When to add cheaters

[mcdermott2]

I'm curious what rules people have about adding black squares that do not increase word count (I call them cheaters). Obviously I try to keep them to a minimum, but are there instances other than total necessity where they can sneak in without degrading the quality of a puzzle?

One example- I have a puzzle where hemps and asks intersect at the s and are in a place where I could insert a cheater at this intersection. I think the s-less versions of these words are better, but I am not sure if using a cheater is acceptable.

Thanks!

[dirtywater]

Wouldn't a cheater be worse than an S? Leaving it as-is seems better...

[worldofcrosswords]

I agree with dirtywater. However, I also don't see any harm in using a cheater if it just adds two additional black squares to the puzzle. JMHO.

Marya

[mcdermott2]

Yes, you are probably right. I felt like I had an inordinate amount of plurals ending in esses, but adding black squares seems like a bad solution.

[AlexBajcz]

I think plurals are a fact of life in puzzles and, unless there's dozens (figuratively speaking!), I wouldn't be too worried about it.

I'm of the mind that "the fewer black squares, the merrier," but I, too, would be curious to know what people thought about adding cheater squares in general. For example, I'm currently at 36 black squares (my limit, generally), but two cheater squares would really help me out. Plus, because my theme involves 4 13-letter answers, I already have 6 cheater squares elsewhere in the puzzle. How hesitant should I be about adding more cheater squares?