After seeing a preview of Outbreak: Undead (warning: has sound) online, and then again at GenCon, I knew I wanted to review this game for the Stew. I love zombies, horror, survival horror, and all the intersections thereof, and Outbreak: Undead (OU) is a zombie survival horror RPG — right up my alley.

It’s also an “avatar game,” an RPG in which you play a character based on yourself (though you can, of course, also play any kind of character you like), which presents some unique challenges and opportunities from a GMing perspective. So I wrote to the publisher and asked for a print review copy, which I tucked into right away.

So how is it? Here’s the short version: It’s got some flaws, but if you like zombies and want to try a different kind of RPG, you’ll love Outbreak: Undead.

Let’s Get Physical

Outbreak: Undead looks great. It’s a monster hardcover, 452 pages, with a full-color cover and a B&W interior. The cover art depicts a woman in two states, uninfected and zombie, with a polaroid showing her as a zombie with a note on it: “She is still pretty to me.” It’s printed on heavy, high-quality paper, and it looks and feels nice. It’s $45, which is about right these days.

The whole book is presented as if it were a survivor’s journal found after the zombie apocalypse: The background for every page is a lined page from a spiral-bound notebook, and the pages are plastered with sticky notes, taped-on bits of paper, polaroids, and sketches. It’s a slick, immersive presentation that really sets the tone for the book.

The photos are actual photos of scenes that, for the most part, fit the subject matter: empty parking lots, abandoned buildings, blurry people as zombies, shots of weapons, etc.

The sketches are almost uniformly awful, but I think they’re supposed to be — in the context of a zombie survival manual cobbled together in a post-zombie-apocalypse world, they look like what a non-artist would sketch to illustrate the stuff they’re writing about.

Unfortunately, OU uses a handwriting font throughout, interspersed periodically with typewritten text. This was cool for the first few pages, but became grit-my-teeth annoying by the end of the book — handwriting fonts aren’t as easy to read or as conducive to the kind of clarity you need from an RPG rulebook as plain text. (It also means that while this is a ginormous hardcover, there’s a lot less text per page than you might expect, which isn’t necessarily good or bad.)


OU is broken into 10 chapters: About, Characters, Skills + Traits, The Turn, Zombies, Combat, Vehicles, Equipment, Gamemaster’s Section, and Glossary. It lacks an index, which in a book this large is a huge oversight and a crying shame.

After spending some time with the book, I found it annoying to navigate because of the lack of differentiation between sections within chapters. There are no page-level headers to help you orient yourself within a chapter, so I always found that it took longer to locate specific things than it should have.

Trying to look up specific things is equally aggravating. For example, I wanted to look up how to make a skill check, which isn’t covered in the Skills section. It’s at the front of the book in an unlabeled sidebar in the About chapter, which explains that checks are all based on statistics and are impacted by Difficulties, and it took me several minutes to find. If there were an index, I would have found it right away.

I get the impression that graphic design and staying true to the “survivor’s journal” theme were prioritized over creating a usable, well-organized RPG book. The saving grace is that the system isn’t terribly complicated, so once you’ve looked something up once it shouldn’t be too hard to remember. Still, the end result is a book that’s pretty annoying to read and somewhat annoying to reference.


Now that you know what OU looks like and how it’s set up, let’s talk specifics:

I want to play me!

Although OU pitches playing a character based on yourself right up front (“avatar play”), the book doesn’t actually include all of the rules for playing yourself. It’s not until page 28 that OU mentions that you need to go to the game’s website to generate a character based on yourself. What the book does include is the rules for creating a template character, IE a normal PC that isn’t based on you.

This is pretty annoying, since one of the selling points on the back cover and in the early pages of the book as that you can create a character based on yourself. Not including those rules feels like a bait-and-switch. That said, the avatar creation system is free, and offering it online has the advantage of letting your players start their avatar characters without needing to own the book.


Okay, so how does avatar PC generation work? You go to the game’s website, click on “SPEW-AI,” and take a test designed to measure your physical and psychological capabilities: SPEW-AI, which stands for “Strength – Perception – Empathy – Will Assessment Inventory.” You then answer multiple-choice questions like this one (the first):

Think of exercises that involve being flexible (yoga, rock climbing, etc.) and think of how your flexibility compares to the average person. Which of the following describes you?

Which is followed by five options ranging from “I’m a contortionist” (A) to “I’m not flexible at all” (E). The whole test is 40 questions, took me about five minutes, and is fascinating. Why? Because it’s stated aim is realism: From the introduction to OU: “The purpose of this game is to create a simulation that can quantify and allow for an accurate representation of a person’s ability to survive any possible Outbreak scenario.” (More on the “simulation” thing later.)

