Designer Diary: It's Grave Business from Start to Finish

Where to start?

Some games start with a mechanism, others with the theme. Grave Business is one that started with the core mechanism, but I think the end product feels like it started with the theme, so that's where I'll begin.

In Grave Business, you're a dangerous necromancer because you're a necromancer with a business plan. You're going to send forth your zombies to dig up and loot graves, and while they're there, they'll check for fresh body parts that they can bring back to you so that you can construct MORE zombies to loot MORE graves. Unfortunately your plan is so good that now other necromancers are jumping on the bandwagon, so expect a bit of stealing and fighting over the choice bits.

By this point you're either enticed – "Ooh, zombies!") or repulsed ("Oh... zombies." – so I'll continue in that direction and see what else I can say about the game that may polarize you: Grave Business is a worker-placement auction game with directly aggressive actions, what some call "take that!" actions. Each of those aspects has its fans and not-so-fans, so let's delve into them a little deeper to see whether I can convince you to try the game regardless of your misgivings.

Evil Auctions

I tend to dislike auction games. I feel they rarely do something interesting mechanically and frequently aren't implemented thematically, so they are often a patch thrown over an otherwise lackluster experience because a standard auction "self-balances" whatever is being auctioned and provides player interaction where none might have existed otherwise. So for me to like an auction-centric game, it needs to avoid those tropes. The auction needs to have a reason for being in the game other than being a band-aid – and preferably it should do so in a non-standard way.

In Grave Business, as I said, I started with the mechanism, specifically an auction but one that was interesting mechanically as it's a spatial auction. Your bid is placed on the contents of an entire row or column of a grid, and if someone else places a bid on a row or column that overlaps yours, whoever has the highest value on any given space will win its contents. You place several bids over the course of the round, so it's possible for you to bolster your influence, as your bid values combine where they overlap on the grid. Thus, you're not bidding in the normal auction manner, and knowing what won't be fought hard over is as valuable as knowing what everyone will be fighting over. Moreover, since this style of auction isn't just "have vs. have-not", it can function with only two players, which tends to be a major shortfall in a lot of other auction games.

This concept (along with several others) came to me while I was doing my daily perusing of gaming articles. Certainly there are other games in which you're exerting influence in columns and rows, but I felt this particular implementation had a character all its own, and I spent the next day or so thinking about what flesh I could put on that character to further make it not feel like another ho-hum auction game. Why are we gathering things in columns and rows? Farming obviously fits, but why would farmers be competing over the same area? What else comes in rows and columns? Houses... marching soldiers... gravestones. Why are we fighting over graves? Because we're graverobbers, of course.

Once gravestones clicked for me, the rest started to fall into place. A basic concept in games is that you distribute resources, then score them. Most games can be boiled down to this, even if the distribution or the scoring is presented in a convoluted way. I had a distribution mechanism that was interesting and thematically relevant – now I just needed to ensure that scoring would also be interesting and thematic.

Pre-testing

Stealing valuables from the graves is the obvious thing for graverobbers to do. So I went with that, giving the graverobbers a "digging value", applying some standard set-collection and flat-value treasure, and pushing bits around the board by myself. This was okay, but not particularly enthralling. Moreover, patterns started to emerge in the way I was arraying my graverobbers. So I needed to make what you were acquiring more interesting, and I needed to make the bid values not as apparently static, so that the same patterns didn't emerge every time. This led to the first version I tried with another person:

You were still graverobbers, but you also used zombies to perform various tasks. Each graverobber was assigned a type of die, which was rolled before placement so that assignment didn't shake out into the same thing every time. Each player had a graverobber that was based on one roll of the D12, for example. Zombies could help you dig and could also attack each other, steal things, etc. You could make new zombies with the corpses you dug up.

Going to the Grave for Real

The first two plays were very informative. The robber values changing from round to round did provide variance in placement patterns, but also felt somewhat clunky. More importantly, since the robber values were public knowledge, the last two or so placements each round bogged down as people crunched numbers to determine their maximum possible gain.

On top of that, the zombies felt unimportant, partially because they could damage only zombies (out of necessary sanctity of the robber bid tokens) and partially because what they were doing didn't feel like it fit with the rest of the game. People liked the zombies, and allowing you to gain zombies from digging was a good way to make the possible graveyard acquisitions more interesting, but they wanted the zombies to feel like...zombies. They wanted more of them, and they wanted a reason for them to fight and rip each other apart.

There were lots of other problems, of course: the set collection in scoring seemed to add only length and not interest to the process; the action system was sort of loose; and there was definitely a dominant strategy cropping up. Oh, and another minor thing, I didn't have a game end condition yet.

