How do you know when your game is good?

Back in 2009, it was far from obvious that League of Legends had the potential to become one of the world’s most popular games.

There’s this story the OGs from the League of Legends dev team used to tell about the first moment they realized their game was good.

This was probably 2008 or early 2009—before the game’s open beta. LoL has never been a particularly welcoming game, but in those early days, even the dev team had to be forced to join mandatory daily playtests. Initially, there was a lot of grumbling about it.

But at some point in the process, something changed. People started playing the game not because they had to, but because they were having fun.

Suddenly people began playing even after regular playtest hours had ended. They were shouting more, laughing more, and staying late to play just one more game.

Moments like these are rare in game development. Because of how specialized and zoomed in each dev’s role is, you often work for years before you have any sense of whether all of a game’s parts are going to fit together.

But there are signposts along the way, little magic moments when you start to think that the thing you’ve been toiling over might actually be good.

I’ve started to wonder: What do those moments look like for other developers?

This week I reached out to some devs who were there for the critical magic hours during the creation of games like Apex Legends, Caravan SandWitch, and Age of Empires. I asked them: How do you know when your game is good?

Apex Legends — It’s All About the Feel

By the time Chad Grenier and the rest of the team at Respawn Entertainment began work on Apex Legends, Grenier had already spent over a decade shipping some of the most critically-acclaimed shooters ever made, including not just Titanfall and Titanfall 2, but the Infinity Ward-led Call of Duty games too.¹

This was a team that knew very well how to make a fun first-person shooter. So Grenier — whose FPS credentials stretch all the way back to the original CoD and who later became Game Director on Apex before founding his own studio—says the challenge was to make the game fun even when it was played in suboptimal conditions:

“The game was often fun because you were partying up with your work friends, chatting, and knowing who you’re playing against,” Grenier told me in an email. But “it didn’t simulate a real online environment.”

To force the team into a more creative mindset, Respawn employees embraced limitations. “We forced the team to playtest without parties, voice chat, and randomized gamertags,” Grenier says. That experience was in part what inspired the team to develop Apex’s acclaimed ping system for communication between un-mic’ed players.

Grenier says the Apex Legends team also heavily prioritized improving the “game feel,” which he describes as “the connection from the player to the game, via whatever control input device you’re using.”

“We spend a lot of time just moving around, strafing, snapping in and out of ADS, shooting dummy targets in a test map over and over again to smooth out the experience,” Grenier says, before listing a litany of examples he looks for when testing for game feel:

“Basically, we try to make sure the player never gets frustrated with the game itself,” Grenier says, “only by being outplayed by their opponent or choices they may have made. If a player gets annoyed with our motion model, collision, weapon state, or anything control related we have failed at making the game ‘feel good.”’

The focus on game feel makes intuitive sense, but what about harder, more scientific forms of evaluation? How do you balance your intuitive sense of game feel with metrics?

“Use metrics to learn where not to go, not where to go.” – Chad Grenier

“My philosophy is to use metrics sparingly, and in the right areas,” Grenier says.

“Never design a game based on metrics, design a game based on a vision the team is passionate about. Use metrics to learn where not to go, not where to go. Take the learnings of other games that have succeeded or failed before you, but don’t let them dictate your every decision.”

Caravan SandWitch — Fun Without Fighting

If designing a fun game is hard even for people who’ve been refining a formula for decades, how much harder would it be if you’re making a game that intentionally strips out standardized game mechanics?

That’s the challenge that Émi Lefèvre and the team at Plane Toast had to solve when designing Caravan SandWitch, a breakout hit adventure game that features no combat.

”To be honest we realized very late that the game had potential to be a success,” Lefèvre, the game’s creative director, told me this week. “For a big part of the development, the game wasn’t very fun to play and felt really clunky, but in the last six months before the release of the demo we started to feel like it was all coming together.”

Caravan SandWitch pitches itself as a “cozy” game, though unlike many that share the label, it’s structured more like a traditional narrative adventure game than the usual cozy farming or community sim. It’s got item-collecting elements, vehicle driving, and a sprawling world to explore, but no death or damage mechanics.

Lefèvre says that at multiple points in the game’s development, she found herself “struggling to explain the game and to communicate what it was all about,” she says. “I really thought we’ll never manage to sell it.”

In part, the challenge came from the team’s willingness to venture into rarely explored game design territory. They committed early on to a no-combat design, but Lefèvre says the team struggled with doubts about whether they could make a platforming/exploration game work without it.

“As you can imagine, we had to replace the combat with something else,” she says. “It took a while before we were fully convinced it would work, as the exploration and “treasure-hunt” kind of gameplay we got was very dependent on having a nice and interesting open world to navigate, and we had to rework the whole map multiple times to try different ways of approaching it.”

Much of the pleasure of playing Caravan SandWitch comes from discovering hidden areas of the game’s beautifully-designed environments.

The same challenges applied to the game’s various “slice of life” interactions with characters in the towns. Lefèvre says the game’s script and mechanics required multiple iterations before playtesters enjoyed talking to the characters.

