Ex-Riot Games, Bungie, and Blizzard devs reveal new game Project Loki

Former developers behind League of Legends, Valorant, Apex Legends, Overwatch, Halo, and Destiny are trying something new. Or, rather, an ambitious new twist on combining components of all of those games — plus even a few more, like Super Smash Bros. and Hyper Light Drifter — into a single game, which they’re calling Project Loki and are inviting more players to try out this week.

Project Loki, developer Theorycraft Games says, is a “squad-based hero battleground.” At a glance, it looks a lot like League of Legends. But keep looking and you’ll see influence from battle royale games, heroes gliding around a map like they just dropped onto Fortnite’s island, and white-knuckle player deaths that look ripped from Smash Bros. Unlike the slower-moving strategy of MOBAs like League and Dota 2, Project Loki moves quickly, as four-player teams battle each other on a huge map set in the sky. You’ll also notice they’re shooting each other; this is a game with a shooter’s “soul,” its creators say.

In a video from Theorycraft Games released Wednesday, studio co-founder Joe Tung — former executive VP on League of Legends at Riot, and former producer on Destiny and Halo: Reach at Bungie — says the team behind Project Loki is “out to make the deepest games in the world, games that are worthy of thousands of hours of play.” And it’s looking for more players to help inform them what’s working and what’s not in its work-in-progress PC game.

Tung says that early players who have gotten their hands on Project Loki describe it as “League meets Apex meets Smash,” a description that the team loves and “can only hope to live up to.”

While that soup of IP and genres may sound like an overly ambitious recipe, snippets of gameplay that Theorycraft has shown of Project Loki look promising. The studio appears to be borrowing bits and pieces from the most popular games of the past decade, weaving advanced traversal and battle royale mechanics into a game that Tung says will be “community-driven.”

Ahead of Project Loki’s reveal, Polygon spoke to Theorycraft executive producer Jonathan Belliss, the former game director for League of Legends, about the team’s influences and how their “squad-based hero battleground” morphed from more traditional MOBA to what it is now.

Read on for our full interview, which has been edited for clarity. If you want to sign up for Project Loki’s PC playtest, you can do so at Theorycraft’s website.


Polygon: Project Loki. Why are you calling it that? What did it communicate to the team and to potential players in terms of what you’re trying to convey here?

Jonathan Belliss: Project Loki is just a codename for now; we intend on giving it a more public name later. Internally, it means a lot of different things to different people. I do think that it signifies the chaos of an average game. There’s all these different things happening and you’re trying to thread the needle through the chaos, but there’s not too much behind the name because it is temporary for now.

What do you think differentiates Project Loki from the heavyweights on the market, like League of Legends and Dota 2?

I think most people, when they see it, they’re like, Oh yeah, this looks like a MOBA. But when you put your actual hands on it and you play it, it actually feels pretty different. A lot of our playtesters tell us that it’s more like Hyper Light Drifter or Enter the Gungeon — a top-down shooter with fine-grained WASD controls with aim-based combat. Less like how a MOBA controls. In a MOBA, you’re issuing a command to a hero and then they are moving on the battlefield. For us, you are that hero you are moving.

There are elements of MOBAs that are alive and well in our game. There’s a diverse set of character gameplay fantasies, and they’re all constructed in a way where you’re meant to team up with your friends and use them [cooperatively] and have big wombo combos — things that you’d expect from League of Legends or Dota.

There’s also a range of agency powers and abilities that are intended to be really expansive in terms of the possibility space, whereas I think if you look at other battle royales — a lot of them are military simulations/gun-based, so they’re a bit constricted — we have technology, fantasy, magic, and all these different things. We try to lean into that and give you a lot of agency on not only how you play, but how you affect the sandbox.

Artwork of the Skylands map and an airship from Project Loki

Image: Theorycraft Games

I wanted to talk about where your starting point was with the game. You’ve got this MOBA perspective, but you’ve also got these battle royale elements where you drop in from a transport ship, the glider elements that let you move throughout this big map. I’m curious how this kind of gelled for the team as a gameplay project.

The origin of the game was a bit more MOBA. The pace of combat was slower — it was more about intentional movement and strategy. A lot of the feedback that we got from our playtest was like, “Hey, this feels slow,” or “This feels really samey.” It was only in the last year, where a lot of what we thought was working was the action roots of the game — having it be faster and feel more like a shooter. So we started pushing in that direction. When we did, we saw a huge response from our players, and they further reinforced, Hey, this is fun. This is the right direction to go. A really crystal-clear example of this was [that] we actually didn’t have gliders for most of [Project Loki’s] existence, but somebody on our team — we have this thing called Freaky Week where every so often we’ll just do a week where we say, “Hey, put whatever you think makes sense into the build” — and somebody had put in gliding. The initial reaction internally was, This is really cheesy. You can just fly away from people! It just felt like now there’s no consequence to the Abyss we have — people falling [to their deaths]. Now that has no teeth. We were mixed internally [but] when we playtested it with our audience, they were like, My god, traversing the world is so fun. It was so overwhelmingly positive that we were like, Are we wrong? Do we just not know it? It was really our playtesters that kind of nudged us in the direction that brought us to where we are — a lot more of a hybrid.