So I took the main problems and worked at fixing them by giving people what they wanted. Now your graverobbers were zombies. You could gain more effective bidding tokens by building additional zombies (which were easy to build), and the action system was completely overhauled. There was a centralized worker-placement style board with actions which required different numbers of zombies to perform them; new action spaces would pop up in the graveyard and be added to the board, and when the spots for new action spaces filled up, the game ended. Additionally, zombie values were now hidden, to prevent the heavy number crunching. I changed most of the treasure to flat rate stuff, and I left the dice in because I felt they may turn out okay with the values hidden.

Protospiel

My next test was at the Protospiel convention, and while some things smoothed out, others did not. The action system was deemed okay, but not particularly thematic – and if have a game with zombies, the theme has to come through.

So it was suggested that all or most of the actions occur where they mattered. If you wanted to steal someone's stuff, you would need to place your zombie in that player's pile of treasures, not on a central board. The same would apply to attacking, digging, etc... The zombie should go where it's actually doing the action.

What else? The crosses were used to give +2 to a particular square's bid value, but that also felt a bit flat; why can't I place a zombie straight on the square instead? The dice were too clunky; they had to go as the hidden bid values were adequate for differentiation. The care and feeding of your zombies felt a bit fiddly, in no small part because there were so many out there now. The name I was using – "The Corpse-taker Covenant" – too serious. Lots of other useful suggestions were noted and filed for the future.

Back to the Drawing Board

So I did my third (or fourth?) major scrapping of the way actions were handled and further tweaked the way bids took place. Out with the dice entirely.

The way zombies were hurt and killed was good, but very same-y, so I scaled the construction of the zombies so that you had different "sized" zombies. This also helped with the quantity of zombies that could be out and about at the same time as now you couldn't build infinitely many small ones. Since the fights were now 1-on-1, there was an easy way to implement zombies fighting back. I kept the primary thematic actions I wanted the zombies to be able to do and removed all others, now placing the action spaces where the action actually occurred. This coincidentally helped re-centralize the core mechanism of the game (the graveyard digging) and made everything feel like your zombies were going forth to do your bidding (pun intended).

The next several playtests went much better. Tons of small and large changes were applied and tried, and values, quantities, and game length were all worked on. The game now ended for the most intuitive reason: You no longer had tokens left to fill the graveyard, and the number of tokens was chosen so that the game wouldn't drag past the point where it was obvious who was in the lead. As a consequence, I removed graveyard scaling based on the number of players. (The scaling was not that useful in any case as the two-player game tended to be too long and the four-player game too convoluted.)

This change also allowed me to reduce the player mat to the actual number of zombies you would tend to build in one game. Notice the player boards from a prototype during this period in the upper left, which have way more spaces for zombies than the finished game does:

Lots of changes happened as issues surfaced after repeated plays. Instead of starting with small zombies, for example, you start with some of the largest ones possible. This serves two purposes: (1) your initial zombies were a little harder to kill (which is good for stability at the beginning of the game) and (2) while building zombies is still important early on, it's not the only thing that matters.

Look again at the player mats in the upper left of the image above. When your starting zombies are the 3, 5, and 7, as soon as someone builds a zombie that's bigger than that, that player starts dominating the game.

The values of treasures were bell curved so that the normal equity change from a steal or a particularly successful digging turn was flattened. As a result, the game became less "swingy", and good decisions over multiple turns mattered more than a couple of good hits.

Players didn't feel like they were necromancers; they felt more like they were a horde of zombies – which wasn't a bad thing, but thematic explanations for things worked better if the players were necromancers, so the boards were given specific necromancers, and starting zombies were more individualized (using my very poor artistic skills).

Players wanted to be able to prevent thefts or to take over thefts from particular spots, so I tried letting combat allow the attacking zombie to directly usurp the position of the defender, but this made attacking way too strong. Instead, you could attack someone to knock that zombie off its spot, and if that spot was still available on your next placement, you could then place a zombie there. All this assumes, of course, that the defender didn't successfully fight back. To make this work correctly from turn 1, the starting zombies were designed to give them an average number of combats they could be expected to survive, and to be able to estimate average success rates for actually removing them from their positions. Your starting zombies are now tougher than later-built zombies (in general), but also tend to give up their spots more easily. You can even custom-build a new zombie so that it always fights to keep his spot, if you save up the correct kinds of corpse pieces.

Gen Con 2010

I started branching out my tests to different groups and worked on the full rules sheet. Then, at Gen Con in 2010, I ran into one of the people who had been at the Protospiel playtest. He was working his booth at Gen Con because he happened to be the owner of Minion Games. He asked how my zombie game was coming along, and I outlined some of the changes I had made in the interim. (It was a very different game from the one he had played before at this point, aside from the core auction mechanism.) He asked whether I had a prototype with me, and since I did, he looked over it more thoroughly later that night.

