Good Custom Position, Bad Custom Position (Dos and Don'ts for NCPs)

As a CGA team (Custom Games Admins), we've decided that it might be time to have a talk with all of you eager position-creators. We love your enthusiasm, and have been impressed by the creative arrangements you've come up with thus far.

However, it seemed necessary to point out a few key things we like to see to approve a custom position. These are all mostly subjective, and are certainly not requirements per se, but are given to you as guidelines. Just like any guideline, there are exceptions, and breaking one or two with a justifiable reason doesn't mean your position will be automatically declined, but that said, if we can't see a reasonable justification for why a guideline was ignored, it will be.

1. Positions should be balanced and fair.
   • Quick or forced mating lines after two or three bad moves from any player should not exist. There are some cases where a position is meant to be played by players who already know what they're doing, and will be expected not to blunder in the opening, but in general, custom positions should be able to be learnt on-the-fly.
   • Symmetry is a good place to start, but is not a requirement for this principle. There are plenty of ways to design an asymmetrical position in which play is balanced for all players.
   • No one player should have an easier time attacking or defending than any other player.
2. Having too many capturing or attacking moves available on move #1 should be avoided.
   • In general, a minimal amount of developing moves should need to be made before any player can launch an attack.
   • Positions where players can make attacking moves from the start against one or more opponents tend to lead to a game that's too chaotic. The goal of any position should be clear, and achievable (get the king to the hill, land 3 checks against the player to your left, control the centre, etc.). Typically attacks aren't just one move; there's some initiative built up with a building series of attacking moves. If a player has to deal with this from one or more opponents from the beginning, it makes the goal of the game unclear, and un-achievable.
   • No one wants to lose rating because he was randomly assigned to play as Green, and got attacked three ways before even having the opportunity to make a move.
3. Having a position where two or more players can "gang up" on another in the opening should be avoided.
   • If two or more players can attack another opponent from the beginning, it leads to a position that's farmable by a particularly nasty form of cheater: the pre-arranged teamer. This strategy is clearly forbidden by the rules of play, but that doesn't mean it doesn't happen. The CGA team has a responsibility to actively discourage pre-arranged teaming by rejecting positions that are ready-made for teaming.
4. Attacking ideas shouldn't be served up on a plate, but a position where setting up an attack takes too many preparation moves will be boring.
   • Good custom positions will give players the space they need to mount some threats. If things are too "blocked-off" and a player needs to go the centre to even interact with another player's pieces, the opening won't be exciting enough.
5. Defensive fortifications should be easily achievable.
   • In the same way that there should be a clear avenue of attack for aggressive players, there should be ideal ways to parry those attacks for defensive players.
   • If fortresses are attainable, there should be secondary ways of winning besides giving checkmate. i.e., king-of-the-hill, by point-spread (+40 or oxN are good gamereules which can be used to encourage this).
6. Players' available moves should be plentiful.
   • Positions with too few pieces, or too many walls will limit where and how players can move, leading to repetitive games. One reason for classic chess' enduring popularity as a board game is that it's so open-ended. After just three moves, there are already over 120 million different games that could have been played. Creators of custom positions should consider how much variation exists in a custom position.
7. Each piece added to a position should have a purpose.
   • Positions where pieces exist solely to be traded are dull.
   • Consider why a piece was specifically chosen to occupy a square. What properties does that piece have that can't be served by lower-valued piece? How can that piece be used most effectively from it's given position? These questions should have clear answers.
8. The goal(s) to winning a custom-position game should be reasonably simple.
   • Examples of game-winning ideas include, finding a forced mate, getting your king to the centre, checkmating at least one opponent for a guaranteed first- or second-place finish, etc.
   • If one avenue to victory goes astray, there should be secondary goals that players can achieve.

As mentioned above, these are not rules, but rather guidelines. I'm sure there are many positions which have already been approved that break one or two of these principles. But in general, these are the aspects of custom positions which make the games fun - and that's what it's all about after all right!?

I'm appending an article penned by @GustavKlimtPaints regarding the final step before submitting your newly-created custom position: TESTING. It's important that after you find a setup you like for a new custom position, that you verify all of these guidelines hold up. The only way to do so is through extensive gameplay with computers, and players of varying strengths. We get frustrated when looking at a position where there's an obvious problem that could have been avoided with minor tweaks, and you may be blind to these if you're only playing through the games quickly to meet the five-game requirement for submission.

I’ll just read the whole thing twice through later

hmmm are there CGA nominated testers? (just a thot)

As an aside, I suggest topic posters spell out CGA the first time it's referenced. If the goal is to educate the hoi polloi about the do's and don'ts, then better to make the content more accessible. 

No. Although actually, when we're running our tests, we're always eager to find some folks willing to hop into a game and try it out. Usually, a member of the CGA team will host a queue publicly, so you'd be free to join. Another way would be to add some of the CGAs to your friends, so if we are testing, we can send an invitation to you. 

Good point. Fixed =]

Does anybody know where I can find the requirements for a WOF

https://www.chess.com/clubs/forum/view/4pc-wheel-of-fortune-information-1

Requirements are the same as NCPs. =]

This was helpful