Happy one year of letters! It's officially 12 months since I started my project of writing a newsletter every month. Here's a picture of me deep in thought, I think here I was pondering either Boggle or world domination.
Ava on Triple R
This month, I was asked to do an interview on Triple R about HTML Day: the website free-write I'm going to be hosting this August 8. You can listen to it here. You can find out more on the website https://html.beer/, but basically we are all going to Palomino Bar in Brunswick to write HTML code together. If you're curious about the reach of radio, I think we've had about 10 more signups since the show.
Boggle in Data
My game Beat Me at Boggle has been a much greater success than I was anticipating, which means many more players that I was expecting, which means lots of juicy (GDPR-compliant) player data to analyze.
The most obvious question of course being, was I beaten at Boggle? Yes, in fact. Of the 21,000 games that have been played, I've lost about 35% of them.
Although that stat isn't technically (☝️🤓) accurate, given the design of the game. Winning is obviously more rewarding than losing, so by default the game is actually slightly weighted against me. When the player finds a word that I don't find, they get a bonus point, but I don't get a bonus point for the inverse: when I find a word they don't find. If you account for that, I actually win close to 75% of games, but hey— who's counting?
Ava won (true) | Ava lost
████████████████████████████████░░░░░░░░░░░░░░░░░░ 64.24%
Ava won (technically) | Ava lost
██████████████████████████████████████░░░░░░░░░░░░ 75.20%Longest word
The longest word ever played is an equal seven-way tie between the nine letters words "timeshare", "reheating", "nastiness", "overshoes", (which was not a word I was familiar with), "aggressor", "tempering", and "chastiser". The first player to get a ten letter word will win a prize.
Most common word
The most commonly played word is "set". Entertainingly, "she" and "her" are the third and fourth most played words. It's good to know my players are feminists. Less so than me, apparently, as my most played words are "era", and "ore". Fun fact, if you find the word "era", you can also always find the word "are", because they are reverses of each other.
Most commonly played words
set | 1694 ████████████████████
tea | 1682 ████████████████████
she | 1549 ██████████████████░░
her | 1517 ██████████████████░░
rat | 1477 █████████████████░░░
A decision I made early on was to only use simulated Boggle dice on Monday and Tuesday, and a custom algorithm the rest of the week, because with real dice you sometime get dud games where it feels like every letter is a consonant. So it's interesting just how different the most common words are for Monday and Tuesday, where the biases of my custom dice rolling algorithm are on display. It really over-favours the letters R, S, T, and E. I've personally found some days recently to be monotonous, so it might be time to make some adjustments to the algorithm.
Most commonly played words (on Monday and Tuesday)
tag | 803 ████████████████████
lag | 511 █████████████░░░░░░░
bee | 494 ████████████░░░░░░░░
soy | 494 ████████████░░░░░░░░
gear | 488 ████████████░░░░░░░░
Most commonly played words (not Monday or Tuesday)
rat | 1328 ████████████████████
she | 1321 ████████████████████
set | 1310 ████████████████████
tea | 1274 ███████████████████░
her | 1237 ███████████████████░
When do people play?
My favourite chart is this one, which indicates how well people did, against the time of day they played, against the total number of games played. As you might expect, most people are playing during the day, with a sharp drop off between 2AM and 7AM.
What's also interesting is the sharp increase between the hours of 9am to 5pm. I cannot possibly imagine what activity typically goes on between 9-5 that could be causing that. It may forever remain a mystery.
As you expect, scores start gradually getting lower throughout the day, as people get more and more tired, before scores start plummeting at 2am. What's interesting is that even though the number of games is still low, the scores are very high at 5am. I imagine there's a group of early risers who are predictably very good at Boggle.
Where do people play Boggle?
On to the geographic data. This graph which shows the win rate per country. As you might expect, the Anglophone nations do the best, as well as countries with very high English literacy like Norway. There are a few outliers, like Afghanistan, which has a very dark shade of green for a country where only 6% of people speak English. That's because exactly one game has taken place in Afghanistan, where I was beaten at Boggle.
