Game The short answer
In all 3 positions the X-square was not the best move, and the gap to the best move ran from 6 to 20 discs. It costs that differently from position to position, so neither "the X-square loses the game" nor "it hardly matters" survives the numbers.
What we are testing
Othello has a long-standing rule: avoid the X-squares (the diagonal neighbours of a corner — b2, g2, b7, g7) and the C-squares (the edge neighbours of a corner — a2, b1, and so on), because they tend to hand the corner to your opponent. But "it just feels risky" is easy to say. Here we take a real position, solve every move to the end of the game, and see exactly how costly these squares are.
How we tested it
Every number here is output from our own solver (negamax with alpha-beta). We take an endgame position with few empty squares, play each legal move in turn, and solve what follows to the very end of the game. There is no evaluation function and no approximation, so the margins are exact rather than estimated. One position cannot show how typical something is, so each article solves three.
Position by position
Black has seven legal moves. Solving all of them to the end of the game gives the table below. The margin is the final disc differential from Black's point of view with best play by both sides: positive means Black wins, negative means Black loses.
| Move | Discs flipped | Final margin | Type |
|---|---|---|---|
| g4 | 1 | +12 | — |
| e1 | 5 | +12 | — |
| a5 | 2 | +6 | — |
| b7 | 2 | ±0 | X-square |
| e8 | 1 | −2 | — |
| g1 | 1 | −10 | C-square |
| a7 | 1 | −12 | C-square |
Where the difference comes from
The best move is g4, finishing +12, and it flips just a single disc (f5) — a very quiet move (e1 ties it at +12). Once again, the move that flips a lot is not the strong move.
The X-square b7 finishes ±0 — a draw. That is 12 discs worse than best: it throws away a clear win and settles for a tie.
Worse still are the C-squares: a7 is −12 and g1 is −10, both losses for Black. Playing right next to an empty corner opens a line for the opponent to seize it — and the full solve turns those moves into decisive losses.
Exceptions
The X-square is not what costs the most here. b7 holds on at ±0, a draw, while the two C-squares lose by more: a7 at −12 and g1 at −10. Filing X- and C-squares together under "dangerous" hides which of them is actually the heavier mistake.
Nor is avoiding the marked squares enough on its own. e8 is an ordinary move - neither X nor C - and it still loses, at −2. With this few empty squares, reading the position out beats memorising a list of squares to avoid.
Playing the best move
Watch the best move, g4, being played: the disc goes down first, then the captured discs turn over one at a time. The gold frame marks the square played; the gold rings mark the discs that flipped.
As stated in our editorial policy, the moves and numbers are the real computed results, printed as-is.
Does it hold in other positions?
To see whether the same thing happens elsewhere, we solved more positions where an X-square was available.
| Position | Best move | Move tested | Gap (discs) |
|---|---|---|---|
| Case 1 | g4 +12 | b7 ±0 | 12 |
| Case 2 | e8 +20 | g2 +14 | 6 |
| Case 3 | d1 +12 | b2 −8 | 20 |
Across the 3 positions the gap ran from 6 to 20 discs — that is how differently the same mistake can be punished, depending on where it happens.
Case 2 — 10 empty squares
Best was e8 (+20). the X-square move — g2 — gives +14, a gap of 6 discs.
| Move | Flips | Final margin | Diff vs best |
|---|---|---|---|
| e8 | 4 | +20 | best |
| d2 | 2 | +14 | -6 |
| g2 | 5 | +14 | -6 |
| d8 | 5 | +12 | -8 |
| h6 | 2 | +12 | -8 |
| g4 | 6 | +10 | -10 |
| h4 | 3 | +10 | -10 |
| h2 | 5 | +6 | -14 |
Case 3 — 10 empty squares
Best was d1 (+12). the X-square move — b2 — gives −8, a gap of 20 discs.
| Move | Flips | Final margin | Diff vs best |
|---|---|---|---|
| d1 | 6 | +12 | best |
| a8 | 1 | +10 | -2 |
| a4 | 3 | −2 | -14 |
| b2 | 8 | −8 | -20 |
| a3 | 5 | −10 | -22 |
The same question over 400 positions
The positions above are solved one move at a time. The same question was also counted over 400 random positions (10 empty squares, 2617 moves solved in total).
An X-square was legal in 153 of them, and it was the best move in 14 (9.2%). Playing it every time one was available would cost 16.77 discs on average against best play.
What this does and does not show
- It shows: for these 3 positions, an exact solve of every legal move to the end of the game. Not an approximation.
- It does not show: anything about every position — 3 is a very small sample. For a wider count see the 400-position study.
- The assumption: the margins assume best play from both sides to the end.
- Reproducing it: the positions and results are in
assets/othello-analysis.json, regenerated bytools/generate-othello-analysis.mjs.
What to remember at the board
- In all 3 positions the X-square was not the best move
- Across 3 positions the gap ran from 6 to 20 discs — how much it costs depends on the position
- Over 400 positions it was best only 9.2% of the time
- While a corner is still open, leave the X-square out of your shortlist
Try it yourself
You can set these positions up and play them out. The evaluation bar moves after every move, so the difference between a good move and a poor one shows up as you play.
Check the same position yourself
Feel how costly the danger squares are, with the evaluation bar and hints. Free, no account.
⚫ Enter the Dojo