LEFT/RIGHT SYMMETRY

Will Nediger, of London, Ontario, has a Ph.D. in linguistics from the University of Michigan. These days he makes his living by constructing crosswords and writing trivia questions. His quiz bowl team won the 2016 ACF Nationals and the 2017 Intercollegiate Championship Tournament.

The reason for this puzzle's unusual grid pattern with left/right symmetry will soon be clear.

This one was brutal to construct because the non-theme fill couldn't have any L's or R's; avoiding one common letter is hard enough, but two is a nightmare. So there's more stuff like SMEE/SSTS/DOEST than I like, but overall I'm happy with how the fill turned out.

There's one entry I'd avoid if I were writing this puzzle today: OBESE. My strategy used to be to clue it matter-of-factly, and I'm glad to see the editors kept my clue, but I've since just removed it from my wordlist entirely since I think it's an inherently loaded word that implies a negative judgment.

Back in 2016, I built a 15x15 puzzle using the exact same L/R idea, except with better pairs. Will (Shortz) quickly accepted it in early 2017.

It had been sitting in the queue until a few weeks ago when he sent me a message: Will (Nediger) had recently submitted a Sunday 21x21 version that they couldn't say no to, so they were going to pay me $300 to kill mine. (Normal print fee is $750.)

I was irate at first — what kind of messed up business practice is that?

But then I thought about it from Will (Shortz)'s perspective. Last time he gave an inventory count, he only had 4 Sundays on file. (For other days of the week, he had about 30-50.) That's shockingly low — especially considering that the Sunday puzzle is the NYT's flagship puzzle.

So I figured it'd be better to let it drop. In the long run, I enjoy having puzzles in the NYT. I reasoned that Will and Joel might give me enhanced consideration in accepting future submissions, given today's bitter pill.

Back to the puzzle. When I created my version, I had three things in mind:

1. I wanted long entries, with multiple L/R replacements. I found hundreds of pairs where there was only one L/R difference, so that felt too easy.
2. The concept would get too repetitive over a 21x21, so a 15x15 was 100% the way to go.
3. Pairs had to be clueable as opposites — that's essential to the "mirror" concept. (I'm an identical twin — the evil one, of course.)

One example: COLONELS [Ones at the start of wars] and CORONERS [Ones at the end of wars].

I ended up also using MOLASSES and MORASSES (my only pair with a single L/R difference), but I clued them as [Causes of some good sticky situations] and [Causes of some bad sticky situations]. Much more interesting than today's clues.

I do like one of Will (Nediger)'s pairs a lot: GO TOPLESS and GO TO PRESS features a fun parsing difference. The rest didn't do much for me.

And I did appreciate the gridwork challenge. I had a hard time working with a 15x15 using no other Ls or Rs in the grid. Big kudos to Will (Nediger) for his smooth fill today.

My puzzle recently got accepted by the WSJ, if you want to compare and contrast.

I've worked in corporate America for a long time. Sometimes you have to remind yourself that it's just business, nothing personal.