A Move Without a Receipt: Erigaisi–Laohawirapap and the Accounting of the 5.e5 Recommendation

I stopped at one place while reading installment number 740 of the series. The piece opened with the modern branch of the Italian Game — 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.d4 exd4 5.e5 — and stated in plain language that in this position it is difficult for Black to equalize. The moment that claim appeared, my hand started hunting for the number: on how many games, in which database, at what time control, on which engine version. None of them were there. The file that stays open on my desk has three columns: sample size, source, publication date. For that installment all three cells stayed empty. One move, one game, two names — Arjun Erigaisi and Laohawirapap — and a recommendation standing on top of them. As a story it is clean. As an accounting entry it is incomplete. The branch is not new, and that is the first relevant fact. After 4.d4, if Black takes exd4, White pushes the central pawn with 5.e5. Black's knight on f6 is forced away, and White accepts tension in his pawn structure in exchange — the risk of a backward or isolated d-pawn for space and initiative. That is a trade of structure, not a profit in strength. From years of watching games from the side of the board, I can say this kind of position is easy for White to play and demanding for Black, who has to think through every single move. There is a second fact that is the real selling point of the recommendation: the same position can be reached by another route — the Scotch move order, 3.d4 exd4 4.Bc4. A player who already plays the Scotch can enter this branch without learning anything new. The article states clearly that Black's three principal replies are covered — 5...d5, 5...Ng4 and 5...Ne4 — and claims full coverage of all three. Alongside them appears the name of Evgeny Sveshnikov, a Russian grandmaster who has played the line himself. Authority-by-association is an old device in opening literature, and this is where the gap opens. If a theorist-grandmaster plays a line, it shows the line is playable. But 'playable' and 'good for White' are two separate claims, and the second one needs a separate receipt. I write from a Moscow desk; the board culture of Dhaka I now read out of the archive, not out of a reporter's notebook. I state that limit explicitly, because from a distance it is tempting to pretend the eye sees everything when it does not. The sample arithmetic is simple. One game, two names, zero ratings, zero dates, zero event. 'Difficult for Black to equalize' is a practical claim, not a proven one. Turning a practical claim into a proven one requires three tools: an engine, a multi-game database, and a little patience. The article contains none of the three. What I have is a clear sentence and a vague basis. On top of that sits the showcase-selection problem. The game shown is between an elite player and a lower-rated opponent. White won; that may be the line's merit, or it may be the gap in playing strength. When showcase games are chosen, the ones that worked get chosen — that is selection bias, not conspiracy. But a reader who does not check the number ends up buying the bias. In May 2026 I tried to avoid exactly that error. I published a reconstruction of Spartak Moscow's title run: sixty league goals against a shot-quality model that said 47.3 — a +12.7 overperformance that, in my reading, was finishing luck rather than structure. Two analysts shared it the same day; a television producer said the eye test ended there. I declined two panel invitations because I had not yet validated the model against a second season. I built the Russian Premier League model because the silence was measurable. Since that week every number I publish carries a method footnote — sample size, competition, season, model version. A model is a receipt, not an oracle. The next year, at the 2026 World Cup in my own city, I was running a live shot-quality model across sixty-four matches and 169 goals. Between shifts I was finishing a scrape of forty years of FIDE rating lists. A sports desk in Dhaka wrote about the numbers. I ran the query: Bangladesh had produced five grandmasters, each in a specific year — Niaz Murshed in 2026, Ziaur Rahman in 2026, Reefat Bin-Sattar in 2026, Abdullah Al Rakib in 2026, Enamul Hossain Razib in 2026. In the decade after that, the count was zero. From that moment I lead every comparative claim with a denominator — per decade, per crore of taka, per million people. Why does Dhaka belong in this piece? Because choosing a sharp line is a decision, and that decision needs four resources behind it: a coach, an engine, a database and time. In Moscow a chess school is treated as infrastructure; in Dhaka the system sometimes hangs on a single corporate patron or a single club. The federation's annual budget is about eleven lakh taka — I keep printing that figure because without it every discussion collapses into complaint. After 2026 home norm tournaments effectively stopped in the country. In 2026 the school pipeline was switched off. Academies run by former stars do good branding work, but the money grassroots coach education needs, nobody pays. A 5.e5 recommendation does not fill that gap — it is only a move. 2026 handed me another controlled trial. Across the first sixty Bundesliga matches after the restart behind closed doors, the home win rate fell to 32.7 percent, from 43.2 percent in comparable fixtures the previous season. When the crowd disappears, the data stops lying about home advantage. Three months later the FIDE Online Chess Olympiad final ended with India and Russia declared joint winners after a server disconnection, and I understood that my chess dataset was the cleanest of all: no crowd, no weather, no pitch. Since that year I have abandoned match reports for model-first pieces, and I add a paragraph at the end of every one — what this data cannot see. Let me write that paragraph for this piece. This information cannot see where an engine evaluates 5.e5; it cannot see how often the move is played at the top level; it cannot see what event the featured game came from or what the opponent's rating was. What it can see is a name, an installment number and a promise. The easiest mistake hides right there, and it is taking correlation for causation. A good player beat a weaker one — that does not prove the line is hard for Black. It proves that good players are good. To learn the line's real value you need games between equally rated opponents, at classical time controls, and you need to look at the results from both sides of the line. The second point is about traffic, and I will put it plainly. Erigaisi belongs to the Indian wave, and his rating is widely reported to have reached the neighbourhood of 2800 — but the source I worked from for this piece does not contain that number, so in my ledger it is pending verification. A name pulls readers; it does not add weight to analysis. That is good marketing and a bad receipt. The third point is about time control. A trap-rich sharp line works at classical, where the opponent has time to think and time to suffer the consequences of a mistake. At blitz or short rapid the same position can simply collapse, because the value of a trap depends on the opponent's error, and in a fast rhythm nobody chooses to err — the rhythm decides for you. The fourth point belongs to the content economy. Installment number 740 means a series running for years, which is to say a mature franchise. The natural destination of such a franchise is opening courses, subscriptions and data products. Traps sell fast; a subtle truth sells slowly — and what works for a platform that needs one installment a week is the trap. This is not an accusation against the author; it is the arithmetic of the system. When a recommendation has to go out without numbers, the numbers fall away, and that is exactly the place where the reader should keep his own receipt. So what should a club player take from this recommendation? The line is available — it is legitimate, Sveshnikov-attested and genuinely sharp. But before adopting it, three tasks: run the main branches through an engine, count the games at 2026+ level in a database, and write down separately how many of your own losses came from 5...d5 and how many from 5...Ng4. When those three tasks are done, what remains is not a recommendation but a receipt you built yourself. Looking forward, the signals I will keep watching: whether the engine evaluation of 5.e5 drifts toward equality; whether the move's usage rate at top level rises; where Erigaisi's rating trajectory settles; and whether Black develops a new antidote. None of these is known to me now. One question therefore stays open: in the next installment, who will be the first to write the number down?