44 Ludwika Zamenhofa Mem · 2025.12.28 · Elo 2411/2504 · defensive · ★★★★
The daily position for Saturday 5 September 2026. It is still open to solve — the grader reads your whole line, sidelines and evaluation included, before anything here is shown. Cover the solution and solve it first →
White to move. the model gives this to 1 in 21 at this level.
The solution
63.♗︎c1!(not63.♗︎a1?b364.♗︎g7b265.♔︎b4b1=♕︎+−+)63…a2(if63…b3then64.♗︎xa3b265.♗︎xb2=)64.♗︎b2(not64.♔︎xb4?a1=♕︎65.♗︎h6♕︎d4+−+)64…b365.♔︎c4♔︎d666.♔︎xb3♗︎g8+(66…♗︎b1is as good:67.♗︎a1♔︎d568.♗︎g7=)67.♔︎b4=
63.♗︎c1 — almost nobody finds this
64.♗︎b2 — the standard move
65.♔︎c4 — almost nobody finds this · worth +1
66.♔︎xb3 — the natural move
67.♔︎b4 — almost nobody finds this · worth +1
“hard to find”, “the natural move” and the rest describe how often this move is played by Maia-2, weighted across the 1500/1900/2000 audience the difficulty model is fitted on — a model of human play, not a count of anyone who solved this position.
Members can take this away as a PGN — the diagram, the verified line and their own analysis in one file, for a database, Anki or Chessable.
What happened at the board
Pidluznij found it · ½-½
the game went on65…♗︎c266.♗︎h8♔︎f567.♔︎b4♔︎e468.♔︎a3♔︎d369.♔︎b2♗︎d170.♔︎a1♔︎e4½-½full game ↗
The field
87% of the 15 committed lines here found the key move — the model predicted 5% at this level.
Notes from the room
Johncsimmons · 2026-09-05
This site is an useful tool but some of the software analysis is silly. It was clear that once only one pawn left and bishop on diagonal, which particular king moves didn't matter. To be fair the autual game had a lot of unnecessary moves too.
AnlamK · 2026-09-05
John C Simmons, I completely agree. After 64.Bb2 it's obvious that it's a draw.
AnlamK · 2026-09-05
And the puzzle rating of 2505 seems really high for this puzzle.