4pc Antichess - General Theory

I like playing chess variants, which is how I found my way to 4pc chess a little more than a year ago. A few months ago, I slowly transitioned to play more antichess than regular 4pc games. I do not know if there is an elaborate theory discussion anywhere, but this is my 2-cents. Should there be interest, or if I will have the time and muse, I may add additional posts.

This post will focus on the general theory of 4pc-antichess, the way I see it. Before we get into it, let me start with the disclaimers:

1. These ideas are mostly true for 4pc antichess FFA, in regular/960 setup. Most of the ideas are also true for Solo. However, the further we get away, by making it teams or adding KOTH, or variations like diamond-wall or knight-chronicles - less ideas may apply, and some might break.
2. There are elements of luck in the game, somewhere on the spectrum between none to coin-flip, and more than in a standard 4pc game (here red decided 1st/2nd/3rd place). However, there are many other games that involve both luck and skill such as bridge and poker. Quantifying the amount of luck involved is out of the scope of this post.
3. I added examples from games that I played in the last two weeks. Not too many examples, but enough to hopefully pass my points. They are supposed to be instructive on specific points, not to criticize any specific player (you may note that some criticized scenarios are exemplified by poor positions of my own).
4. This is only how I see things. If someone disagrees, or have anything to share, I would be interested in exchanging ideas. I suppose the differences between FFA to teams and/or KOTH is especially interesting.

Ok, let’s begin... This post will review 4 general concepts without getting into too many details: control, an equivalent of “tactics”, timing, and the pieces.

A: Staying in Control

Because capturing in antichess is forced, the most important intermediate-goal is to stay in control. Obviously, the end-goal is to lose all of your pieces first, but in order to do that one needs to stay in control. When things get out of hand, usually in an unfavored endgame, it feels like you have no choice but to capture pieces while the other players keep feeding you.

There are two types of control: “strong” when you can move however you like (i.e. not forced capture) and “weak” when you have several options to capture. The game begins with strong control, which quickly transitions to weak control, with phases in which players lose and regain strong control.

Usually, the opening and middle game are the first few moves in which most of the players are in weak control. One of the most dangerous things is to have only a single piece that can capture anything, and worse is when it only has one choice. It is not too bad if it happens in the opening, or happens on a single move, but lack of control in the endgame when the board is sparse enough could lead to disastrous results. In this game, a wrong promotion of mine (should have been a king) led to a long sequence out of control. This is especially devastating in an endgame when the chances to regain control are fewer than in a middlegame.

B: “Tactics”

Well, there are no proper tactics in antichess, but their analogy are capture-chains and capture patterns/positioning that should be useful to notice. You want to help your opponent to capture your pieces, which can happen in several ways.

First, if a piece happened to get deep into your territory, you want to keep it there. A classic example is a rook on the 1st/2nd rank who can do a lot of capturing, or a knight that also stumbled too deep. This game is an example of trapped rooks (blue, red).

When these things happen, you should try to keep these pieces around by making sure they can keep capturing your pieces. This means that you do not want to threaten them, and if you do, it is preferable to be able to capture something else, elsewhere (having weak control). Moreover, it is not guaranteed that the opponent will capture your pieces if they can, so another common idea is to block the piece with your pieces (if you can). For example, if a bishop has a chance to capture you neighbour’s piece, move a pawn to block that diagonal, so you lose at least another pawn before the bishop escapes into another’s territory.

Second, the other perspective: if you have a piece too deep in enemy territory, try to capture with other pieces, or escape, before you capture too much pieces of the same player. It is quite common to see player A captures a piece of player B where both are in the territory of C. This reduces the effectiveness of C losing pieces.

Third, capture-chains become more obvious in an endgame. The middle game is too chaotic to force an opponent to execute a capture-chain, but when the board becomes less populated, it gets easier. It is also possible to prepare a chain in advance, but do it carefully (see “Timing”). A classic capture-chain can be set up in front of a pawn, or a king when pieces are clustered. In general, any piece can have its own chain, depending on the way it moves. Here is a nice example with a knight.

C: Timing

Tempi are important in antichess. The worst blunder happens when you are in strong-control, and move a piece to be captured, just to find out that the other player has weak-control and doesn’t capture it, so now your piece is loose.

In general, it is important to take note on how much time it takes a piece to complete its captures, and if there could be intermediate-moves that cause the timing to change. For example, when two rooks (/bishops) face each other, if the first to move can capture another piece while maintaining vision with the second rook (/bishop), it will be captured afterwards, unless that other can also do the same, or if there is another capture choice to that player.