The graph below tries to correct for these outliers, by plotting the number of games played, in blue, against the win rate, in red. In this graph, you want to be purple. Purple countries have a high number of games played, and high win rate.
You really don't want to be blue, like Brazil is. That means you have lots of games played, but a low win rate. The same can be said for most the non-native English speaking parts of Europe.
I made this chart, to represent the Boggle nations by English proficiency. It highlights how amazingly well the Philippines, Mexico, and Croatia are doing, given their relatively low percentage of English speakers, and how the Netherlands and Sweden really should be doing better. The crown for the best Boggle nation then goes to Norway. You may not have won the World Cup, but this is surely almost as good.
Who's playing Boggle?
Melbourne, Australia | 559 ██████████████████████████████████████████████████
Sydney, Australia | 526 ███████████████████████████████████████████████░░░
Montreal, Canada | 142 █████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
Los Angeles, United States | 137 ████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
Brisbane, Australia | 129 ████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
São José do Rio Preto, Brazil | 127 ███████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
Dublin, Ireland | 114 ██████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
Auckland, New Zealand | 113 ██████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
Calgary, Canada | 107 ██████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
Toronto, Canada | 103 █████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
Goiânia, Brazil | 88 ████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
New York, United States | 84 ████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
Markham, Canada | 83 ███████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
Seattle, United States | 77 ███████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
Ogden, United States | 67 ██████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
Unsurprisingly, most of the Boggle games have come from my beautiful home city of Melbourne. Curiously, the second highest is Sydney, which may not seem strange, but I really don't know anyone from Sydney, and most of my web-traffic comes from the US. I believe this is due to the Australian population being a lot more concentrated in a few big cities (being Melbourne, Sydney, and Brisbane), where as the US is very spread out.
Who's the best at Boggle?
Elgin, United States (20 games) | 100.0 ██████████████████████████████████████████████████
New Windsor, United States (23 games) | 100.0 ██████████████████████████████████████████████████
Bowmanville, Canada (53 games) | 98.11320754716981 █████████████████████████████████████████████████░
Ithaca, United States (30 games) | 96.66666666666667 ████████████████████████████████████████████████░░
Sechelt, Canada (25 games) | 96.0 ████████████████████████████████████████████████░░
Maple Grove, United States (33 games) | 93.93939393939394 ███████████████████████████████████████████████░░░
Aurora, United States (15 games) | 93.33333333333333 ███████████████████████████████████████████████░░░
Middleton Corners, United States (26 games) | 92.3076923076923 ██████████████████████████████████████████████░░░░
Madison, United States (23 games) | 91.30434782608695 ██████████████████████████████████████████████░░░░
Overland Park, United States (22 games) | 90.9090909090909 █████████████████████████████████████████████░░░░░
Sanford, United States (21 games) | 90.47619047619048 █████████████████████████████████████████████░░░░░
Eisenberg, Germany (31 games) | 90.32258064516128 █████████████████████████████████████████████░░░░░
Bluffton, United States (20 games) | 90.0 █████████████████████████████████████████████░░░░░
Berkeley, United States (18 games) | 88.88888888888889 ████████████████████████████████████████████░░░░░░
Baltimore, United States (34 games) | 88.23529411764706 ████████████████████████████████████████████░░░░░░
Amazingly, there are plenty of cities with a 100% or a near-100% win rate. I predict this is just cities that are home to one GOAT who can consistently beat me at Boggle. Indeed, Madison, USA, is home to the true Boggle GOAT, who has the two highest scoring games ever, played across a single weekend.
Highest-scoring games ever
Madison, United States (2026-07-19) | 678 ██████████████████████████████████████████████████
Madison, United States (2026-07-18) | 636 ███████████████████████████████████████████████░░░
Bowmanville, Canada (2026-07-18) | 559 █████████████████████████████████████████░░░░░░░░░
Melbourne, Australia (2026-07-18) | 555 █████████████████████████████████████████░░░░░░░░░
Hildesheim, Germany (2026-07-12) | 537 ████████████████████████████████████████░░░░░░░░░░
They played 113 words in that game, meaning that in four minutes they played a word every 2.1 seconds. Unsurprisingly, it was a board with a lot of S, R, and E, which means that they could double or sometimes even triple-up on almost every word: turning "wash" into "washer" and then "washers", for example.
