Last Place Wins: Inside Centumth

Centumth started as something you did with friends. You’d pick a topic, write a Top 100 in secret, and they’d argue their way down it. What was one of the first topics you ever did that with, and what was the moment it stopped feeling like a thing you did at a table and started feeling like something that should exist online?

The first topic I ever did was “Top 100 most streamed artists on Spotify.” It was an easy one to come up with a list for, so it was a good way to test the idea.

I have a solid group of about six friends that I get together with from time to time. We mostly play D&D (Dungeons & Dragons), and after we finish a session we usually end up playing different games. One day I brought Centumth to one of those nights and it just worked really well. There were a lot of debates, everyone was trying to convince each other that their answer had to be higher or lower, and I actually really enjoyed being the person hosting it.

At the same time, I’d always been a big Wordle player and had wanted to make a daily game of my own. So eventually it just clicked that I could take this game we were playing at the table and turn it into a daily game.

I already had experience building websites, so making a rough prototype and putting it in front of my friends was pretty easy. Once I saw that they were still having fun with it outside of the original game-night setting, I thought, “Okay, maybe there’s actually something here.”

And yeah, that’s basically how Centumth started.

The scoring is inverted from how ranked lists usually work. The obvious answer at #1 is worth almost nothing and the deep cut at #100 is worth the most. Was that the rule from the beginning, or did you try scoring it the conventional way first and find it boring?

It was actually the rule from the beginning. The whole idea of Centumth was built around making the obvious answer almost useless.

If you get the number one answer, that’s great, but it’s not really impressive. Everyone knows it. The fun is when you somehow pull out the answer sitting at #73 or #96 that you didn’t even know you knew.

I also wanted people to have those moments where they answer something and think, “There’s absolutely no way that’s on the list,” and then it turns out to be #87. That’s the feeling I was trying to build the whole game around, so the scoring naturally followed from that.

You mentioned that finding fresh topics everyone can participate in was harder than you expected. What’s a topic you loved but couldn’t make work for Centumth, and what broke it?

Yeah, this is probably one of the hardest parts of making the game.

There are a lot of topics that would be absolutely perfect for Centumth in theory, but then you realize there simply isn’t a reliable Top 100 list for them. A lot of these are actually geography-related. You might have a really interesting idea, but when you start researching it, you discover that there are only 30 or 40 meaningful answers, or that the rankings become completely arbitrary after a certain point.

The other problem is that I always try to find topics where the answers aren’t too niche. Ideally, almost everyone should be able to think of at least one or two answers, even if they have no idea where those answers rank.

So there’s a bit of a sweet spot: I need enough answers to make a genuine Top 100, but I also need the list to be something that a normal person can actually play.

In July you posted that too many lists felt subjective or rested on thin sources, and that every Top 100 would get a stricter research pass from August. Two parts: what does that pass actually involve now that it didn’t before, and was there a specific list that made you decide to change it or was it general feedback?

This was definitely a lesson I learned the hard way.

Early on, I would sometimes find a random Reddit post or forum list and use it as a starting point. The problem is that those lists were often just someone’s opinion, or they weren’t based on anything particularly concrete. That doesn’t really work when you’re asking people to argue about whether something should be #43 or #67.

There was one list in particular that made me realize I needed to change things: “Top 100 first words babies say.” I remember getting a bunch of messages through the feedback form that day, and let’s just say they weren’t exactly positive.

Since then, I’ve made the research process much stricter. Every list now goes through multiple checks and gets proofread before it goes live, and I also have two of my friends verify the lists independently.

They’re working for free thankfully… for now.

You’re upfront that sources are the starting point and that the final order can include small editorial calls so the puzzle stays balanced. Most games would hide that behind the word “algorithm.” How often do you overrule a source, and what does balanced mean when you’re the one deciding it?

I actually get asked about this quite a lot.

The simplest way I can explain it is that I want the #100 answer (the Centumth) to be something that a player could realistically arrive at if they think hard enough. If it’s something completely obscure that nobody could possibly guess, then the game stops being fun.

