Thursday, February 19, 2009

Degree of Success

I've been thinking recently about D&D skill systems and why they are sometimes clunky. One reason is confusion between determining Degree of Success and just determing a binary success/failure.

In combat, you use two rolls. The to-hit roll (usually on a d20) determines binary success/failure. The damage roll (usually a D6, D8, etc) determines the degree of success.

However, skills are deemed to be less important so they generally get just one roll, presumably to speed things up. However, some skill checks are binary (go/no go), but others determine degree of success. Here's some examples:

1E Thief Skills: Generally binary. You either roll under, or you don't. Some modules occasionally say, "if the thief beats the % by X, then something happens" but that's the exception, not the rule.
1E Open Door Rolls: This is one check where degree of success enters the picture, at least for characters with Exceptional Strength. They have a chance to break down even a specially sealed door if they roll well.
1E Secondary Skills: No resolution mechanic included.
1E Other Checks: I often times used a saving throw as a "skill" check mechanic for things I wanted to scale with level. This is just a binary yes/no check.
3E/4E Skills such as Diplomacy, Intimidate, etc: These are usually binary. You either hit the target DC and have the desired effects, or you don't.
3E/4E Skills such as Athletics, Jump, Climb, etc: These usually include Degree of Success. The Athletics check determines how far you can jump -- you don't just get a binary success/failure. Failing a climb check by too much means you fall.

This has a few problems.

First, its confusing. Not all skills work the same way. Ever notice how play grinds to a halt when using an Athletics check to jump? A D20 is just not a good die for generating the results desired in this case, and at least in 4E, you end up having to divide and do mental math. Additionally, some skill checks end up being much more important than combat to-hit rolls. For example, a skill check to leap over a lava pit is more important than the numerous to-hit rolls against a monster. But because of the relatively lower fidelity in skill checks, players are often hesitant to gamble it all on just one die roll. Also, rolling skill checks on a D20 makes the variability very high. As has been discussed, players usually dislike variability because it means that they still have a good chance to fail at things they should be good at (example, a swashbuckling rogue with athletics/acrobatics maxed out might still fail to leap from a chandelier and do the classic banister slide move, so he'll probably just walk down the stairs instead -- boring).

I've been toying with using my D6-pool system to generate skill check results. For example, say a character that has a rank of "2" for jumping (however the system does that) rolls 2d6 and takes the highest one. The result is the number of spaces he jumps. Easy, still down to one roll. It reduces variability significantly, making players more likely to try to use skills in combat (breaks up the monotony). The trick would be to figure out how to convert all "skills" to a Degree of Success check instead of the mixed binary/degree of success system we have now. Perhaps a return to something like the 1E Open Door roll would make sense -- a 6 or higher might indicate "great success."

Tuesday, February 10, 2009

Odds of Success

When considering game design, one should think about how often the avatar/PC/player should succeed. If the players fail too often, then they will get frustrated. If they succeed all the time, they will get bored. Finally, there's a connection to reality. Depending on what the game is trying to model, wildly bizarre results may strain credulity; for example, it may not be fun to play a game that gives an untrained but very dexterous individual a 95% chance of success for, say, brain surgery.

Casinos are well aware of this. For example:
"Customers at slot machines receive frequent, small prizes at irregular intervals. This encourages continued betting in the hope of getting a big jackpot." (http://www.scientificpsychic.com/workbook/gambling.html). Scientists have looked into this phenomenon to examine the frequency of wins that a player needs to receive to stay "hooked." Here is one such paper (http://www.camh.net/egambling/archive/pdf/JGI-issue11/JGI-Issue11-turner-horbay.pdf), and another (http://www.camh.net/egambling/issue21/04haw.htm).

It looks like many slot machines reinforce about 1/6 to 1/3 pulls (that is, you win 1/3 to 1/6 of the time, at a minimum). In blackjack, I found these odds: Odds of dealer winning - 45.85% Odds of a tie - 9.53% Odds of a player winning - 44.62%. Additionally, consider the role of "near misses" -- that, is times when the player ALMOST wins. One more factor is that of nearby wins -- notice that slot machines are always in a row of at least 5? That guarantees that with 3 players sitting there (man space...), one will be winning almost every pull. At a blackjack table with 4 players, someone will win almost every hand (obviously sometimes the dealer gets a 21 or busts out everyone, but that doesn't happen every time).

