The essence
You know a fighting game in terms of frames: startup, block stun, advantage, punish windows. That is the right instinct — you are asking what the system is, underneath the moves. This page answers the same question for draughts, with the same rigour. Every number here was measured over 200 real games; none of it is advice.
Start here
The rule everyone learns last, which is actually the whole game.
The win condition is: your opponent has no legal move. Capturing every piece is one way to arrange that. It is not the definition, and it is not the only way.
One game in seven ends with the loser still holding pieces — surrounded, blocked, out of moves. Those games were never about material. And in the other six, capturing was the method, not the goal.
Think of the health bar as a red herring. The real resource is options. Every capture is a way of taking options away, and taking options away is a way of winning — but you can also take options away directly, by blocking, by crowding, by leaving a piece nowhere to go. Material is the most convenient currency; it is not the account.
The mechanic
This is your block stun, and it is a bigger share of the match than you would guess.
At any moment you are in one of exactly two states.
71%
No capture is available. You choose from an average of 5.6 moves. This is neutral. This is where games are decided.
29%
A capture exists, so you must capture, and you must take the maximum. Often there is exactly one legal move. You are not playing; you are being played.
Nearly a third of the game is spent in a state where you have no agency at all. That is the single most important structural fact about draughts, and it is what makes it unlike chess.
FORCED is block stun. When you put your opponent into it you know exactly what they will do — not probably, exactly — and you knew before you moved. A forced sequence is a combo: no choices, no reads, just execution you can verify in advance.
Which means the real skill is not escaping FORCED. It is choosing when your opponent enters it, and on which square. Everything strong players do in the neutral state is setup for that.
The number that decides games
Mobility — how many legal moves you have — is the health bar.
Here is the average number of legal moves available to the eventual winner and the eventual loser, over the closing plies of every decisive game measured:
Look at the shape of it. The winner's mobility barely changes — they are not doing anything special at the end. The loser's mobility falls off a cliff, and it has already begun falling sixteen plies out, long before anything looks decided.
Count your moves. Actually count them. If you have four and they have seven, you are losing — regardless of material, regardless of how the position looks. That single number is a better predictor than the piece count, because it is closer to the win condition.
And when you are ahead: do not just take pieces. Take squares. Every square you deny is a legal move deleted from their list, permanently.
Stage geometry
This is pure geometry. No pieces, no position — just the shape of the board.
A jump needs three things in a line: an attacker, the victim, and an empty square directly beyond. So a square's danger is simply how many of its four diagonal directions have a square on both sides. Compute that for all thirty-two squares and something strange happens:
The answer is not a gradient. It is binary. Every square on the rim has an exposure of zero — a piece standing there cannot be jumped, by anything, ever, while it stays. Every square off the rim has an exposure of four — it can be jumped from all four directions.
There is no middle ground. You are either untouchable or fully exposed.
The rim is the wall, and standing against it is completely safe and completely passive. A piece on the edge has two diagonals instead of four, so it can be hit from nowhere and can act in half as many places. The engine prices this exactly: a piece on an edge file scores −2, and a man on your own back row scores +6.
That is not a positional nicety. It is the direct consequence of the geometry above — the back row is the only place a man is invulnerable, which is precisely why giving it up is permanent and expensive.
Because the edge is safe, weak players hide there. But the win condition is options, and the edge halves yours. You are trading the thing that loses games slowly for the thing that loses games quickly. Use the rim to park a piece you are not using yet — never as a plan.
Your mixup
And the numbers say most people use it wrong.
When you move a piece somewhere it must be taken, you have not lost a man. You have done something much stranger: you have chosen which square their piece will be standing on next turn. Compulsory capture means they cannot decline, and take-the-maximum usually means they cannot even choose how.
That is the deepest mechanic in the game. It is the only move type that reaches across the board and moves an enemy piece to a square of your choosing.
This is your throw, or your unsafe-but-scary launcher. It is the highest-leverage tool you have and it is minus on block — measured, it goes wrong more often than it goes right. The difference between the two is not courage. It is whether you counted what happens after the forced capture, to the point where nobody can capture any more.
Every good offer is a three-beat sequence you verified before you started: offer → their forced take → your strike. If you cannot say all three out loud, you are not making an offer. You are hanging a piece.
When the strike lands, this is how much comes off:
Startup frames
One statistic on this page matters more than the rest.
Not "usually helps". Not "is an advantage". 82% — measured over games, against an engine playing both sides. And the first crown arrives at a median of plies, which is not the endgame. It is the middle of the fight.
A king is worth about 3.2 men to the evaluation. That understates it, because a man can act on two squares and a king can act along four entire diagonals. The reason the number is so lopsided is not the material — it is the options again. One king adds more legal moves than three men, and legal moves are what you are actually playing for.
Crowning is a startup cost you pay once for a permanent frame advantage. Every quiet move you make should be checked against one question: does this get me there first, or them? Being two tempi ahead in that race justifies a great deal of otherwise reckless play — including giving up a man to clear the road, because a man costs 100 and the crown is worth 320.
The seventy-one per cent
Since a third of the game plays itself, everything rests on the other two thirds.
In the FREE state you have about six moves. They are not six equal options. They fall into four kinds, and knowing which kind you are playing is most of what it means to have a plan.
Moves you closer to a crown. Pays off in the race, costs you exposure — you are stepping off the rim into a square that can be hit from four directions.
Puts a man behind another man, so entering the square in front is punished. This is how you make advances safe. A staircase, never a flat line.
Takes a square away from them. Does nothing for you directly and is often the strongest move on the board, because the win condition is their move count.
Forces them into a capture you have already calculated to the end. Your mixup. Correct about a third of the time, which means it needs verifying, not feeling.
The habit worth building is naming which one you are playing, every move. Most club-level losses are an advance that was played as if it were a support — a man pushed forward with nothing behind it, onto a square with an exposure of four.
In a locked position the player who must move is the player who must break something. That is the closest thing here to frame advantage, and it is countable: how many quiet moves can I make that do not weaken anything? Count yours, count theirs. The one who runs out first has to step into an exposed square, and the suffocation curve begins.
The translation
| What you know | What it is here | The measured number |
|---|---|---|
| Block stun — they must respond | Compulsory capture. They have no choice, and often exactly one legal move. | |
| Combo — a verified sequence | The forced chain. No reads, no branching: you can check it before you commit. | |
| Throw / unsafe launcher | The offer — the only move that relocates their piece to a square you picked. | |
| Health bar | Not material. Legal moves. You lose at zero. | |
| Chip damage / pressure | Denying squares. Every square taken is a move deleted permanently. | |
| Stage wall | The rim. Exposure zero — completely safe, completely passive. | |
| Startup for a big payoff | The crowning race. Paid once, advantage permanent. | |
| Neutral game | The FREE state. Advance, support, deny, or offer — pick one and know which. |
The whole thing
You said you can move pieces and win some games without knowing what is happening. What is happening is this: two players are taking options away from each other, and the pieces are only the mechanism. Once you are counting moves instead of men, the game stops being a series of trades and becomes what it is — a squeeze.
The strategy guide is the same material as technique — shapes, shots, endgames, all with playable diagrams. How the engine thinks is the calculation procedure: five habits that turn this understanding into moves you can actually find at the board.
Recent
Everything here
Nothing here by that name.
Keyboard shortcuts