Theme or Mechanics?

I read a post here in which Peter Schweighofer was musing about which comes first, theme or mechanics.  As he said in his post, theme gets people to the table, but mechanics keep them coming back.  As a good blog post should, it got me thinking.  Certainly, I have signed up for games at conventions based on the event description, theme, scenario, or historical period.  Many times I have been disappointed by a game with a terrific theme that uses a set of plodding and pedantic rules and walk away thinking about the four hours of my life I’ll never get back.  That is an example of what Peter is describing in his blog post.

I am always in search of the next project.  For me, the answer to this “chicken or the egg” question is “it depends.”  I think you need to match the right mechanic to the right theme.  A light-hearted theme doesn’t fit well with a set of detailed mechanics.  While many are doing it, I think that some periods and scales do not fit will with mechanics that are too simplistic.  (Simplistic is not the same as simple.)  So I often keep a set of mechanics percolating in my head, but instead of cramming that mechanic into my historical interest of the moment, I keep them in reserve for when they will be a good fit for a new period.

I am always thinking about new mechanics.  I don’t find a mechanic and then build a game around it, however.  Often I think about a mechanic for a couple of years, often while running.  The card-based mechanic that I use in Combat Patrol(TM), for instance, has been percolating for over 30 years.  I first saw something like it in the late 1980s.  I tried it again in the mid 1990s with an abortive ACW skirmish design, called Stealin’ Chickens for Gen’l Lee.

I have been thinking about a dice-progression mechanic in which modifiers change the size of the die, not the die roll.  In such a scheme, my “base” might be a d8, but if I get a +2 for some reason, that means I am rolling a d12.  What I like about a single opposed die roll mechanic with a dice progression is that if one player is rolling a d4 and the other a d24, there is still a chance, albeit a small one (6%), for the d4 player to succeed.  Such a mechanic generally breaks as soon as you pass d12.  Each next higher sized die adds 1 to the expected value of the die.  I am being a little sloppy in my terminology here, but you can think of the “average” roll on a die as the sum of all the faces divided by the number of faces on the die.  As an example, the expected value of a d10 is (1+2+3+4+5+6+7+8+9+10)/2 = 5.5.  For common dice, you can that a shortcut of adding the highest and lowest face and dividing by 2.  So the expected value with a d10 is 5.5; for a d12, 6.5; for a d8, 4.5.  That’s all well and good, but what happens if you are at d12 and have to apply a +1.  Some people will have you roll multiple dice.  The problem with that is that one die gives you a uniform distribution, but the more dice you roll the more likely you are to have a strong central tendency, as a plot of all possible die rolls starts to approximate a normal distribution.  Sometimes that’s fine.  In fact in Santa Anna Rules! and Wellington Rules!, I used a d20 (uniform distribution) to approximate a normal distribution on purpose.  But with a dice progression scheme, I think rolling multiple dice breaks the elegance of it.  Also what two dice would you roll to get an expected value of 7.5?  8.5?  An option is to “cap out” out at d4 and d12, but I don’t think this gives you enough variance.  Recently I found a company that makes uncommon dice, so I have d14, d16, d18, d22, and d24 to add to the normal d4, d6, d8, d10, d12, and d20.

If you are interested, here is a table I worked up that describes the attackers chance of winning (rolling higher than the defender, ties going to the defender) based the type of dice each are rolling.  (The table is not symmetrical because of the ties.)  (This is the kind of analysis I think a good game designer should do.  I don’t think most of the “buckets full of dice folks” look at central tendency and probabilities.)

So this is an example of a mechanic that is percolating and waiting for me to get interested in a historical period where this might be useful.  I was thinking about some sort of small-scale skirmish / pseudo role-playing game, perhaps gladiators.  My attack attribute says I am a d10, and your defense attribute is a d8.  But you have a shield, which is a +2.  I am slightly wounded, which is a -1.  So now for my attack, I am rolling a d8, and you are rolling a 12.  I have a 29% chance of success.

