Today I’ll address in detail some of the elements of a well-made common language system, elements that are, in my opinion, applicable to virtually any system design. In particular, we’ll look at intentionality, which forms the philosophy of the system’s design, and the central mechanic, which forms the core of the system’s execution.

Intentionality and Game Design

The first of these is Intentionality, which simply means always keeping your goal in mind and discarding anything that does not help you achieve that goal. A necessary corollary is drawing together as wide and deep a knowledge base as possible so you don’t spend most of your energy “discovering” things that have already been done before. The process of learning is a continual one, and you will inevitably find yourself editing and improving your common language system as you go. I believe that the process of learning widely about a subject, then focusing on what works (with no regard for ego or the way things have always been), can be applied across the range of human endeavors, from martial arts (as Bruce Lee did with Jeet Kune Do) to business, art, and roleplaying.

Intentionality requires knowledge. So where does that leave those of you who haven’t played a lot of different games and systems, but who want to come up with a good common language system anyway? First, I’d suggest that if you enjoy designing games, you continue to do so, just for the joy of creating. Create as many different kinds of games as possible, and focus on small efforts you can finish rather than epic games that will bog you down and take all of your energy. Try to write games you wouldn’t ordinarily attempt, just to shake things up inside your head.

Second, I’d suggest you read as many systems as possible. There are a large number of games available for free download on the internet, and there are also a great many professionally or semiprofessionally done games available quite reasonably in the rpg.net store. Don't neglect rules-light systems, even if that's not your usual style. I found a lot of innovative and useful mechanics ideas for the Tarafore System while reading and writing rules-light systems.

Get a feel for the each system, focusing on the following: how the mechanics work, how user-friendly they are, and whether they deliver what they promise. Make some sample characters, and see if you can figure out how to break the character creation system. Try to make some unusual but appropriate characters – see if they system breaks down if you try to make a talented newcomer, a skilled but physically slowing veteran, or a person with low ranks in a wide variety of skills. Ask your gaming friends to do the same.

Run some playtests in which you not only get a feel for how the system works, but whether or not you can break it. If possible, run a one-off using the game, to see how things work “for real.” Lather, rinse, and repeat for as many games as you can. Also, spend some time in the various gaming and game design forums, both here on rpg.net and elsewhere.

Intentionality of Design Each of the major elements of a common language system – the central mechanic, the scale of the traits, skills, and difficulty numbers, how you define your traits and skills, combat, and of course character creation – need to be viewed critically with a mind for what will work well in practice with minimal fuss. With that in mind, we’ll look at the central mechanic.

What does a common language system's central mechanic need to do?

In my experience, these three are the most important:

1) Make life easier for the players and the GM. Mechanics should be clear, easy to understand, and mathematically coherent. The GM should be able to calculate the odds of a character succeeding at a given difficulty even if his last name isn't Hawkins, Einstein, or Spock. Players should be able to use the system without any real problems, and with a gentle learning curve. The system should be consistent and predictable (if the GM doesn't want players to know their characters' chances of success, she can hide the difficulty numbers from them, but the system itself should make sense), and should not make itself the center of attention.

2) Provide the level of detail your group desires. Some groups are fine with broad levels or benchmarks of skill (Average, Good, Very Good), while others want fine detail (Average(50) vs Average(51) in a percentile system, for example). We've found the old 10 average, 3 points is a benchmark scale to be ideal for our tastes, but your mileage will most likely vary. But detail isn't just about the Trait Scale: a game that only gives binary "succeed or fail" results will be less flexible and less effective than one that gives several qualities of success and failure. On the other hand, most groups don't need twenty discrete degrees of success and failure, either.

3) Adapt to different settings and genres easily. A flexible, rational, unified mechanic will be easier to adapt to new settings by adding subsystems and additional rules (even additional traits, if necessary).

Central Mechanic

As I said in the last column, the Tarafore System’s central mechanic involves taking the appropriate Trait, adding 1d10, then subtracting 1d10, yielding a range of Trait-9 to Trait+9, with the average result being equal to the Trait itself. With this type of mechanic, every 3 points of Trait is effectively a major step – the character with a 3 point advantage has a 72% chance of winning, a 7% chance of tying, and a 21% chance of losing. This is a major but not insurmountable advantage, and I consider it appropriate for a benchmark.

Why?

Why did I choose this particular mechanism?

1) It’s easy to use for both players and GMs. While any extremely math-phobic might blanch at the addition and subtraction at first, the learning curve is gentle (especially if you’re using a dicebot). The coherency of the mechanic, which gives results on the same scale that attributes are rated, makes it easier to visualize just how well a character is doing. At no point does it require complicated math, frequent referencing of tables, or adding a big pile of dice together. The GM can easily figure out the likelihood of a character succeeding and set appropriate difficulty levels.

2) It makes every one-point increase in ability pretty much equal (a character with 47 Strength has the same chances against a character with 46 Strength as a character with 7 Strength has against a character with 6 Strength), adding to clarity and coherence.

3) It gives the same chances of success regardless of which character “rolls for it.” In other words, if an NPC attacks a PC, the GM can just say “Defend against a 12” and the player can roll the dice and report the result. This has become something of a sticking-point for me, because GMing a fight with four PCs and a small army of NPCs (some under the PCs’ command, some hostile to them) can be a real pain if the GM has to roll for every attack or defense. It’s much easier to just let the players roll the dice for the things their characters are involved in, and let me handle the dice for NPC-NPC actions.

