiser asks: what should be expected of the host player? I have only one honest answer — the host's advantage shows up in the calendar, not the rating. He plays there every year; everyone else plays every other year. The sample is small, so I call it an observation, not a finding.

The opening section is a different business. Ten articles and three video series are training material, and the economics are plain: the half-life of a repertoire idea, on my estimate, is about eighteen months. After that, engines and opposing preparation neutralise it. That estimate rests on my own records, not a published study, so read it accordingly. Werle, King and Ris appearing on video confirms what I have written since 2026: reach moved to video, and what is missing is not audience but long-form written material. The 'ChessBase Book' format is an attempt to fill exactly that gap — the same content on three platforms.

Now the part where I have to argue against myself. The twenty-seven miniature selection was made for entertainment, not representativeness. It is a curated highlight reel, and a highlight reel does not measure the style of an era. In classical chess the base rate of games decided inside thirty moves is low; if it has risen — and I suspect it has slightly — the cause is preparation depth, not a faster rhythm. A miniature is usually not a display of brilliance but the product of one large error by one side. Editors can present that error beautifully. It remains an error.

The second trap is mistaking correlation for cause. The issue carries annotations from Aravindh, Giri, Gurel and Navara, so it is easy to conclude that Prague itself settled these players' rise or decline. It did not. Prague supplied a date; the cause sat in preparation rooms elsewhere. Chithambaram's jump came from widening his classical calendar, not from the Prague stage. Gurel's growth comes from his coach and from age-group structural work. And the market's recent enthusiasm — sponsors, streaming, a franchise league playing at home — is not a vindication of my 2026 model. That belongs to Gukesh, to streaming, and to a domestic circuit that finally ran a season. I keep the prediction apart from the cause, because merging the two is what destroys models.

Sixty years in, I still start with the denominator. This issue's denominator is twenty-seven; the numerator is four annotators and ten opening articles. A reader who asks how many of those twenty-seven were between top-ten players deserves the answer: very few, and that number is the most honest part of the whole 'Special'. Entertainment does not come from the top. It comes from the middle, where the rating gap is small and the cost of an error is large.

For what comes next I have three thresholds, and all three can be proven wrong. First: if the average birth year of the Prague 2026 Masters field moves forward by at least two years from 2026, that is a signal of the invitational circuit tilting toward teenagers. Second: if Chithambaram scores above 50% in three or more elite round-robins over the next twelve months, my old 52–54% entry must be revised — and that will be my loss. Third: if the miniature section in the next issue shrinks from twenty-seven to ten, the editors have conceded that this list was a market product, not a stylistic shift.

Between the market's price and the result sits a thin silence. This time it is eighteen months long, and I am recording it in a notebook nobody asked to see. Elo doesn't flatter anyone; the market does. Next time a teenager wins Prague, look first at what his projected score was before the win.