When you finish the test, it spits out just four stats: Strength, Perception, Empathy, and Will. The range runs from 5-46+, “weak” to “Olympian.” My stats came out as 15 Strength (low average), 22 Perception (average), 28 Empathy (average), 26 Will (average). IE, tubby zombie bait — which, if I’m being honest, is probably about right. Interestingly, that’s just 91 points — whereas template characters get 120. So clearly my goal should be to get myself killed ASAP so I can stat up a muscular cop with heroic levels of empathy…

So why isn’t this test in the book? I’m not sure why, because it’s not that long, wouldn’t take up much space, and appears to just be assigning points based on your answers — something easily accomplished with a breakdown familiar to every Cosmo reader (ahem). If I’m overlooking some complex element, I don’t know what it is.

The SPEW-AI test is neat, fun to take (if depressing, which is appropriate for a zombie survival horror game!), and unique, but it should be in the book.

You’ve got to be fucking kidding me

OK, so I have my stats; now I turn to the “Characters based on Players” section in chapter two, where the game tells me that I can pick a Type (basically a class, like Lawyer, Nurse, or Firefighter, should I happen to be any of those things in real life) or create a Type based uniquely on me. For the latter, the primary draw of the game, I’m told to turn to the section “Yourself as a character”…which doesn’t exist.

No, really.

And on top of that, I’m told that the SPEW-AI test should have generated a recommended skill set for me, which it didn’t — all I got was my stats. At this point, I’m kind of screwed; I have no choice but to wing it, since one of the core elements of the game simply doesn’t exist. I had to triple-check this, since I just couldn’t believe it.

So can I wing it? Sure. The Characters chapter tells me that characters get five Tiers of Skills, which can be sacrificed at a ratio of 1 Tier:5 stat points if I like, and that if I instead pick a Type from the book I get no stat bonuses, two Skill Tiers, and whatever non-stat bonuses are provided by that Type. Plus Gestalt Dice, which are a mechanic unto themselves (more on that later).

But if I’m new to gaming, I’m flipping around not finding what the book tells me should be there and going “What the fuck?!” at this point. Super-lame.

The core mechanic

Moving on, let’s peek under the hood at the game system. The core mechanic is percentile-based, with standard checks being made against your SPEW stat + any bonuses from skills or equipment, roll equal to or under to succeed. For every full 10 below the needed value, you achieve an extra degree of success; ditto for degrees of failure in the opposite direction.

If a check has a higher-than-normal difficulty, that’s applied as a penalty to your stat in 5% increments before you roll. So if you have a Strength + climbing gear bonus total of 40% for a check, but the climb is Difficulty 2, you now have a 30% and need to roll 30 or less to succeed.

This is a simple core mechanic that I suspect would get out of the way and let you focus on the game in actual play, just like it does in other percentile-based games I’ve played.

Mad zombie survival skills

Skills are broken out by Tier, with Tiers representing the amount of training you need to have a basic level of proficiency. There are several dozen skills, and they’re pretty oddball: Agility (1st Tier) gets you a bonus to some skills and checks, while Caged Wisdom (3rd Tier) lets you make shivs because you’ve been to prison, and Sniper (5th Tier) means you’re a trained sniper.

The skills are really a mix of what most RPGs would call two different things: skills and advantages (or traits, or feats, or whatever), with skills letting you do specific things and advantages giving you bonuses. It’s a bit quirky, but it’s workable and gives the game a unique flavor. It also emphasizes that the focus of OU is different than most RPGs.

Those bonuses are important because they’re added to your base stat when you make tests. For example, Agility gives me a +5% bonus to Strength for climbing and jumping, which, with my 15 Strength, would be pretty important. (Unfortunately, I can’t take Agility because my Strength isn’t high enough — unless I pick a Type that grants Agility as a bonus skill, which overrides that prerequisite.)

Gestalt Dice

Gestalt Dice are part of the skill system: If you’re playing an avatar character, you get one Gestalt Die for every year of your age, which represents the fact that the full range of human abilities and specialties can’t be represented by any skill list. (Template characters get them too, but they work a bit differently.)

When making a check, you can add Gestalt Dice to the result (how many depends on your years of experience in that skill). You roll that many six-sided dice, add that number — and that bonus is permanent. So If I’m making a Str – Endurance check and I add Gestalt Dice to the roll, the bonus granted by those dice applies to all future Str – Endurance checks as well. This permanently reduces your Gestalt Dice pool.