Basic, Special, Exceptional

The “three point step” feeds into the quality of results. A tie represents either stalemate or a marginal success, depending on circumstances. A margin of success of 1-3 is a Basic Success, a margin of success of 4-6 is a Special Success, and a margin of success of 7 or greater is an Exceptional Success. The reverse holds true for failures, though I tend to avoid “fumble rules.” In Opposed Tests, there are no Special or Exceptional failures, just Special or Exceptional Successes for one of the characters.

The Odds

The Tarafore System mechanics produce simple probabilities that are easy to visualize and calculate – with two ten sided dice, there are 100 possible combinations (10 x 10), so the chances of winning, losing or tying all come out in nice round percentages.

Odds in Percentages of Different Results Based on Differences in Trait Ratings

(These are the odds of success or failure for the character with the higher rated trait).

(Abbreviations: Diff = the difference between the higher and lower Traits, F = Failure, S = Success, Exc = Exceptional, Spc = Special, Bas = Basic, E+S = sum of Exceptional and Special Success/Failure chances)

Diff	Exc. F	Spc F	E+S F	Bas F	Total F	Tie	Total S	Bas S	E+S S	Spc S	Exc S
0	6%	15%	21%	24%	45%	10%	45%	24%	21%	15%	6%
1	3%	12%	15%	21%	36%	9%	55%	27%	28%	18%	10%
2	1%	9%	10%	18%	28%	8%	64%	28%	36%	21%	15%
3	0%	6%	6%	15%	21%	7%	72%	27%	45%	24%	21%
4	0%	3%	3%	12%	15%	6%	79%	24%	55%	27%	28%
5	0%	1%	1%	9%	10%	5%	85%	21%	64%	28%	36%
6	0%	0%	0%	6%	6%	4%	90%	18%	72%	27%	45%
7	0%	0%	0%	3%	3%	3%	94%	15%	79%	24%	55%
8	0%	0%	0%	1%	1%	2%	97%	12%	85%	21%	64%
9	0%	0%	0%	0%	0%	1%	99%	9%	90%	18%	72%
10	0%	0%	0%	0%	0%	0%	100%	6%	94%	15%	79%

As you can see from the chart above, each single step is significant - even a 1 point difference in ability raises the chance of victory by 10%. A "benchmark" step of 3 points gives the character with the advantage a 72% chance of winning (27% Basic Success, 24% Special Success, and 21% Exceptional Success).

As the difference between Traits increases, the chances of losing decrease (of course), but so do the chances of tying and getting a Basic Success. If a character has a 9 point advantage (the difference between the fighting Prowess of a legendary warrior like Myamoto Musashi and a basic professional soldier), he'll have a 72% chance of getting an Exceptional Success (the same chance a character with a 3 point advantage has of succeeding at all). At the 9 point range, the chance of Basic Success has dropped to 9%, and the chance of a merely Special Success has dropped to 18%. Musashi can’t lose an exchange of blades with Joe Ashigaru, but he does have a 1% chance of getting a tie (as we’ll see in the combat section, a tie can still end the fight, if the damage test goes the right way).

For Example...

Some tests are pretty binary – either you succeed or fail, and margin of success doesn’t much matter. Other times, such as in combat (and any social interactions you choose to handle mechanically), the margin of success is vital.

For example, if Loki wanted to bean Thor in the head with a water balloon full of shaving cream, he’d first have to sneak up on him. That would be a standard Opposed Test, presumably between Loki’s Awareness(Sneak/Hide) of Outstanding: 19 and Thor’s Awareness(Perception) of Very Good: 16. Loki’s player would roll the positive d10 and the negative d10 and add them to Loki’s Sneak/Hide. So if he rolled +5 and -4, he’d add them to 19 and get 20. That’s more than enough to beat Thor’s Perception 16, so Loki gets into position. It doesn’t really matter that Loki got a Special Success, because all we needed to know was who won.

Now we see if Loki can hit Thor. Since Thor is not aware of Loki’s presence, he won’t be able to dodge the incoming water balloon, and Loki will be able to use the base Range Difficulty Number. We’ll say Loki got within about ten feet of his burly barbarian brother. That’s Point Blank Range, with a Range Difficulty Number of Easy: 7. For the moment, let’s ignore any “called shot” rules and just let Loki throw that balloon. We haven’t covered this yet, but in combat, you need an Exceptional Hit to hit someone in the head. When you’re using an actual weapon, there are damage bonuses for Special and Exceptional hits, but they won’t apply to water balloons.

Loki’s player rolls an Unopposed Test of Loki’s Prowess(Throwing) Very Good: 15 against the Range Difficulty Number, which is Easy: 7. Loki’s player rolls +9/-10, which gives him a total of 14. That’s 7 more than the difficulty number, an Exceptional Hit. The water balloon sails out of the Wagnerian shadows and coats Thor Odinson in a messy lather of waters from the sacred river Vimur and Bermashave.

Let’s hope Loki’s “Running” Trait is similarly godlike, else Thor will give him a wedgie fit to shake the halls of Valhalla.