I wanted to ask about your heroes. [In the reveal video for Project Loki,] Joe talks about totally unique builds for heroes and I want to ask what that means. How do the heroes work from a gameplay perspective?

All of the heroes, probably the most common throughline with all of them, is that each one of them has a kind of “shooter soul.” If you’re familiar with Halo, there’s a gun called the Needler, and there’s a hero in our game that’s kind of like an ice-empowered magical girl. She shoots these little needles that kind of behave like the Needler. We have another hero called Rocket Girl, and the soul of that [hero] is, What if it’s Quake [3] Arena, but top-down, with rocket jumping, grenades, etc.? All of our heroes have some sort of primary fire that if you’re coming from a shooter world, [they] make sense, but then the rest of their kit is a MOBA kit.

We have another system that we call our power system. If you play a game like League of Legends, they have this concept of summoner spells that you can add to your hero. In Loki, those are all random and placed throughout the world. You find them either in vaults, or you kill monsters or other players to get them. They combine with your hero kit in really interesting ways. We have this hero called Oathbreaker, who’s this kind of robot paladin with a big hammer. One of his weaknesses is it’s hard for him to kind of get up close to somebody — but if he does, he’s quite devastating. There’s a power called Grappling Hook that you can use to grab and get close to somebody, or there’s a power called Wings that lets you fly and move really fast. There are also other powers that don’t necessarily combine with hero kits but combine with themselves. An example is a power called Money Tree, which lets you just throw down a seed and then after a couple of seconds, a tree grows. Some players use it to block collision or block skill shots. If you kill the tree, there’s some money that falls on the ground, which is also nice. But there’s another power called Timber, [which] is basically a throwing ax, and if it hits a tree, it knocks that tree over and stuns anybody on the other side, so you can actually combine those two together. You can create your own Rube Goldberg machine, effectively with just two powers, or you can combine them with your kit in really interesting ways, too.

We really like to have this attitude of “yes, you can” whenever we’re designing things in the game. [Grappling Hook can] grab enemies […] but you can also grab monsters, you can grab your teammates, you can grab anything that can be grabbed basically. So it kind of creates this higher level of emergence and unexpected outcomes that we think are super fun. You and your team, you’re kind of like scientists in a lab.

You [also] gain abilities as you would in a MOBA. At a certain level breakpoint, you have a set of random passives that become available to you. Kind of like Teamfight Tactics, where you’re given kind of random modifiers across your entire kit. We do that as well. An example might be [that] anytime you destroy an item, you get 500 gold, or when your glider runs out of gas, it starts consuming mana, so you can glide farther than another player. You can combine all of these passives with powers and your hero kit to create really interesting [strategies]. Sometimes you’ll be playing a game, you’ll be like, I can read the tea leaves. I can see this combination that’s appearing before me. Do I take it?

Artwork of a fantasy sniper shooting at a pair of enemies approaching via glider in Project Loki

Image: Theorycraft Games

On the hero side of things, you’re clearly trying to differentiate yourself from some of the MOBAs, but what have you wanted to kind of iterate on and innovate there from their design, personality, and characteristics?

In terms of character theme: representation. We have a lot of folks who have worked on both Valorant, League, and Overwatch on characters. We want to make resonant characters for a global audience, that appeal to a wide variety of different player types, and players with different motivations. I think the thing that we focused on the most, though, is core gameplay. For all of our hero kits, we have always started from a very low art fidelity level where we have this mentality that sometimes art can lie to you about whether or not a kit is good and fun. So we really try to focus and make sure that at a very, very basic level: Is this fun to play?

Part of the challenge there is each one of these heroes needs to have a shooter soul and the more nuanced meaning of that is when you put your fingers on the keyboard, it needs to feel really good. We’ve had hero kits where maybe we have a really interesting utility, or this combo is really interesting, but if it doesn’t feel good to left-click your mouse and shoot a monster then that hero struggles, and we see players not play that hero. Our core focus has been, How do we get a shooter hero to work top-down? That’s been really difficult in a lot of cases.

In a MOBA, it’s a lot easier to get melee heroes to work. MOBAs are a little bit slower. Typically they give you a lot of tools to move around. When a melee hero gets on you and sticks to you, it can be really oppressive. And so I think that’s been a big focus for us too. Which is, How do we get heroes to feel like they’re fair, but also feel like they’re really fun to play and really engaging in a top-down shooter environment?

What have been the difficulties of adapting shooter mechanics to translate to a top-down perspective? You mentioned Enter the Gungeon earlier; are there, like, north stars for you from a top-down shooter genre that you’re looking at to say, We want to make sure that we’re at least matching this or We’re taking inspiration here?