This is a neat system that’s unlike any I’ve seen before. Self-applied bonuses aren’t uncommon, but the permanency aspect is new to me. I love this idea, and it introduces a neat fun factor into what’s otherwise a pretty grim game — particularly because the average schlub (like me) is going to have fairly shitty stats to start out with.


As you might expect, the chapter on zombies is lovingly detailed. It opens with the immutable and mutable laws of zombies (immutable: there’s no cure; mutable: only zombie-inflicted wounds transmit the infection), and then provides a solid overview of zombies in general, rules for zombie perception and movement, and zombie traits.

OU also covers zombie priorities, which fits well with the simulation aspect. Zombies choose their targets in this order: closest, in plain sight, downwind, least infected. Properly enforced, that list will keep you, the GM, on the path to a brutal zombie survival experience — and your players will know it.

The bulk of this chapter is taken up by templates for different zombie types. Each includes stats, a description, special traits, and an illustration or photo. More or less every kind of zombie you’ve ever seen in a movie, comic, book, or game is statted out here, from banshees and titans (Left 4 Dead’s spitters and tanks, more or less) to standard zombies to zombified dogs (Resident Evil) and zombie kids.

This chapter is every bit as good as it should be. If you think a game where you fight zombies 90% of the time would be boring, this chapter will convince you otherwise. There are so many kinds of zombies, special traits, and things you can tweak to surprise your players that you’ll be set for a nice long stretch of play.

The turn

Chapter four, “The Turn,” is where OU’s divergence from traditional RPGs starts to become clearer. Turns are five seconds long and include six phases: surprise, intent, check, resolution, grapple, damage. The wrinkle is that second phase, intent: You go around the table and everyone declares what they’re going to do, then you resolve their actions in the next phase; resolution is simultaneous. Conversation among players is encouraged, and you can change your mind freely until the next phase.

Once you’re locked into a course of action, though, that’s that for that turn. If circumstances change in phase three (resolution), your action could be irrelevant and wind up being wasted. On the flipside, everything happening simultaneously makes it easy to team up against zombies.

This seems like a mechanic designed for maximum realism: In real life, things sometimes happen too fast for you to adjust on the fly. But will it be fun? If you’re in the right mindset to play a zombie outbreak simulator RPG, I think so.

I think of it like the original Resident Evil games, which used absolute controls and a relative camera: up was always forwards, but halfway down a hall the camera would flip to the opposite view — so now pushing forwards meant you were running back towards the zombies. Rather than being bad design, this was deliberate; it nicely represented the fact that in real life, you’d get scared and fuck up periodically.


Your character’s health is derived from your Strength (+1 per point) and Will (+1 per five points). A standard zombie does 1d6 damage with its bite; an average pistol does 2d6. Degrees of success multiply the number of dice rolled: two degrees = double dice, etc. Zombie bait like me (15 Strength, 26 Will) would get 20 Health.

Damage also inflicts wounds, ranging from zombie infection to internal injuries, and both characters and zombies can be killed outright with a single good blow. For characters, that’s a hit that causes 31+ points of damage; for zombies, it’s a percentage threshold representing headshots and the like.

Combat is pretty straightforward: turns proceed as described above, with all damage being applied simultaneously. Zombies are rarely encountered alone, and that plus their special abilities makes them deadly. Zombie bites can infect characters, too, and pain from your wounds plays a role.

Healing is slow and difficult. First aid restores Health, but it’s based on degrees of success, not a simple roll. And natural healing takes a long time, modified by your activity level and how sanitary the conditions are.

Overall, combat in OU leans towards realism, though it’s not overly crunchy (unusual for a realistic combat system). There are specialized rules for automatic weapons fire, grappling (which zombies do a lot), teaming up, and other situations, but no real surprises.

Your chances of surviving a single gunshot wound or zombie bite are quite good, but healing from those wounds can be difficult — and, of course, you’ll rarely be in a situation to get shot or bitten just once. On top of that, with degrees of success modifying damage rolls, any damage can potentially inflict serious wounds or even kill you outright.

On balance, OU is a deliberately deadly game.

OK, what’s with all the “simulation” stuff?

It can be a subtle distinction at times, but Outbreak: Undead is explicitly a zombie outbreak survival simulator, not a traditional RPG. It’s also an RPG, but one that’s colored by its goal of realistically assessing both your moment-to-moment survival and your overall ability to survive a zombie outbreak.

This is why, for example, your baseline stats aren’t so hot. With a 15 Strength, my avatar character sucks at climbing or lifting things when it matters — without bonuses or Gestalt Dice, I’m going to fail a lot. Similarly, it’s why Health is a precious resource that’s easy to lose and hard to recover.

