Not currently surfaced in the X-ray Coach. This is a whole-game pipeline label, computed across an entire game rather than for a single position, so it is not part of the per-position tactic set the X-ray Coach receives.

How Koga decides

Definition A winning position made less winning
  • A missed opportunity is not a pattern on the board — it is a judgement about the move that was actually played versus what was available. Koga raises it when you were already clearly winning, had several much stronger continuations to choose from, played a weaker one, and yet remained winning afterwards.
  • In short: you let a big advantage shrink, but you did not throw it away. The win is still there; it just got harder.
When we flag it The three conditions
  • Already winning — before the move, the position was worth at least a clear advantage (about a pawn and a half or more) for the side to move.
  • Better was easy to find — of the engine's top five candidate moves in that position, at least three were clearly better than the move played (each by at least a pawn). When that many moves all beat yours, the improvement was not a hidden, only-move resource — it was sitting in plain sight.
  • Still winning afterwards — after the move played, the position is still clearly in your favour. The advantage was reduced, not surrendered.
The dividing line Missed opportunity vs. unforced error
  • The same test that catches a missed opportunity also catches its harsher cousin. If you were winning, several better moves were available, and the move you chose throws the advantage away (the position is no longer winning afterwards), Koga calls it an unforced error instead.
  • So the only difference between the two labels is the position AFTER your move: still winning means missed opportunity; advantage gone means unforced error.
  • If fewer than three clearly better moves existed, neither label is raised — the position is then handed to the normal pattern detectors to look for a concrete tactic.

Test positions (the examples that guard against regressions)