There are a lot of top-down shooter games that are an inspiration for us, but the reality is, a lot of them are either single-player or co-op. They’re not competitive multiplayer. A lot of them are also 2D, and something that we’re pushing really hard on is being 2.5D — having some level of verticality. The 2.5D component has probably been one of the most challenging pieces. A good example of this is when you see somebody gliding up in the air, you have to ask yourself: Do I shoot at the person that’s in the air or do I shoot below them where they should be? When we introduced gliding this is something we had to solve for. A really kind of clever solution that a VFX artist on our team came up with was when somebody’s gliding up in the air, why don’t we just have an indicator that’s on the ground roughly where they’re supposed to be and it acts as kind of a bullseye. If you want to hit this player, you just shoot at this location here on the ground, it’ll bring them straight down to the ground. It is a little gamey, but it really helps competitive clarity, and it really helps players understand how our 2.5D works. If we don’t have bits like that, it ends up becoming more frustrating than fun.

Obviously people are going to look at some MOBA comparisons here, but I wanted to check in on whether certain traditional elements like creeps, last hits, and towers are elements of your game, or is that not really relevant here?

One meaningful difference between traditional MOBA and our game is that MOBA is kind of lane-based combat. If I say, “Hey, Michael, let’s go play a game of League,” what I’m really saying is, “Hey, for 20 minutes or so you’re gonna be in top lane, I’m gonna be somewhere else on the map, and I’ll see you at some point. I hope you do well.” [laughs] Don’t get me wrong, there’s a bunch of really meaningful and interesting skill checks, and 1v1 that happens there. In Loki, we instead say, “OK, you should be playing with your team the whole time.” A lot of our game is designed to encourage you to be with your team the whole time. In a game like League or Dota, It’s actually optimal for you to split up and farm in different locations, because you want to maximize the value that your teams are creating. A lot of our abilities and the way that things work encourage group play. If you’re around a minion and it dies, you get the money, you don’t have to last-hit it. Especially when you’re in a group of four people all hitting minions, it can be kind of chaotic. We didn’t want to create competition between friends.

There is a notion of bases. There’s no towers or anything like that, but scattered throughout the map, there are these bonfires — it’s a capture point. If your team captures it, you have a little shop that you can buy stuff and repair armor from, and you also get vision in an area that’s around it. So you get some semblance of safety. Also, your team can recall back to that location if they’re spread out throughout the map. There’s no laning, no towers, no last hitting — the skill check that we really want to test players on is How do you and your friends play together and adapt to the chaos of the game? The skill test in our game isn’t fighting the same team over and over. It’s When do you fight? And where do you fight? Which I think is maybe not novel for a battle royale player, but is more novel for a MOBA player, or even a hero-based shooter player.

You talk about early players informing the gameplay decisions that you made, and Joe said you’re “willing to go to war with the industry for our players.” I want to know what that means, because obviously you can’t satisfy 100% of your players, right? You talk about being unapologetically deep but you also don’t want to make a game for 200 people. You want to make a game for a lot of people. What’s your approach to evaluating your community, integrating their feedback, and is it different from your approach on previous games?

A lot of my core responsibility right now is, how do we simplify the game systems that are in our game in a way that maintains the depth but at the same time is accessible that a broader audience can play? We have a lot of experience from operating other competitive games, and I think that we’ve seen different examples of what good and bad competitive depth can look like. […]

Where we believe we are now is that most players can walk in, put their hands on the game, and say, “Wow, this feels pretty good.” On top of that, we try to make the strategic game that you play simple enough to where you understand what you’re supposed to be doing, which can be really complex in a MOBA. You come in and you’re like, My god, there’s a hundred different heroes that do different things and I have to learn about last hitting but at the same time, this person’s harassing me. It can be really stressful. So we try to take an approach where we have a different type of pacing. Again, the battle royale format helps us a lot here. If you want to just have a chill run, just drop [onto the map] somewhere where nobody else is, then go farm and learn how to play the game — engage with the systems at your own pace.

A lot of us have worked at really big companies, and when you work at really big companies you focus on a lot of different things. We just want to focus on one game and make this one game as good as it can be. That’s really hard in some of those other environments — and it’s no knock on those other companies; they’re trying to accomplish things that are much larger than an individual game. But a lot of us are just really, really passionate competitive gamers. And we just want to focus on making a great competitive game and nothing else.

What are you thinking about in terms of your monetization for the game? How do you plan on doling out heroes?

We’ve thought a lot about it, and a lot of it is based on what we think works in MOBAs and some other shooters: Just offer a bunch of cosmetics. It’s all just personalization content. It has no impact or bearing whatsoever on gameplay. We also have a very strict attitude when it comes to competitive clarity: There should never be a cosmetic that gives you any kind of advantage whatsoever. We actually say it’s better that you pay to lose, we’re fine with people paying to lose, it’s OK if you buy a skin that makes your grab skill that much easier to notice. That’s fine, but pay-to-win is something that’s kind of allergic to who we are as a studio.