Switching this to gaming, a Dragonsfoot Poll asked the question, "% chance of success when attempting a task should be:" is to gaming (http://www.dragonsfoot.org/forums/viewtopic.php?f=11&t=27372&hilit=chance+to+succeed)

The results are reproduced here:
33% trained / 0% untrained (i.e.: d6 roll, 1-2 tr but no untr) 0% [ 0 ]
50% trained / 0% untrained 7% [ 2 ]
45% trained / 15% untrained (i.e.: base Siege Engine check) 10% [ 3 ]
66% trained / 33% untrained (i.e.: d6 roll, 1-4 tr or 1-2 untr) 41% [ 12]
75% trained / 25% untrained 17% [ 5 ]
other 24% [ 7 ]

Note that the most popular result is 66/33. Remember our slot machine odds? Pretty close (for the untrained figure, at least!).

Now let's look at some real world figures. The Armed Forces have requirements for marskmanship with various weapons. These requirements have two objectives that I can think of: (1) practically, to ensure that soldiers can fire their primary weapon with some degree of success and (2) psychologically, to encourage the soldier to feel confident in his ability to use lethal force, and provide a means to reward him for being accurate and eager with his weapon. The services don't want to set the bar too high, or else nobody will pass, which is psychologically damaging and bad for morale -- plus its expensive to have all those remedial fires! You can't set the bar too low, though, or else it poses a danger to the improperly trained soldiers themselves and makes those who are actually trained frustrated because its too easy to succeed.

Here's the AF standards:

EXPERT: 43/50 0.86
A: 32/50 0.64
B: 25/50 0.5
C: 19/50 0.38

Note that group A are people who need to use the skill on a regular basis (security forces, for example), so they have a higher standard. Arming group C are people who do not need to shoot often (office employees), so they have a lower standard. Expert is a meritorious category that anyone can be rewarded for attaining. Note the Category A figures falling right around that magic 2/3 area identified in our DF poll which is good for a "trained" person. Category C falls just about dead on our 1/3 area used for "untrained" persons, and also matches up with our untrained slot machine chances of winning.

Here's the Army Standards:
EXPERT: 36/40 0.9
SHARPSHOOTER: 30/40 0.75
QUAL: 23/40 0.575

In the Army, everyone uses the same standard, but there are two tiers of meritorious performance instead of just one like the USAF "Expert" category. Note the qual and sharpshooter figures falling right around that magic 2/3 area identified in our DF poll, again good for a "trained" person.

Obviously one would need to do a study to find the ideal frequency of wins for addiction/satisfaction, but I think the 66/33 rule of thumb is a good rough hack. It ensures that in a group of ~5 players, one will succeed at the task most of the time (if all are untrained, at least one will succeed about 87% of the time). It matches up with real world marksmanship odds. Its easy to generate on standard dice. It rewards training without making untrained attempts absolutely fruitless.

Monday, February 2, 2009

Ironclaw Dice Mechanic

The Ironclaw system uses an interesting variation on dice pools. Multiple dice are rolled, and the result is the highest number showing on any of the dice. With just that, the system would be logically equivalent to a dice pool system where only one success is ever needed.

Where Ironclaw differs, however, is in allowing the dice to be of different types. The interesting effect here is that adding a small die to a pool with bigger dice does not increase the maximum performance possible (the highest number that can be rolled), but does make it less likely that a very low number will be rolled — thereby making easy tasks less likely to be failed.

For example, consider a character with a Dexterity rated as 1d12 (for anyone not familiar with RPG dice notation, it's (number of dice)d(size), so that's one twelve-sided die). The character can perform tasks with Dexterity with difficulty as high as 12 — they'll succeed at such a task 1/12 of the time. On the low end, this character can succeed at a Dexterity-related task with a 4 difficulty on 3/4 of their attempts.

Now, let's give the character a tiny bit of training at, say, lock-picking — enough to give them a lockpicking skill with a rating of 1d4. The character's chance of picking a difficulty 12 lock has not changed. We can say that their training doesn't cover locks that tough, so they have to rely on their innate talent. However, against a simple difficulty 4 lock, the character will now only fail if neither die rolls a 4 or better — giving roughly an 82% chance of success, up 7% from what they had before. A bit more training will raise their lock-picking skill to 1d6, at which point they'll succeed against this lock about 88% of the time.