When I got back from Gen Con, he emailed me, asking whether I could send him the files necessary to make and try the latest version of the game himself. He did, and sent me the feedback results from this playtest (and as any designer knows, feedback from external groups is perhaps the most valuable thing you can get), so I made more changes and sent him the files for the next iteration. From then on, I was getting roughly double the normal amount of playtesting and feedback on the game (his groups plus my own), and was able to iterate even faster and more thoroughly. Weak points were shored up, and the strong points were enhanced.

The actions and their average value were tightened so that you always have something interesting to do, and the tension arc of the game increases as the game goes on. The two-player rules in particular had to have some changes applied; although everything worked, an early advantage could be pressed too hard, and the dominant player was difficult to stop because there wasn't enough time to debilitate him. So a few small rules differ in the two-player game to ensure that the overall flow of the game is enjoyable and full of meaningful decisions throughout. The amount and value of the corpse pieces available needed to work at all player counts as well (so that adequate zombies could be expected to be created and killed over the course of a game), and so that was poked and prodded thoroughly.

We Mean Business!

Then, when it seemed like we were running out of functional things to change and the game had stabilized, he said he was interested in publishing it! Now the playtests continued, but the major changes being made were usability and theme oriented. All the zombies were named to go with their master; token size and board spacing was changed and changed again to make them work well for handling and placement; and the number system on the zombies was altered slightly to the "brainpower and bones" system in the final product. The game is functionally the same about 99% of the time, but resolving who gets what in the graveyard is MUCH faster. I was actually surprised at how well that change worked, and how much time it shaved off the game.

Numbers and play arc were tightened even further. You might be surprised at how much thought I gave to a tiny two point difference in the combined total value of the graveyard tokens.

Recently, the (almost) final art was finished up and is awesome – I really like Chuck Whelon's style as it makes me want to pick up the box and find out what's inside – and we got to test it at the 2011 GAMA trade show. In addition to passing the color-blind test for both red-green and greyscale – many thanks to the person who helped us with that – the whole package went over swimmingly. The rules layout has been finalized, and I believe all that's left is some tweaks to the box.

But How Does It Play?

Okay, here's a brief run-down. Everyone starts with three zombies. You will take turns placing them, and only when they are all placed will they perform their actions – with the exception of attacks, which occur immediately. Zombies will steal, then dig up the graveyard, then move the start player marker. Any zombie that's taken too much damage is then removed, and you may build new ones with your freshly-gathered parts, then equip them. You then fill up the graveyard and do it again. If you can't fill up the graveyard, the game ends, and the player with the most points in treasure and unused body parts wins – assuming that no one resurrected the Master, that is.

Remember the set collection that kept getting lessened as playtests went on? Now there is only one set: the pieces of the Master. If someone has all of them at the end of the game, that player wins regardless of score. If this happens, though, it's generally because the players LET it happen. You should have plenty of opportunities to prevent someone from grabbing all of the Master, and this challenge serves to shake up the expectations and focus of players' actions throughout the game.

Everything is worth doing: stealing, attacking, choosing the start player. What matters most is when you do an action and with which zombie. You are making "blind" bids, but these are definitely "blind yet informed". You know where the zombies are going, and you know what their possible values can be, and people's placement tendencies are much more predictable than you might think. Even if you aren't particularly observant, the way the actions work will provide you with additional tools to figure things out.

It's interesting watching a player progress in understanding the game. The first few turns of their first game, people rarely attack each other. By your third game, there's almost always bloodshed on the first turn. Similarly, the "Start Player" action (which lets you choose who the new start player will be... usually you actually want to go last) is often considered weak by new players, even though it has several small perks attached to it. An experienced player can get huge value out of those perks in the right situation, with the start player choice being just an added bonus. I've been told the best way to win is to aim only at treasure. Or building zombies. Or destroying other players' zombies. Or snaking whatever low-value stuff is uncontested. In reality, you can approach the game a lot of different ways – it's how well you pick your spots that matters the most.

So I hope that I've convinced you to take a look at my worker-placement/auction/zombie game. You get to build, you get to destroy, you get to have fun and make interesting decisions while commanding your small horde of undead, all in (usually) under an hour. During the whole course of its evolution, every change was designed around taking the parts the players enjoyed, then taking the changes necessary to balance the game, and trying to find thematic ways to combine the two. If you're tired of zombies, think of it as a worker placement game. If you don't like worker placement, think of it as an auction game. If you don't care for auction games (like me), think of it as a zombie game.

Andy Van Zandt