It’s also why there are some interesting programmatic elements in OU — for example, the game is designed to be played out in missions, with victory conditions for every mission. This isn’t fundamentally different from most RPGs, where every adventure has a successful/failed outcome, but it’s made explicit in OU through subsystems to determine random encounters, by the mechanics for each mission type, and more.

The simulator aspect of OU also means that it’s more focused than other similar games. Take Eden Studios’ All Flesh Must Be Eaten: It’s a much more freewheeling, traditional RPG, with sourcebooks for zombie pirates and zombies in space. By contrast, OU uses the modern world and, by default, your actual players as characters — that provides a lot of focus right out of the gate.

Encounter Checks

Let’s dig into Encounter Checks as a good example of the simulation aspect of OU. Because the world is swarming with zombies, as well as sparsely populated with panicked humans, resilient survivors, and vigilante bastards who want to take your stuff, OU uses a random encounter system to generate a constant sense of danger and ramp up the tension. This is nifty, and I think it will really set the proper tone during play. It also funnels the game into the simulation it’s designed to be — here’s why.

You, the GM, make an Encounter Check whenever one of five things happens:

  1. The characters move for one unit of Time or look for a place to rest
  2. The PCs search
  3. After a certain number of turns during an encounter (so zombie “adds” become a real threat)
  4. The characters make a survival check or rest
  5. The PCs flee an encounter at a dead sprint

Taking that first condition as an example, if the party moves at a normal walking pace, you make an Encounter Check once per mile. The check is rolled against the senses of the zombies in the area; for standard zombies that’s 30% (because they can see and hear, 15% each). So after a mile, you roll d100 and on a 30 or less, the PCs run into zombies (or survivors, vigilantes, etc. — but it’s usually zombies); on a 31+, no encounter. The number of zombies is also dependent on a formula.

Are you free to ignore that? Sure thing, just like any aspect of any RPG — but played as written, Encounter Checks set the pace of a game of OU, and that pace is likely to be pretty brutal. After running into a few random packs of zombies, with each bite potentially resulting in one of the PCs becoming a zombie, your group is going to start focusing on finding shelter, fortifying their shelter, etc. pretty damned quick.

In other words, the programmatic elements of OU, especially Encounter Checks, will automatically populate your game with all of the elements of a good zombie movie, will create palpable tension, and will “funnel” your players into doing the kinds of things real people would do in a zombie outbreak.

I usually hate random encounters, but I think this is fucking awesome — it’s perfectly matched to what OU wants to be, and it should happen organically during play. Among the many things that make me want to play OU, this is at the top of the list.

Mission structure

To run OU, you need to determine the outbreak scenario. As the GM, you sit down and determine the type of outbreak, whether the game will start from the first infection or pick up mid-outbreak, what kinds of zombies are involved, what traits and unique properties those zombies have, and whether the goal is survival or stopping the outbreak.

In a nutshell, that’s your campaign. It’s a pretty straightforward process, but one that offers a significant amount of variety. Combined with fairly quick and dirty character creation, this makes OU simpler to pick up and play than the size of the rulebook suggests.

Once you have the world and the zombies all sorted out, you’re ready to run some missions. Each mission type includes a short description, prerequisites (having a stronghold, for example), the objective, how much Time it will take (which influences the number of Encounter Checks), conditions of success and failure, and special stuff (rules, conditions, whatever).

The first mission is a good example: All-out Defense. In this mission zombies are attacking your stronghold, which can be determined by the GM or programmatically by having zombies trailing the PCs after another mission roll higher than the stronghold’s level. The objective is to eradicate all the zombies, which is made more difficult because the population of your stronghold can’t move freely without PC escorts.

The characters get 10d10 rounds to succeed, and if they don’t succeed they have to flee — abandoning the stronghold in the process. There are a few special rules for this mission, including rolling for the time of the attack and a roll to determine if characters begin the mission in Panic (a state that has mechanical consequences).

All of that takes up less than three pages, and all you need to add is a map. Google up some floor plans, doodle a strategic-level map for your players, and you’re off and running. Provided you created the circumstances of the outbreak in advance, you could play this mission at the drop of a hat, which is handy.

On the flipside, this type of formal structure can make some players feel stifled. Provided you set expectations upfront, including an emphasis on the simulator aspect of OU and what makes that so much fun, this shouldn’t be a huge hurdle for most groups.

