My experience making a Wordle non-clone

A year ago, I launched wordhuddle.com. It's a hangman / Wheel of Fortune game, but with a unique scoring system. (In fact it has had two unique scoring systems, because I decided that the complexity of the initial scoring system was too complicated.) So it's not a Wordle clone, but it does follow the magical quick-daily-puzzle format.

The results have been depressingly like my earlier attempts with native mobile games (from about 2009 to 2015, all done as weekend/evening side projects) -- the result being that lots of people say they love the game, and there are a bunch of dedicated players, but it's hard to grow that number to anything really significant.

The number of players is remarkably steady, at about 500 players (that is, 500 finished games) per day. Here is the history:

I launched the game because I wanted to try my hand and the daily puzzle format. I was thinking about a number of projects, and couldn't decide which to start, so I just went ahead and started this game since it seemed like it might not take that much time, and at least I'd be doing something. I had done some trivia games in the past so I felt like I could think of good puzzles and good clues, which I feel is accurate.

My initial design was you get a certain number of ticks, and each time you guess a letter not in the puzzle, you lose a tick. So it's exactly hangman, without the violent imagery. But I didn't like that the goal was just to keep guessing letters until you had guessed all the letters in the puzzle. I wanted to reward the player for solving the puzzle, for knowing the complete answer as early as possible. So I made the rule that a solved letter (that is, a letter you get right when putting in a solution for the full puzzle) is worth 10 points, but just guessing a consonant is worth 2 points, 1 point for a vowel. I set a goal score called "par". So if you just keep guessing letters, even if you finish the puzzle and have few or no missed guesses (guessing a letter not in the puzzle), you won't reach the par score because all those 1 and 2 point scores won't add up to enough. To win, you have to figure out the puzzle with some letters not yet revealed.

That was the scoring system for quite a while. During that time I put in a "Rate Today's Puzzle" button. The feedback from that has been extremely helpful. Among the various feedback I got, it came through that a lot of players were not really understanding the scoring system. So I kind of fixated on that as the reason the game wasn't continuing to grow. By this time, I had announced the game in various places, including the Wordle group on Facebook, who are very friendly and allow makers of other games to post about them. After my first post on that group, the number of players shot up from about 200 per day to 600 per day, but soon settled back at about 300 per day, and since then has risen to about 500 per day, and has been stuck at 500 for a number of months.

So I was focused on the idea that the scoring system was confusing to people, and turning them off. I modified the game so that there was no longer a separate step of "solving" the puzzle. Rather, you get 10 guesses, and each time you guess a letter you lose a guess, except if you guess the next non-guessed letter in the puzzle. You can see the effect at wordhuddle.com. The spinning ring shows the first unguessed letter, and if you can guess it, you "save" a guess. Thus, when you know the solution to the whole puzzle, you can just type every letter in, in order, and you finish with a perfect score of 10 unused guesses. I realize this still sounds a bit complicated, but I was pleased that it was a way to simplify the game by getting rid of the "solving" phase, and by getting rid of the arbitrary seeming "par" system, replacing it with a simple 0 to 10 scoring system. (I liked that it was similar to the 1 to 6 scoring or Wordle and the 0 to 5 scoring of https://wafflegame.net/.)

The result of this design change was a veritable mutiny among the players. I got so much negative feedback. But some players did appreciate the new system. Furthermore, I was feeling that something had to change because the game wasn't growing, so even if this change lost half of the players, it would just be a drop from 300 to 150 and it was all insignificant. So I left in the new system, but ended up creating a back door to the old game (see the violin icon in the game). That at least stopped the hate mail. And a certain number of players have since apologized, saying that they didn't appreciate the elegance of the new system. Lesson learned: offer people a new version, but don't just suddenly replace the old version.

The other problem with the game is that the puzzles take some time to create. It takes me about 10 solid hours to make the 30 puzzles for each month. At first, of course that was no problem, but after a year it gets tiring, as I try to grow this game in the background while doing other projects, all in parallel with my full-time job. I've gotten faster, and I've gotten better at making good clues, so it doesn't take as many passes over the puzzles to make them interesting, not too hard, and not too easy. But another problem is that there are only a certain number of well-known sayings in English (sayings, adages, aphorisms). Early on I started to add idioms and expressions, not just sayings. For example I've made puzzles for expressions like PLAYING DEVIL'S ADVOCATE, THROWING GOOD MONEY AFTER BAD, TAKING A CURTAIN CALL and many, many others. One big problem was that it seemed like any time I used a piece of "trivia", such as an actor, movie name, song title or lyric, I would get complaints that some sizeable number of players had never heard of it. From the beginning I have had a lot of international players -- no matter where you're from, to be a good English speaker you need to know the idioms and expressions, but you do not need to know Talking Heads was a popular band in the 70s and 80s, that Judy Garland was a popular actress, or that there was a movie called Ghost Busters. In summary, over time I have "used up" a lot of the interesting sayings, expressions, and idioms, making new puzzles harder to create. Recently I have started putting in some Rhyme puzzles (similar to twofergoofer.com) and some of the old Wheel of Fortune wordplay style puzzles. So this gives me a new pool to draw from, but it also starts to make the whole thing feel like a different game.

In terms of promotion, as I said I introduced the game on the Wordle Facebook group, which I actually did a few times. I also sent the link to game bloggers, reddit puzzle groups, and YouTube puzzle reviewers/influencers. I did get one nice article written (https://www.express.co.uk/entertainment/gaming/1628460/Wordle-Wheel-of-Fortune-new-Hangman-game-WordHuddle-puzzle).

With the puzzle experts, I got a lot of good feedback. One person in particular summed it up by saying the game is fun, but at the end of the day it's just hangman / Wheel of Fortune.

I have been considering a number of enhancements including:

• The ability to create your own puzzle, and send it as a link to a friend (or share it with the whole player community).
• The ability to make a group of friends that play against each other and chat.
• Offering puzzles in different categories (so for instance those people who do know all about movies and music can play puzzles focused on that kind of trivia).
• Offering the ability to play past puzzles.

All of those ideas at one point were also things I considered as premium features that players might pay one or two dollars for each month. But at this point it's not clear to me that even if I offered all of those for free it would make a significant difference in the player count.

So I find myself coming back to the idea that, yeah, at the end of the day it's just hangman, and even though I put a lot of love into the puzzles, images, explanations, and overall design, nothing is really going to move the needle much. I sometimes feel like instead of trying to make better puzzles, or add the premium features listed above, I should try a daring stunt, such as making AI videos of Donald Trump expressing his grievances about that day's puzzle, how unfair it is, and similar ideas with other celebrities (AOC for instance would generally much more positive about the puzzles).

And then I start to think of other game ideas. Two things about Waffle stand out to me. One is that I read an interview with the developer who said he never really promoted it. He just mentioned in once on the Wordle Facebook group and it took off from there. The second thing is that, with friends and family, I have tried the experiment of having them play my game and then play Waffle. I've noticed that when they sort the last letters to complete the Waffle they often have a big smile on their face. And maybe it's as simple as that...