Still, the Plane Toast team soldiered forward with iterations on the game. Lefèvre says the team took inspiration and encouragement from other combat-free games like A Short Hike and Alba, because “they were proof for us that this kind of game could exist, despite the third person adventure genre being so full of games with lots of combat!”

But it wasn’t until the release of the Caravan SandWitch demo during Steam Next Fest earlier this year that the team began to become more confident they had something truly special on their hands. Reactions from critics and players of the demo were extremely positive, giving the Plane Toast team more reasons to believe in their vision.

Now, with over 1,000 reviews on Steam and a 91% positive reviews average, the Plane Toast team has been fully vindicated. Lefèvre says she now feels more positive about the game’s development process, though she admits to some remaining uncertainty about the approach they took.

“It would have helped a lot if we had taken the time to do more market research and make a game that could have marketed itself,” she says, “But I also think it limits the authenticity you can put in your design. I’m glad we took a lot of inspiration from our environment and our life experiences and I’m sure people could tell how much of ourselves was put into it.”

The Age of Empires Series — Trust the Vision

Over the past 26 years, Greg Street has been a lead on countless iconic games, including World of Warcraft (where he was Lead Systems Designer),² on various projects as a producer and VP at Riot Games, and now at his new studio, where he’s cooking up a next-generation MMO.

But Street’s first adventures in the games industry were at Ensemble Studios, where he was a designer on every Age of Empires game for a decade.

Street says that it’s always hard to tell when a game has high potential—harder than almost any game developer is willing to admit.

“I think most developers, and nearly all stakeholders, are actually really bad at squinting,” Street told me in a DM via X. “They all say they are good at looking past the gray boxes, the bugs, the crashes, the ear flicks. But then they are grateful when a buggy playtest ends, or grateful when the whole playtest is cancelled, because it is a painful experience.”

Street says that over the years, he’s become increasingly convinced that even executives who’ve been in the industry for decades often fail to see a great game’s potential in the early stages—a fact revealed by the questions they ask:

“The questions are like ‘I know this isn’t the real art style, but will the real art style be a bunch of gray boxes? Will the knight unit always be a pony with the word KNIGHT floating over its head?’” Street says.

The executives may often be clueless, but for devs themselves, there are often ways to tell that you’re on the right track.

“Ensemble Studios playtested our games every single day,” Street recalls. “I think playtest was probably 4 o’clock, and back then we all met in a central playtest area versus staying at our own desks.”

Street says that for each of Age of Empire II, Age of Mythology and Age of Empires III, he can vividly remember the moments when the team began to prefer to play the new games over the previous one.

“You’d reach a point where just enough content had gone in, where the bugs didn’t completely detract from the experience, where game tuning went from wtf random to [a spot where] you generally felt smart for building the right counter.” When the team preferred to play the new game, Street says, “that was when we knew we had something good.”

Greg Street on Metrics vs. Creative Vision

After speaking with both Chad Grenier (Apex Legends) and Émi Lefèvre (Caravan SandWitch) I realized that they had a surprising commonality in their approach to game design: Both stressed the importance of trusting in your own creative vision instead of letting market research dictate your path.

I decided to ask Greg Street directly about this question, and his answer was so interesting that I’m including it here in full.

PUSH TO TALK: Much of big studio game dev is driven by market research over the gut feelings of creators. Any thoughts/rants you’d want to get off your chest on this topic?

GREG STREET: I am really stubborn about this view, which is that the vision needs to come first and it needs to be a really personal commitment to the vision. The vision can come from your knowledge of the market (“Gee seems like players are souring on battle passes. Hmmm.”) But this is not the time for market research. Once you have a design and ideally a playable prototype, that’s when you start the market research, and what you are doing at that point is validating your ideas, not casting around for ideas in the first place.

I was a scientist, so maybe no surprise I view the process very scientifically. You make a hypothesis. Then you gather data to see if it supports or can falsify your hypothesis. But you are the one with the intuition and experience and those should count for a lot. The market testing and user experience testing are to look for hiccups or things you might have missed.

Riot did a good job here IMO that I have tried to emulate with the new studio. Be very clear on who your audience is and then hyperserve that audience. You might be surprised that you attract more than your core audience. But if you don’t have a core audience and try to be all things to all players, then you’ve probably compromised your game.

Our new game is an MMO with an emphasis on other players. It’s possible players who love to play MMOs as a solo experience won’t enjoy our game. And we are fine with that.

Unexpected Agreements Between Wildly Different Devs

When I set out to ask devs the question in this article’s headline, I expected I’d get wildly divergent answers. There is little obvious in common between a cozy game creative director, a former Call of Duty designer, and an RTS/MMO legend.

And yet, their responses revolved around extremely similar themes.

How do you know when your game is good?

The answer is: You won’t. Not for certain, at least.

But there are some signs you can watch for, and ways to make forward progress: Playtesting and honestly appraising the ways players respond to your game, relentlessly iterating on the basic interactions that comprise game feel, testing your own hypotheses about what’s working and not working… these are tools you can use to make games better.

And if you hold true to a vision and work hard at it, there’s even a chance you might make something great.

That’s it for this week. I’m gonna go navigate the Dallas Fort-Worth airport while trying to avoid any combat encounters.