What Makes a Good Chess Construct?

A chess construct is a type of chess problem but it does not necessarily follow all the conventions of traditional chess problems. For instance, the first (or key) move might be a check or capture. You can read more about them here. After years of developing Chesthetica, arguably the world’s most advanced automatic chess problem composer, it has always fallen upon me to choose the compositions that not only do I like, but that most chess players would probably find aesthetically appealing, interesting, educational or perhaps even all three.

Chesthetica Composing Chess Constructs

So there are many criteria (and exceptions) that I usually follow when having to sift through hundreds or even thousands of compositions by the program every few weeks and line them up for publication in various places such as Steemit, WordPress, Minds and Reddit. If a steadily increasing number of social media likes and shares by chess-playing community members are any indication, I am likely doing something right in this regard. Here are some of the characteristics I consider and have tried to simplify for the purpose of this article. These are beyond the default conditions of only having one key move in the solution lest the problem be ‘cooked’ and (as far as possible), not having any unnecessary or irrelevant pieces on the board, i.e. an economical setup.

1. White (i.e. the winning side by convention) has equal or less material than Black (yet still wins).
2. Ideally, there should be no check, capture or promotion in the key move.
3. If White has more material than Black, there should be no pawn advances to the 7th rank either in the key move; this would be seen as an ‘obvious march’ toward a pawn promotion.
4. If White has even more material than Black, there should be more necessary/relevant pieces, in total, on the board as this usually means a more sophisticated or complex composition.
5. Underpromotions that are not necessary should be avoided; for instance, checkmate by promoting to a rook or bishop when a queen, obviously, would have also worked.
6. Compositions that feature White sacrificing material (at a minimum, equivalent to a minor piece or three Shannon points).
7. Dense or ‘cluttered’ positions, such as this one; the closeness of the pieces adds perceived complexity.
8. Positions where White may have more material than Black but has fewer pieces; these at least tend to look like Black has a better chance of surviving.
9. Where the next winning line is at least three or more moves longer; for example the shortest mate in the position being four moves and the next shortest mate being seven or more moves away (ideally, no other move mates at all, or even more ideally all other moves lose).
10. If possible, there are no duals in the main/chosen line of the solution (more a preference than a strict condition); so if there is more than one white move for each black move, there is a dual.

One of the distinguishing features of chess constructs is that they are somewhere in the middle-ground between interesting sequences that might arise in real games and traditional chess problems. The advantage of this is that they are not as ‘uneconomical’ as a position in a real game yet still appear ‘composed’ but easy enough to understand or process. This is because many/most players, myself included, often find traditional chess problems to be too sophisticated for their taste and too time-consuming to solve or understand. Traditional chess problems also tend to be extremely unlikely to arise in a real game and therefore less useful to chess players hoping to improve their play.