Ironclaw has a "fumble" rule when all 1's are rolled on the dice for a task — if that happens, then something's been messed up badly. Increasing the size of a single dice doesn't decrease the fumble chance by much; going from, say, 1d10 to 1d12 moves fumbles from 10% of the time to about 8%. However, adding extra dice decreases the chance a good deal; a character going from 1d10 to "1d10+1d4" (that's Ironclaw's notation, even though the dice aren't actually being added) has gone from a 10% fumble chance to 2.5%.

Although Ironclaw doesn't make use of it, such a system also can offer a choice for advancement — expanding the limits of one's ability, or increasing the reliability of it on low-difficulty tasks. That is, someone with, say, a 1d8 skill could have the choice of advancing to either 1d10 or to "1d8+1d4". The former allows the character to do things that they couldn't before, while the latter makes the character's skill more reliable on low-difficulty tasks.


FYI I stole the above text from somewhere else (sorry, no link).

This would be good for an RPG. But it also works great for, say, a wargame (with some modifications). All units on the battlefield might have Attributes: Training, Supply, and Morale. They might have skills like "Bombard" (artillery), "Engineering," "Attack," or "Defense." You could either use the system as is, or flip it around, so you use the LOWEST value -- i.e., an Elite Trained Unit with High Morale is still limited if it has poor supply. Or perhaps the second lowest unit.

Playing Cards

I've recently found out a few interesting things about playing cards.

First, the traditional associations of the suits.

HEARTS (chalice) - Clergy (Clerics?) - Water
DIAMONDS (coins) - Merchants (Rogues?) - Earth
CLUBS (clubs in the agricultural sense, or wands) - Agricultural (druidic or wizardly?) - Fire
SPADES (swords) - Military (fighters/warriors?) - Air

Obviously those all lend themselves nicely to classical fantasy archetypes. Whether it be the core classes, key magic items, or the classical elements (and all the symbolism THAT entails!) they relate pretty nicely.

Next, each of the face cards has a historical allegory in the French tradition. I won't list them all here (check out Wikipedia for the list) but there's lots of room there for magic items (Charlemagne's sword, anyone?), on-the-fly NPC development, etc.

Also, some of the face cards are notable. From wiki:


Though specific design elements of the court cards are rarely used in game play and many differ between designs, a few are notable. The Jack of Spades, Jack of Hearts, and King of Diamonds are drawn in profile, while the rest of the courts are shown in full face; these cards are commonly called "one-eyed". When deciding which cards are to be made wild in some games, the phrase "acey, deucey, one-eyed jack" (or "deuces, aces, one-eyed faces") is sometimes used, which means that aces, twos, and the one-eyed jacks are all wild. The King of Hearts is the only King with no mustache, and is also typically shown with a sword behind his head, making him appear to be stabbing himself. The axe held by the King of Diamonds is behind his head with the blade facing toward him. This leads to the nickname "suicide kings". The Jack of Diamonds is sometimes known as "laughing boy"[15]. The King of Diamonds is traditionally armed with an axe while the other three kings are armed with swords, and thus the King of Diamonds is sometimes referred to as "the man with the axe" because of this. This is the basis of the trump "one-eyed jacks and the man with the axe". The Ace of Spades, unique in its large, ornate spade, is sometimes said to be the death card, and in some games is used as a trump card. The Queen of Spades usually holds a scepter and is sometimes known as "the bedpost queen", though more often she is called "Black Lady". In many decks, the Queen of Clubs holds a flower. She is thus known as the "flower Queen", though this design element is among the most variable; the standard Bicycle Poker deck depicts all Queens with a flower styled according to their suit.

There's possibly some work that can be done there, especially with the Queen of Spades!

The cards themselves could be useful for quite a few things. I like the idea of using them as a play aid. Perhaps standing in for magic items, action/hero/karma point rewards, or character powers. For example, when buying an item, maybe you just draw X cards (determined by character charisma, size of the local market, price paid, etc) and choose your favorite -- you don't always get what you ask for! They could also be plot cards. Given the connotations of the suits/face cards they could stand for all sorts of plot twists in a vague tarot-sort of way.

Just some musings...

Variability in Dice Mechanics

I just read about an interesting technique for RPG dice rolls. Instead of manipulating difficulty, instead, manipulate variability.