A good example is the “Top 100 loudest sounds in the world” list. If I followed the source completely literally, ocean creatures and underwater sounds would take up a huge part of the list. Technically, that might make sense, but most people aren’t going to know anything about those sounds, so it would make for a pretty terrible game.

So I removed the ocean sounds and actually mentioned that in the description for that particular day. That allowed the list to have a much wider range of things people could think of.

And the #100 answer ended up being “a normal conversation,” which I thought was a really fun Centumth. It’s the kind of answer that sounds completely wrong at first, but if you think about it, you can get there.

For me, “balanced” basically means making sure the list is still faithful to the topic while giving players a reasonable path to those really obscure answers.

The archive pages publish the complete ranking for anyone who came looking for the list rather than the game, and revealing it locks that puzzle for good. You built a door that lets people skip the thing you made, and then made it one way. Walk me through why that’s the right tradeoff.

That’s actually a growth strategy.

I started noticing in Google Search Console that Centumth was getting a lot of traffic from searches like “Top 100 X.” So I realized there were people who were specifically looking for the list itself rather than the game.

Instead of trying to force those people into playing, I figured I might as well give them exactly what they’re looking for: the complete list, immediately available.

It does mean they can skip the game, but I think it’s a worthwhile tradeoff. I’m still getting the search traffic and fulfilling what that person was actually looking for, while also putting Centumth in front of a bunch of new people who might end up playing the game anyway.

And once someone has seen the list, that puzzle is obviously no longer playable for them, so I think having it be one-way makes sense. They got what they came for, and there’s always another daily puzzle waiting for them.

You’ve said player feedback is why the lists got stricter, which suggests people do write in and argue. Has a player ever actually won one of those arguments and changed a ranking after it was published? What was it?

Honestly, not really. Most of the time I’m the one trying to convince them of why something belongs where it is.

But player feedback has been extremely important to the game.

One good example is the arcade mode. I never would have thought of adding it myself. The original idea was that you get your five official guesses and that’s it. But I started getting a lot of players asking for a way to keep playing after they finished their five guesses, so I added an arcade mode where you can keep guessing as much as you want.

People really liked it, and it’s now one of those features that feels like it should have been there from the beginning.

So even if players don’t necessarily win individual arguments about rankings, they definitely influence how the game evolves.

Your about page lists Mellow the cat as head of morale. What does that role actually involve on a day you’re deep in research for a list?

Mellow is actually very grown now compared to the picture on the website, but I don’t think I’m ever going to change that picture because I love how tiny he looks in it.

His main responsibility is basically jumping onto my lap while I’m working on a list, purring, and demanding to be petted.

Honestly, it works. I’ll be deep into some completely ridiculous research trying to figure out whether something belongs at #64 or #72, he’ll jump onto my lap, and suddenly I’m in a much better mood.

So yeah, I’d say he’s doing a pretty good job as head of morale.

You recently launched Majority, and one of the Reddit comments said it needs to be a physical card game. Given that Centumth started as a game at a table with friends, that seems like it could go either direction. Did Majority start on paper too? And now that it’s out, what have players told you that you plan to act on?

Majority actually didn’t start on paper. It’s an idea I’ve had in my head for a while.

I’ve always loved watching things like Family Feud, and Majority is basically the polar opposite of Centumth in a weird way. Centumth is about finding the answer that almost nobody else thought of, while Majority is about understanding what everyone else is going to think. So it kind of scratches two completely different parts of my brain.

After Centumth started taking off, I also really wanted to start growing a portfolio of daily games. I genuinely love making these little games, and I have way more ideas than I have time to build, so Majority was always something I knew I wanted to make eventually.

And it’s actually been getting some really good traction since launching, which has been really nice to see. I’ve already had people suggest turning it into a physical card game, and honestly, if it keeps growing, I’d definitely consider it. It sounds like a lot of fun.

For now, I’m mostly focused on watching how people actually play it and listening to what they want changed or added. That’s probably the biggest thing I’ve learned from Centumth too: you can spend ages imagining how people will use a game, but once you put it in front of real players, they’ll immediately show you things you never thought about.