Not much GMing advice

On the one hand, you don’t actually need that much GMing advice to run OU. Setting it in the modern world, with your players as the PCs, and having systems to programmatically address many aspects of gameplay means that you’re free to focus on improvisation, atmosphere, and other GMing details.

On the other hand, having your players as PCs is unusual, given that most modern RPGs don’t attempt this, and an overview of how that works in practice would have been useful. Similarly, even though OU strives for realism, it’s still an RPG at the end of the day — help me out with some advice on enhancing the tension, setting the right mood, etc.

I can see the decision-making process with regard to avatar characters and the lack of GMing advice revolving around the fact that no one needs advice on playing themselves, and GMs shouldn’t need advice on dealing with PCs who are in fact their friends, but as a GM I would have appreciated some guidance.

Cool little details

OU is full of nifty details that make it clear how much effort and TLC went into making this game the best possible “death by zombies while the world falls apart around you simulator” around.

For example, every character has a bite resist stat (half your Perception + armor/clothing + skills), which is the chance that a given zombie bite will raise your infection level. And every shot you take at a zombie has a chance of killing it outright (Ranged Attack + weapon lethality modified by zombie protection) — headshots!

I also love that noise attracts zombies, with consideration given to whether it’s instantaneous noise (a gunshot) or sustained noise (a chainsaw…), all matched up against the zombie perception subsystem. OU is packed with these kinds of details, and they really make it shine.

Should I buy it?

Assuming your group wouldn’t be turned off by the simulator aspect of Outbreak: Undead, yes.

The lack of organization and the annoyance of the handwriting font are more than offset by everything else about OU, from the badass zombie chapter to the Encounter Check system, simulation elements, and the level of detail and thought that went into creating what’s ultimately a very cool zombie horror RPG.

It’s a testament to just how cool this game is that even though it fails to include complete rules for creating yourself as a character — one of its biggest selling points! — I still recommend buying it. Was I pissed to find that out? Yep — but the rest of the game made up for it.

Outbreak: Undead is different from most RPGs (closer in some respects to an indie RPG than to, say, D&D), but I found those differences to be pretty nifty overall. I’d love to play this game with my group, and it seems like it would shine in the convention event format, with its tight scenarios, low prep time, and the sexy, easy-to-describe hook: “It’s a zombie survival horror game where you play yourself as a character.

Questions welcome

As always, I’m happy to answer questions in the comment, and feedback is welcome! Even at over 4,000 words, there was no way to cover everything I could have covered about OU, so if I missed something you were curious about just let me know.

Update: One of the folks behind OU, Ivan Van Norman, emailed me this (thanks, Ivan!):

Outbreak: Undead developer here… Must say, thanks for such an honest review. We appreciate your candor with both our strengths and weaknesses, Martin.

In regards to a couple of of specific questions raised: yes, we are having a PDF available soon (slated for late November) and it will be revised based on the Q&A of our awesome forum members and the extremely insightful reviews such as this one. A complete errata will be available for free on our site, so those who own the book will benefit from the revision as well.

This revision will contain an index. The initial lack of an index was not a style choice, nor was it an oversight… It was unfortunately a sacrificial lamb to the dark deadline gods in order that we meet our GenCon release date.

In regards to the missing ‘Yourself as a Character’ section, that was a glaring oversight in the regard that we deleted the header and moved the contents without realizing that the section itself was still being referenced. What would comprise the “Yourself as a Character” section was distilled into the content at the top of pg. 29 and expanded in the Gestalt Dice section in the Skills chapter. So the content is still in the book, it is just improperly labeled. This is obviously addressed in the revision.

In regards to the SPEW-AI: it is a test that is still in its infancy and, at best, a blunt psychological testing instrument. Its exclusion from the book was so that each improvement we make would not require a re-print of the entire 456 page core rulebook. Such scheduled improvements include the recommended skill set (a feature that we had hoped to have implemented before our release, hence its inclusion in the body copy of our book), built-in lie factors, an expanded question set and gender/age-factored questions. Although, we’re focused more on smoothing out our game mechanics, this test remains a very high priority of ours and demands more than the passing attention we have been able to give it. We would actually feel morally remiss if we didn’t approach this subject seriously. Involving actual player avatars requires that we do our best for honest portrayal especially when the in-game lifespan can be so brutally short. For this reason, the SPEW-AI will never be complete, as it will be improved upon and revised with the passage of time.

We also put out free content each month that we try to make a combination of new material and GM tips that we provide to our community, so we don’t like leaving people in the dark… We know we’re a brand new game and we are generating material accordingly. Anyway, this has gone on long enough. We really do appreciate such an honest review. Thanks again. Take care and prepare.