The Design Philosophy of Battletech

I do a lot of my design work by gut. I intuitively grasp whether something is going to be fun or not; whether something is a good idea or not; whether something is going to be worth the time to develop or not.

In the moment, I can’t explain why something feels right or not, but that doesn’t mean there isn’t a lot of thought and analysis behind each design decision. I’ve just done that analysis well in advance, building a kind of foundation of good design practice, experience of game systems and mechanics, and conversations about what is and isn’t fun.

So it’s often not until well after the fact that I’m able to articulate the reasons for some of my ideas. This is frustrating for me, because it makes it hard to argue for what I know to be the right answer to a problem, and frustrating for the people I work with, who want more to go on than ‘Kiva feels it really strongly’.

I’m far enough away from both Battletech and my frustrations and anger about it that I think I can start to explain why some of the things in the game are the way they are.

The First Principle: A Game About People

I didn’t want to make a game about robots. Frankly, the Battletech IP was my least favorite part of the project, and I nearly turned down the job because of it. Robots don’t interest me, people do.

I gave a talk many years ago at a game conference about whether games are art. My conclusion was that they aren’t art, but they could be. The reasoning isn’t really important here, but the central thesis is: games become a distinct art form when they learn to leverage their unique properties, those elements that make them distinct from paintings or movies or novels or television. The art form of games is interactive experiences.

So when I make games, I’m not looking to tell you a story. That’s better done in other media. I’m not looking to show you a cool movie or an exciting cinematic. I want to give you an experience you can only get from games. I want to give you the tools to tell your own stories, with my game as the context for those stories.

I used to make jokes about wanting to make you care enough about your pilots that you’d be devastated if they died, and I got a reputation for pilot-murder, but the truth is that I didn’t care if they died, specifically; I just wanted you to care about them. I wanted you to feel like you’d followed these people, these complex messy people, through their adventures and their hardships. I wanted, at the end, for you to look back at your roster (and your memorial wall) and think about who those people had been, and remember moments of amazing success and crushing defeat.

My favorite moments of games like XCOM were always those where a soldier became a person — when they elevated themselves above the rest of the team and made me notice them. My favorite moments of Crusader Kings were always those where I learned something new about who the people in my court were — when their weird quirks and personalities shaped the course of the game in unexpected ways.

I like games where afterwards I want to tell other people the story of the people in my game.

Battletech is a game that was designed, from the first, to be about the people on your ship. The people you hired, trained, fought with, and mourned. Battletech was never the story of Kamea Arano, or Darius, or Santiago, or any of the other named NPCs. It wasn’t even really the story of your commander (something I’ll return to later). It was the story of Glitch and Behemoth and Medusa and every other weird misfit you hired to drive those big stupid robots.

Every time I’ve read someone talking about how they’d always save-scum to save Glitch, it’s made me smile. She’s not unique; her stats are mediocre and her voice acting can be found on dozens of other pilots. But for whatever reason she mattered to you, and that mattered to me. It’s what I wanted from the game; it’s why I made it.

I’ve told people before that the heart of Battletech is really a romance game, and while that’s a simplification, it’s basically true. Battletech began life in my brain sometime around 2008 as a concept for an anime high school sim. It evolved through a number of incarnations, becoming a pitch for a Harry Potter casual game, the inspiration for the Bakugan game I designed, and a Facebook game I created about dark fantasy mercenaries. The thread that links all these things together is the same: you shepherd a group of people through hardships, and in doing so, you learn and shape their stories.

The Roster

On the Argo, the barracks is always bigger than the mech bay. This is intentional. You have more capacity to hire and maintain pilots than you do ’Mechs. From a practical standpoint, injuries are brutal and will take your pilots out of the action long after the robots have been repaired and readied for battle, so you need extra pilots available to pick up the slack and keep the company in the black.

But truthfully, you have more pilots than robots because I didn’t want you to associate a specific pilot with a specific robot. That turns the pilot into a piece of equipment on the robot, not a person. That strips away the pilot’s distinct individuality, and replaces it with their stats and skills.

When Glitch is out with an 80-day injury, and you have to slot some chump named ‘Pamcakes’ into the Vindicator instead, you stop thinking of Glitch as the particular robot with the big gun, and start thinking of her in the medbay, out of action, while Pam racks up kills in ‘her’ robot. When Glitch comes back, you’re now trying to decide between her and Pam when you go out on a mission, because they’re both good and you’ve gotten used to both of them being around.

When you hire a new pilot, it’s always cheaper to hire a rookie and train them yourself. This isn’t by accident. I want you spending time with these people, thinking about who they are and where they’re going. I want you to be guiding them, because when you level their Gunnery you’ve got a commitment to that specialization in a way that hiring High McGunnery from the hiring hall doesn’t provide. And if that pilot ends up in the medbay, you’ll feel every one of those decisions about experience investment and you’ll think about every mission you’ve taken them on to get them a few more points.

Every pilot you hire has a lifepath. I spent way, way too long on this system, and it’s easily the most overdesigned system in the game — eclipsing even the Plot system that allowed maps to be re-used in almost any context. I sunk that time into the lifepaths because I believed (and still believe) they’re the absolute minimum necessary feature set for me to convince you the pilots are people.

I don’t know if anyone outside of the forum hardcore ever explored the complete extent of the lifepath system. Every pilot you ever see that isn’t a special backer pilot has their own unique life. The jumps from one career node to another are non-random; every transition tells another story, of promotion and demotion, of failure and triumph, of crime and prison and escape.

Once, I hired a woman who started as a criminal, joined a pirate crew, was captured and sent to prison, joined the Canopian navy after her sentence was over, and then jumped right to ‘pirate captain’. My narrative for her was that she never stopped being a pirate; she just bided her time until the right moment, and then seized the ship she was assigned to and returned to her pirate comrades triumphant and firmly in command.

