Here is the most reliable way to spot a beginner at Reversi: they are winning at move twenty. The board is covered in their colour, they have flipped huge rows all game, and they look comfortable.
They will lose. Not occasionally - almost always, and usually badly, in the last ten moves. The reason is the single most counter-intuitive principle in the game: for most of a game of Reversi, owning fewer discs is better than owning more.
Why disc count is nearly meaningless mid-game
Only the final count matters. Every disc on the board can flip, repeatedly, until the game ends. A disc you own on move thirty is not a point - it is a temporary loan.
So what does matter in the middle game? Mobility: the number of legal moves available to you. In Reversi you must move if you can, and if you cannot, you pass and your opponent moves again.
A player with fifteen legal moves can choose a good one. A player with two legal moves must play one of two, and both may be terrible. A player with none must pass and watch their opponent play as many moves in a row as they like.
Flipping lots of discs destroys your own mobility, because a legal move in Reversi requires an enemy disc to flip. If you have converted most of the board to your colour, you have removed your own targets.
Frontier discs, and the real cost of a big flip
The precise version of this idea is the frontier: discs adjacent to at least one empty square. A frontier disc is one your opponent can potentially play next to, so every frontier disc you own is an opportunity you are handing them.
Interior discs - fully surrounded - are safe. They give the opponent nothing.
This gives a working rule that will immediately improve most players' results: prefer the move that flips the fewest discs, especially in the first thirty moves. A move flipping one disc usually keeps your position compact and your frontier small. A move flipping seven usually sprays your colour across the board and hands your opponent a dozen new options.
Once you start counting flipped discs as a cost rather than a gain, the game reorganises itself.
What a mobility squeeze looks like
Play it out and the winning pattern is consistent. One player stays small and compact through the middlegame, taking quiet one-disc and two-disc moves. The other expands, flipping happily.
Around move forty, the expanding player starts running out of legal moves, because most of the board is their colour and they need enemy discs to flip. They are forced into the moves they were avoiding - usually the squares adjacent to corners, which hand the corners over.
Corners are permanent: nothing can ever flip a corner disc, because flipping requires a bracketing line on both sides and a corner has no outside. From a corner, entire edges become stable. The player who was compact now converts, and the disc count swings thirty or forty in a handful of moves.
The whole game is really about arriving at that moment with moves in hand while your opponent has none.
The squares to fear
Because corners decide games, the squares next to them are the most dangerous on the board:
- The X-squares - diagonally adjacent to a corner - are the worst. Playing one usually lets your opponent take the corner immediately.
- The C-squares - orthogonally adjacent to a corner along an edge - are nearly as bad, though occasionally playable.
The catch is that you cannot simply avoid them forever. If your mobility collapses, you will be forced to play one. Which is the point: mobility is not an abstract nicety, it is what keeps you out of the squares that lose the game.
The general principle underneath
Strip away the discs and Reversi is a game about not running out of good options - which is a shape that shows up in far more games than people notice.
It is exactly the trap that patience players know as filling an empty column. An empty column in a solitaire game is manoeuvring capacity, and spending it for an immediate small gain is the classic losing move: Spend an evening with FreeCell online and the formula that decides how many cards you can shift depends on empty space, so hoarding options beats taking material.
It is also what separates a competent falling-block player from a poor one. Anyone who has kept their stack flat rather than chasing an immediate clear when they play Tetris online is doing the same thing: preserving the ability to accept whatever comes next, rather than optimising the current move.
And the compulsion element - having to move even when every move is bad - is shared with the draughts family, where running out of useful moves is itself a losing condition. If you want to feel that mechanism in a different setting, play Checkers online and pay attention to who runs out of safe moves first.
How the variants differ
The variants here are mostly experiments in how board geometry changes the mobility calculus. 6x6 and Mini compress everything - corners arrive sooner and the squeeze happens faster, which makes them excellent for practising the principle. Grand Reversi enlarges the board, lengthening the middlegame and making frontier management even more important.
Anti-Reversi, where fewest discs wins, is the best training tool on the list. It inverts the goal while keeping the mobility rule intact, which forces you to think about legal-move counts explicitly rather than by instinct.
Rolit adds more players, which dilutes the two-player squeeze and turns it into a genuinely different tactical problem.
One thing to try
Play one full game where you always choose the legal move that flips the fewest discs, with a single exception: take a corner whenever it is available. That crude algorithm beats most casual players, which is a slightly deflating demonstration of how much work the principle is doing.
Then read the rules hub for the exact flipping and passing rules, the corners explainer for why they are permanent, and the glossary for frontier, X-square and the rest. The main game is where to practise.
For the endgame counting rule that decides close games once mobility is equal, see the companion piece on parity.