Note: I thought it’d be interesting to accompany some of these posts with original puzzles where I try to put the ideas I’m talking about into practice. If you’re inclined to try out this post’s related puzzle, feel free to solve before or after reading!
If you can’t tell from my previous posts, I get a real rush from analyzing crosswords from a solver point-of-view – particularly when it comes to the theme. (Earlier this year, I wrote about how putting a puzzle’s revealer in different grid locations changes the “aha” moment for solvers.)
You know what I don’t get a real rush from? Admitting I’m wrong.
Well, not exactly wrong: Last year, I wrote about ways to sequence theme entries to create the most satisfying “ending.” I don’t necessarily disagree with what I wrote; but I do think there was an aspect of the solving experience that I glossed over – which I learned about firsthand last month from a Thursday New York Times crossword by Kit Sheffield.
(So consider this post an addendum to that older post, “Ending puzzles with ‘OOMPH!'”)
Spoilers (click for solving link):
New York Times: Thursday, January 29, 2026 by Kit Sheffield
The Puzzle in Question
This puzzle features three repetitive, seemingly nonsense 15-letter theme entries. The trick is to reinterpret them as a “row” of letter combinations, which leads to a “___row” and results in a new word.
Themers
- 20a: [Expand]– G-G-G-G-G-G-G-G-G-G-G-G-G-G-G (alternately parsed as a “G”-row, to result in GROW)
- 37a: [Sadness] – SOR-SOR-SOR-SOR-SOR (alternately parsed as a “SOR”-row, or SORROW)
- 52a: [Day after today] – TOMOR-TOMOR-TOMOR (alternately parsed as a “TOMOR”-row, or TOMORROW)
One notable decision, to me, is that there’s no revealer to explain what’s happening. Besides the theme entries being italicized, there’s nothing in the clues that definitively refers to the existence of a theme.
You could imagine an alternate version with a cheeky bottom-right-hand-corner entry for ROW, clued like [Line of things, as displayed in the theme answers]. I don’t think this alternate version is necessarily a bad choice – it would probably make the “aha moment” more easily accessible, i.e. reduce the number of crossworders flocking to online blogs that tell them the theme.
But this published version doesn’t use any sort of revealer – not in the grid nor a title – which leads to an interesting effect. It means that the theme entries have to collectively do that work and show what the theme is about.
Forgoing a revealer isn’t unheard of. For instance, you see it a lot in “creating wacky phrases” types of puzzles – e.g. the themers are all common phrases, except the first letter is changed to a Z to create a new effect (maybe [Offering from a drag-racing car shop] clues to ZOOM SERVICE, a transformation from “room service”). For those types of “letter-changing” themes, the trick is pretty obvious early on.
In contrast, this NYT puzzle has a more complicated mechanic – one that I certainly didn’t understand immediately. Yes, it’s possible that I only solved this puzzle through a combination of luck and divine intervention. But maybe there was something else at play…
A Nudge in the Right Direction
Fortunately, the puzzle wants to see us succeed. (I think.)
In particular, I think the last theme entry – TOMOR(row) – does a great job of “giving the game away.” The clue [Day after today] is a very simple and straightfoward way of hinting at the word “tomorrow” – I think even a non-crossword person has a good shot at getting the correct “answer” if you told them that clue.
To me, this clue-answer pair offers the solver a much clearer path to successfully untangling the trick. By hinting at TOMORROW so overtly, you can at least be a little more confident in the wordplay “destination.” In contrast, [Expand] and [Sadness] have a lot of potential answers/synonyms, so they’re harder to confidently interpret as “grow” and “sorrow.” It’s not necessarily hard to “solve” that first theme entry (the line of Gs), but that doesn’t mean you feel like you understand the theme – more likely, that G-row still allows the puzzle to maintain an element of opacity/mystery/intrigue.
Yes, it’s ultimately up to the solver to make the mental leap of TOMORTOMORTOMOR → TOMOR-row → TOMORROW, and then extrapolating that trick in order to understand the previous themers. But the less difficult cluing makes it possible, at least.
Though, with that in mind: Why not make ALL of those theme clues easier, rather than just the last one?
“Only Connect” and Progressive Reveals
You might have heard about the British television series “Only Connect,” a trivia gameshow with a final-round game that the NYT puzzle game “Connections” is heavily, um, inspired by.
But a certain earlier round of the show has a different structure, where teams are tasked with guessing the theme category. To do that, they’re gradually shown clues of items from that category, which they can ask for more of one at a time. You’re allowed to “solve” the puzzle at any moment, but the more clues you need the fewer points you get.
More clues inherently means a greater quantity of information for the teams. But also notably, each subsequent clue tends to be more pointedly helpful than the previous one.
Similar to that Thursday crossword, there’s no revealer to explain what’s happening; that’s your job to solve. And that can be tough: One consequence of this show format is that many times, the contestants simply don’t solve the puzzle at all, which is not very satisfying.
But there are benefits to the difficulty as well. It’s super satisfying to solve the puzzle early, of course – you feel like an absolute genius. However, I’d say there’s also a pleasing element of reflection in looking back at a clue – that was previously incomprehensible – with new eyes and new knowledge. You get that little dopamine rush of feeling like you got smarter (even in an inane, crossword-smart way), because now you can understand something that you couldn’t 20 minutes ago.
And that’s a large part to why I think the NYT puzzle is successful: the order of the themers. I do enjoy the nice top-down progression, where the lengths of the repeated sequences increase (G → SOR → TOMOR). But more importantly, I also think that a result of this progression is that each theme answer becomes more approachable, a clearer illustration of the “trick.” Just as the TOMOR-row is easier to solve than the SOR-row, it’s probably more likely to solve the SOR-row than the G-row.
So there’s a twofold benefit: Yes, I definitely would have felt a little more like a genius if I understood the theme trick on an earlier themer (which I clearly didn’t). But ultimately, I’m satisfied that I had to do some work to understand the puzzle; and that the resulting knowledge I gained helped me understand the entire puzzle as a changed, slightly-smarter solver.
Final Thoughts
In my previous post, I focused on the huge importance of that last theme entry to create a satisfying puzzle conclusion. Obviously, constructors won’t place their other themers willy-nilly, but oftentimes I think the question becomes more about “are these entries symmetrical?” instead of “what is the ‘best’ order for these themers?”
But I think there’s more potential to explore this second question. First, as we saw from this above puzzle, there are ways to control difficulty aspects through the cluing, regardless of the actual order of theme entries. Plus, asymmetrical puzzles are becoming more widely acceptable, which offers increased freedom for ordering the themers.
Particularly for puzzles where the trick isn’t immediately obvious, this ordering feels even more important. Obviously, you probably still want the most-revealing themer at the end. But the “Only Connect” structure also seems like a nice way to think about a progressive reveal for solving the trick – where each themer is deliberately meant as an increased opportunity to solve the theme. Whether it’s through cluing choices or the entries themselves, creating a smoother difficulty curve for understanding the theme seems more mindful and rewarding to the solver.
Anyways, what do you think? Feel free to comment your thoughts; no snarky emails, please, unless you’re funny. And subscribe so you get a lil alert for the next post!
(And if you forgot to solve the puzzle I made based off these thoughts, now’s your chance!)