She was a person. When I hired her, she felt like someone who could be the main character of her own game. But she was just one of eight pilots I had around, each of whom had their own comparable story (though none quite as extensive as hers; the lifepath system could spit out green novices as well as leathery veterans).

The lifepath gave every character a handful of tags which served as narrative hooks for the event system to use to continue their stories. A criminal would be caught up in criminal stories, figures from their past, black market opportunities, and so on. These tags and their effects were, for a few reasons, never really explored to their full potential; I will probably explain why, and what I did to try and address that, and what the (fairly dire) consequences of that attempt were, in a later post.

The Medbay

We went around and around about injuries and how they’d work. I tasked another designer to figure out how pilots could take damage, how often they’d take damage, and what the consequences in battle would be. Meanwhile, I tried to develop the simulation side of injury and recovery, treating combat as a black box. In retrospect, this was a mistake, but frankly we had too much to do and not enough people to do it, and I couldn’t possibly tinker with every piece of design myself.

In my plans, a pilot coming back from battle with ‘minor’ injuries would be stuck in the medbay for weeks, at least. If the injury was enough to make a difference in battle, it had to be something serious and nasty — a broken limb, a severe burn, a concussion. I hoped for moments where you’d have to consider whether you could dump an injured pilot off at a nearby planet and replace them, rather than carry them through a hundred days of downtime. If I’d done my job right, of making you care about these pilots as people, you’d find that choice wrenching and difficult.

Major injuries would always have a serious risk of death. I didn’t want you to ever be comfortable letting your people take hits. If you came back with any injury that knocked your pilot out completely, I wanted at least a 50/50 chance you’d be saying goodbye to them shortly. This, like so many systems meant to be driven by events, was mostly notional; I couldn’t push our focus towards evolving the event system any further than I had.

Initially I had the idea that really serious injuries, if you survived them, would permanently damage you in some way. Perhaps we’d lower your stats, or perhaps (this one was weirdly popular but never made sense to me) you’d develop some kind of psychological affliction that would impact you in battle in some unspecified way. I mostly discarded this idea for the simple reason that it wasn’t fun.

Remember, the point of all this was to make these characters feel real, and to get you invested in their stories, and get you to follow their adventures and tell other people about their lives. If I broke a character mechanically, made them less useful, made you less likely to keep them around and take them on missions, their story would end prematurely. What I wanted was for you to take that battle-scarred veteran out of the medbay and shove her back into the pilot seat, whether she wanted to go out again or not. I wanted the process of recovery from injury to be a process you eagerly anticipated the end of.

Most of the brutality of the injury system was stripped away, though you can still get some of it back through the difficulty settings. I ended up loading a lot of the more player-unfriendly ideas I’d had into those settings, actually. I know the kind of fun I like to have, and I also know that it doesn’t necessarily work for everyone. I’ll probably come back to those difficulty settings, and the compromises they represent, sometime in the future.

End of Service

In The Company, my Facebook game that was in many ways Battletech 0.1, soldiers would grow old and retire (or die). Old soldiers would request retirement; middle-aged soldiers would want to get married and leave service. And a soldier injured frequently enough might get sick and die; even a brush with disease could leave a soldier vulnerable to a wide array of lethal events that might crop up, without even considering the potential of their death in battle.

I couldn’t let your pilots die of old age; we weren’t passing that much time. The Company’s timescale was ‘seasons’, while Battletech’s scale was ‘weeks’ or even ‘days’. We initially planned to track age and increment accordingly but we cut that when I realized: what would we ever do with that information? How could we use it in a way that was fun?

In The Company, you could see an old, experienced soldier’s retirement coming. He was the backbone of his squad, a veteran in his 50s with twenty years of campaigning under his belt. You could start planning his replacement, looking through your roster for someone who you could put in as his second-in-command, ready to step up when Sarge finally retired (or died).

That wouldn’t work in Battletech, because the population scale was entirely different. The Company had potentially 200 soldiers available at a time, and organized them into 8-person squads. Battletech has, for most people, somewhere between 6 and 10 pilots, with four on the field at a time.

Still, I needed to ask: what happens when someone needs to be replaced? The obvious catalyst is death, but I also wanted less brutal ends — pilots might decide to retire, or join a different company.

When you level up your pilots, there’s a combination of stats that generates a title for you. This has no real effect, but I love it and insisted on it for its flavor. There was a practical purpose, though, which had to do with losing pilots.

If you’d developed Behemoth into a Guts/Piloting character, with the consequent title of ‘Brawler’, you could have a second Brawler you’d leveled as a replacement. Knowing that label and what it meant for the available abilities and traits of the pilot, you could quickly slot in a new Brawler into her ’Mech and pick up the slack immediately. In theory (though I suspect no-one ever did this) you could have a complete set of specializations — all twelve possible combinations — and a backup set of twelve more pilots with the same specializations. (This is, in case you’re curious, why there are 24 berths total in a fully-upgraded Argo.)

Ultimately I didn’t get the kind of pilot turnover I’d been hoping for. I think the game just wasn’t lethal enough, and losing a pilot was never the storytelling beat I wanted it to be. In retrospect, I’d probably emphasize non-lethal ends to careers; a happy ending for Glitch might be retirement to colonial administration, or marrying the girl of her dreams and raising grapes on Coromodir. Making those sorts of story-endings would probably go a long way towards making players feel positive towards pilot turnover.

Next Up

Next time, I’ll talk about the event system, why it exists, why it failed, and what could have been done differently.

Battletech is property of Topps, Microsoft, Catalyst, HBS, Paradox, etc. I don’t own any of it, nor do I represent any of those entities. I’m only talking about my own experiences as the lead designer of the 2018 computer game.