This issue is not only relevant in active faceoffs, but also when two players want to make another capture their pieces. If both of them approach at the same time, the third player can choose their capture, and potentially force the un-captured piece to capture. This happens for example when two kings approach two different pawns, one will be captured and the other will have to capture the pawn it approached.

Sometimes, it is not worth the risk, and one player would yield by not approaching until they can guarantee to be captured. This may seem like suddenly teaming, but it is in fact the rational thing to do, to guarantee no unlucky choice happens. Usually if one approached the second won’t risk approaching too, and sometime if you have more to lose you should not approach even if the other player plays after you. Things are obviously different if this is a race for the win. For example, if two players have a single king remaining and compete for the first place, when one approaches the other must too, and hope for being lucky on being captured. Note that if the luck is for getting 2nd or 3rd place, but risking it could mean 4th, then it may be better to concede and not risk it, especially if the capturing player can secure the 3rd place by capturing correctly.

Overall, the main point above about timing is to be careful of moves that make sense in a “sterile” game but might have disastrous results due to intermezzo moves or luck due to another player’s choice. Another point to keep in mind is that while a player is forced to capture, it is possible to approach them to set up your pieces to be captured. This is great when you are not the one being forced to capture.

D: Pieces: Strengths and Weaknesses

Bishops. Unlike standard chess, in antichess there is no clear “winner” among piece strength, but the clear “loser” is the bishop. This is the worst piece, because of its long range and low control, which makes it easier for opponents to exploit it to capture their pieces. In this example I didn’t manage to lose my bishops for a long time. While bishops are mostly bad, they have unique endgame features and sometimes a pawn should be promoted to a bishop, especially against knights. [Note: I suspect that in team antichess they are even worse, because they can’t run-away on the diagonals to the player across, so it is harder to get rid of them when they run wild.]

Rooks. At first glance the rook is ranged like the bishop, so it must be as bad... This is not the case, because it moves in direct lines, so unlike the bishop which immediately becomes open to the other players, rooks can be controlled better and dislodged only when appropriate. Sometimes they get freed by capturing the pawn in front of them, and then it is useful to get rid of them as soon as possible. Because they are ranged and are more easily controlled, they do provide benefits in endgames by pulling other rooks into open files, or initiating a capture-chain of a pawn, because if there is an empty file near the pawn, they can approach it without risk. Also, when a pawn is promoted in an endgame, if there is no risk of forced capture in the next few moves – it would usually be good to get a rook. In the following examples, three players were left with a rook at the very-end, two hiding behind their pawns for a few moves.

Queens are also ranged pieces, but allow wide choice of captures once they are loose. They are not recommended to survive to an endgame, because then there is less choice in captures, but in the middle game it is useful to keep them around, in order to maintain weak-control and refrain from capturing opponent pieces that roam in you camp. That being said, you shouldn’t keep the queen “at all costs”, but rather find the delicate balance how long to use it to capture other pieces (if you capture too much, you help your opponents).

Knights are a short-ranged piece, which is why it is usually safe to keep, and many of them survive to an endgame. Like rooks, they are useful to trigger capture-chains against pawns, and they can jump around to harass other players that will need to be careful of having to capture the knight and losing critical tempi in which other players might approach them as well.

Like knights, the king is another safe piece. Since I don’t consider KOTH, the king is simply just another short-ranged piece, which is useful. It is slower than a knight, but steadier, so which of them is better is a question of the board’s setup. Generally, I would say a king is better, but if the remaining pieces are at the other side of the board, a knight would reach it twice faster, which may be the difference between winning and losing. Also, a king is the most common promoted piece since it is usually a safe (and useful) choice. [Note: in KOTH, the king obviously has a special role, and losing the king is bad since the only other way to win would be to lose all the pieces, which is usually slower to achieve.]

Pawns are a good cover for your rooks, but one should aim to lose many of them relatively quickly. The problem with pawns is that in the endgame they are very slow, and therefore an easy target for other pieces to be captured by them. For this reason, few pawns are better than several, and double-pawns are bad because they are less dynamic. Getting into an endgame with only pawns also means that you can only move forward, and unless there is a blocking piece to stalemate a pawn, it is preferable that your last remaining piece would not be a pawn. It is not something that can be fully controlled, but what should be aimed for. In this example, 2 pawns lost to 5 pieces due to mobility.

That is it for the general ideas. If you read this, I hope you found it useful.