I have been thinking about players searching around the table for the right die.  The search for the right die might work against the speed of play desired in a dice progression mechanic.  Most of us can pick out the normal polyhedral dice from across the table through years of familiarity.  The newer dice tend to look very similar and may be difficult to distinguish.  I can see players not liking having to hunt around the table in the heat of battle.  As I rounded up the dice for this mechanic — for the time in which the right period provides a muse, and there is a good fit between that period and this mechanic — I made sure that each die type is a standard color.  D14s are all green.  I would have liked the colors to follow a coherent color scale, but the non-standard dice come in limited colors.  The best I could do was ensure that each die type is a different color.

I was reading one of the Wally Simon books that have been published by On Military Matters.  In one of the articles he describes the difficulty of getting two units to stop shooting at each other once their lines halt.  This was echoed in the Nosworthy books and showed up as an optional rule in Wellington Rules.  I have been mulling over a mechanic for an 18th or 19th century battle game that uses an action, reaction, reaction, reaction mechanic, not unlike Ed’s Two Hour War-games’ chain reaction system.  My worry is that several players may be sitting around watching two people “do stuff.”  This brings up another point.  I will often come up with an idea for a mechanic and then spend the next couple of years thinking about all its drawbacks and how to address them.  This is what I did with the card activation scheme in the Look, Sarge, No Charts(TM) family of rules as well as in Combat Patrol(TM).  Every mechanic has pros and cons.  Do the pros outweigh the cons?  Are there ways to mitigate the cons?

I have long wanted to create a jousting game that was interesting enough that you are interested in playing more than once.  Most use some sort of mechanic in which I choose my attack and you choose your attack (and perhaps defense) and then we ride toward each other.  Sometimes there is a mechanic that affects the outcome of the run based on impetus.  These games are fun for an hour or so between other games or as a filler at the end of the night when the main even ends early.  But they quickly become repetitive and boring.  I’ve collected many different sets of rules, and they are more similar than they are different.  I’ve been tinkering with the idea of using the diagram of a finite state machine to enable players to try to react to what they are seeing the other player doing as they run down the list.  The idea is something like this:  I initially had my lance high, but I see that you have your shield high.  From a high position, I can adjust my lance to medium high or medium left.  The finite state machine doesn’t have a transition from high to low.  You know that I see that your shield is high.  You have the option to keep your shield high or move it to medium.  What will I do?  Staying the same might mean that my defense if more effective, but it might be the wrong defense.  Moving might give me a better defense, but since I moved, I would get some negative modifier, since I was not “set” the whole time.  I might even combine this with a dice progression scheme as described previously.  This is an example of the theme in search of just the right mechanic.

My point of this rambling is that while mechanics and theme are both very, very important, you have to match the right mechanic to the right theme.  You don’t want to sit down to play a “serious” war-game and find that mechanics are like Munchkin.  (I like Munchkin, but you don’t want goofy artwork and silly double entendre when you are repeating Leipzig.)  Similarly, you don’t want to sit down to a Dr. Who game and find that you are tracking detailed physical damage (e.g., that bullet nicked Amy’s spleen, ricocheted off of something and exited between the third and four rib on the right side), ammunition usage (e.g, you have 9 bullets left in your 9mm automatic), fatigue, or a dozen other details.

In a couple of years, I plan to start on a project for the American Revolution.  (I have a few other projects to clear from the painting table and the game design queue first.)  I have been thinking about doing it in 10mm so that you can do the major battles on your kitchen table.  On the other hand, the sizes of the units are so small and vary so widely, that 25mm/28mm might be better.  When the time comes, I’ll start by reading several books that describe the Revolution at the tactical level.  Then I’ll go to a couple of conventions and play every RevWar game I can find to first see if there is a set of rules I like.  Maybe I’ll just use Brother Against Brother by Herb Gundt or an adaptation of Wellington Rules.  If not, then I will begin thinking about the period.  What are the salient ideas that make RevWar different from the wars that immediately preceded or followed it?  What mechanics can best bring out those ideas?

So, is it theme or mechanics that are important?  Both are equally important, but it is more important that the mechanics and theme fit together.