Written by the team behind Word War: Duel, a daily word game for iPhone
Most word puzzle leagues start in a group chat on a Monday and are dead by the second Friday. It's rarely because people stopped playing the puzzle. It's because nobody wrote the rules down, one person ran away with it, and whoever was keeping the spreadsheet got bored. Here's how to set one up so it lasts a season, and the four decisions worth making before anyone's ego is attached to a number.
Two to four weeks. Long enough that a single lucky Tuesday doesn't decide it, short enough that somebody who has a terrible first week can still see a reason to play in the third.
The version that fails is the open-ended one with a running all-time total. It seems fairer, because nothing is ever thrown away, and that's exactly the problem. The gap between first and last only ever grows, so by week six the bottom half is playing for nothing. Ending a season resets the gap to zero, and that reset is the whole point of having seasons at all.
For two people, counting daily wins is fine. Past that it's the rule that kills leagues, because only first place scores. If one person in a group of six is a bit sharper than everyone else, they take most days, and the other five spend the season competing for nothing measurable.
Averaging guesses fixes it without any extra bookkeeping. Solving in four still counts as a good day even when somebody else got it in three, so the middle of the table has something to play for, which is where most of your players are.
The four rules to agree on day one: what a miss counts as, what a skipped day counts as, whether hard mode is required, and how ties break. Each one gets discovered late, in the middle of an argument it would have prevented.
Weaker players don't quit because they're losing. They quit after the day they needed six guesses and watched it sit in the table for three weeks.
The cheapest fix is to drop each player's worst day or two before you total the season. No baseline data, no calculation, no argument, and it forgives the morning where somebody opened with the wrong word and never recovered. A per-player handicap based on their first week is the more precise option, and it's also the one that needs a volunteer to maintain it and gets relitigated every time somebody improves.
A miss counts as seven. It has to be worse than a six, and seven is the convention people already expect.
A skipped day also counts as seven. This is the one leagues forget. If a no-show scores nothing, skipping is free. If a no-show is dropped from the average entirely, skipping is actively rewarded, because a short average is an easy average.
Hard mode is all or nothing. If one player has to reuse revealed letters and another doesn't, they aren't solving the same puzzle, and the numbers in your table aren't comparable.
Ties break on days solved in three or fewer. You already have the data, and it rewards the sharper player rather than rerunning the same average.
Time zones and archive play cause the other half of the arguments. We covered both, along with the reasons two grids can look identical and mean different things, in how to compare word puzzle scores.
The admin. Somebody has to collect scores from people who forget to send them. This is the single most common cause of death, and it has nothing to do with the game.
One dominant player. Handled by averaging and dropped days, if you set those up before it's obvious who the dominant player is. Doing it afterwards looks like a rule change aimed at one person, because it is.
Latecomers. Somebody wants in on day eight. Either they start on the next season, or they join with the group's median score for the days they missed. Pick one now, in writing, because you'll be asked.
Every problem above except the dominant player is really the admin problem wearing a different hat. That's the gap we built Word War: Duel for, and we'll be upfront that this section is us describing our own game.
Everyone gets the same word each day. Fewest guesses wins the round, ties break on solve time, and the table updates itself whether or not anybody remembers to post their grid. A battle is the first player to win 10 rounds, and a war is the best of five battles, which is the season structure described above with the reset already built in. Teams score on the average of their members, so a bad day costs a round rather than a season.
You don't need to recruit anyone before you start. Every player has a Universe war running by default, which scores you against the average of everyone who played that day.
The honest version: a shared spreadsheet does all of this too. It just needs someone to keep doing it, every day, for the whole season.
Still deciding what to run the league on? There's a roundup of multiplayer word games that covers five others alongside ours.
Three is enough for the averaging rules to matter. Above about ten, daily chat updates stop working and you need a table somewhere everyone can look at without scrolling.
It's a real advantage in a chat league, because knowing the word took four guesses tells you something. Either agree not to discuss the puzzle until a set time, or use something that doesn't show you other results until you've played.
No. Word War: Duel is an independent game, not affiliated with, endorsed by, or connected to Wordle or The New York Times Company. It's a Wordle-style daily word game, and that comparison describes the format. It's the only claim being made.
Free to download on iPhone, with an optional subscription.
iPhone · free to download