lassoer, washers, mashers, hoovers, showers, tamers, wheres, horses, homers, shares, shores, shames, washer, masher, barres, hovers, hoover, shower, tamer, hares, wares, where, heres, horse, hoses, sorer, sores, shoes, homes, homer, barer, bares, morse, mores, maths, meths, teems, shawl, share, arses, masse, mares, tares, sheet, shore, shame, lasso, barre, roses, overs, hover, halt, hams, lame, lams, tame, hare, wars, here, hose, roos, errs, sore, ores, shoo, shoe, hems, home, bare, more, math, meth, meet, teem, arse, mass, mare, tare, tars, rate, sate, shat, wash, lash, mash, hash, bros, roes, rose, over, show, wham, what, were, ham, lam, war, err, ore, hos, hem, tar, rat, sat, ass, bro, brr, bar, roe, law, haw, hah, who
So thank you to my dedicated boggler from Madison, but also thank you to all my beautiful boggleheads for playing. 21,000 games is so many more than I had hoped for. When we reach 100,000 games, I'll look at this data again and I'll release it publicly for anyone to have a play around with. In the mean time, keep playing sharing the game: I've added a challenge feature so you can ask that your friends beat YOU at Boggle.
I wanted to include a very special thank you to one particular boggler, my friend Ava, who helped me complete the task of removing every slur from the dictionary.
It was just me and her at the Ballieu Library, my laptop open. We would sit back in our chairs in thought, then one of us would perk up, and kind of excitedly but also kind of uncomfortably go "does it have this one?" and write the heinous ethnic or queer slurs into the search box. When we would discover that the slur was included we turn to each other and say, "Ooh, probably shouldn't have that one in there," and I would add it into the filter. This was a taxing process.
We had to make a couple of executive decisions on which words to include and which to disallow. For example, I decided to remove the term "wog". As someone of Greek-Australian heritage, I am a wog; and I'm not particularly offended by the term wog. I feel that in Australia in 2026, it is mostly used neutrally. I have even heard it used uncensored on the ABC. But, I have also heard that the word is incredibly offensive in the UK, where it is a very old fashioned slur for people of South Asian heritage. As an international game, I thought it best to remove it.
Also this month:
I helped facilitate a LARP (Live Action Role Play) around the Melbourne CBD, where we pretended we were in a dystopian future Melbourne. I carried a giant speaker on my back, as I played as the character known as Future iPad Baby. It was very silly and very fun. This was the Open House Melbourne, which we also did last year, but since then LARP has accrued a new negative meaning for someone who pretends to be something they are not.
What am I listening to?
100 Objects: Lowe’s Gas Bag
I have previously called this section of my newsletter "Music", but when I'm out and about, I mostly listen to podcasts. 99 Percent Invisible's series on American history in 100 objects is sating my Americophilia, and keeping my occupied on the tram. This episode on the humble hot air balloon, optionally known as the gas bag, is very good.
Audrey Hobert – Who's the Clown?
There's so much I want to tell you about this album, but I worry it's all just meme-adjacent rabble that I'm using to hide from an ugly truth. That this album is a guilty pleasure. Or that Audrey Hobert is a nepo baby that writes for Gracie Abrams. Or that I am very first unemployed person to listen to coworker music. The truth is, I like this kind of music. Corny overwritten indie-slop corporately-backed quirked-out 80/20-bisexual white girl music. So sue me.
Swapmeet – Mount Zero
In the past few months, I've become an avid Wikipedia editor. It was such a real joy to discover that no one had added Swapmeet's new album, Mount Zero, to the mountain's Wikipedia page. There is now an In popular culture section on the page, that says the mountain became a symbol of "in-betweenness and disconnection" for the band.
Finn Keane, Iiris – Be Your USA
I've been wanting to host an America-themed house party for a while now. The opportunity will come up, but then America will do something like kidnap the leader of Venezuela and the already gauche party theme becomes a little too politically incorrect. I missed the perfect opportunity for America's 250th birthday this past Fourth of July, where I would have surely played Be Your USA.