For example, take a task in a standard D20 system. Generally, if performing that task under stressful conditions, the DM will either increase the target number (TN) or apply a penalty to the player's roll (mechanically, the same thing). This can have a chilling effect -- if the player judges the odds of success to be too small, they will probably not try at all for fear of failure.

For example, say a D20 character is +5 and is attempting a TN 14 task. They have a 60% chance of success -- not bad! Say conditions are terrible and the DM imposes a -5 penalty (or raises the TN to 19). The odds of success have decreased to 35%. If the penalty for failure is significant, most players won't even make the attempt.

What if instead, we could adjust the variability of the attempt?

I see three ways to do this.

1) The FUDGE system uses special dice labeled +1, 0, and -1. By adjusting the number of dice in the FUDGE pool added to the base roll, you adjust how variable it will be.
2) You can have a "good" die and a "bad" die (both the same kind of die). Roll both and retain only the lowest die. If you retained the good die, add it to the base roll. If you retained the bad die, subtract it from the base roll. By using a large die such as a D20, you create a highly random, variable result. If you use a small pair of dice like a D4 or D6, you have reduced the variability. The advantage of this method is that you retain the original possible range of results.
3) You should be able to add a fixed modifier and reduce the die size. For example, you could have 1d20+MOD vs. a TN. Or, you could have 1d12+4+MOD vs. the same TN. In the first case, the Minimum is 1+MOD, the maximum is 20+MOD, and the average is 10.5+MOD. In the second case, the minimum is 5+MOD, the maximum is 16+MOD, and the average remains 10.5+MOD. You've just moved the tail ends of the distribution closer to the center.

Usually, the players have the upper hand. After all, if they didn't, then most heroes would be short lived indeed! Thus, they are interested in reducing variability. The underdog is usually interested in increasing variability. Only through higher variability does the underdog have any chance of winning.

For example, take a poorly skilled character trying to hit a difficult TN (it could be a kobold trying to hit a high level fighter's AC). They may need a base roll of 18, 19, or 20 in order to actually succeed. If they use the lower-variability D12 method, then they CANNOT succeed. That lowly skilled individual wants the scenario to be as variable as possible!

Here's another example. A common D20 rule is allowing someone to "take 10." This is the ultimate in reducing variability; you're basically rolling 1d0+10. Most novices are not interested because they can't succeed. However, an expert is usually perfectly happy to take 10 -- they know that they will likely succeed and that variation just gives them a chance to fail.

If your dice mechanic allows the DM to adjust variability instead of difficulty, it gives them an interesting tool. In our original scenario, say the DM increased the variability due to difficult circumstances. A novice still has similar odds for success, and indeed, may even benefit from the increase in variability. But the expert is no longer facing a sure thing. This "levels" the playing field, which is handy for challenges that the entire party is expected to meet.

Dice Rolling Articles

http://www.rpg.net/columns/rollthebones/rollthebones1.phtml

http://www.darkshire.net/~jhkim/rpg/systemdesign/dice-motive.html

Principles of RPG design

PRINCIPLES OF GAME DESIGN

I've been thinking and reading a lot about good game design, and I think I've identified a few principles that should be present in a good game.

- Rule of Three
- Magic Sevens
- Granularity
- Odds of success
- Simple Math
- Essential Elements of D&D

THE RULE OF THREE

Good things tend to come in Threes. Check out Delta's article on the subject at http://deltasdnd.blogspot.com/2008/03/1-2-3-infinity.html.

Note how WOTC has incorporated three into many elements of 4E. How many encounter and daily attack powers do you get? 3 (except PP powers). Your third encounter of a day is a milestone, and most typical sessions probably only take you to the third or MAYBE fourth encounter anyways (and if you go four, one of them was likely a cakewalk/quick). There are three tiers of play. If each session has three encounters and you throw in a quest or two, it takes about three sessions to level up.

Sticking to the rule of three keeps development manageable. It has a defined start, middle, and end. Progression is quick enough to be pyschologically satisfying.

Thus, I think the Rule of Three should be considered for use for elements that (A) should feel special or (B) are related to linear development (plot, session pacing, mechanical growth, etc). The mechanics of the system should help GMs use the Rule of Three to make more pyschologically sound scenarios.


MAGIC SEVENS

Human memory seems to have a thing for 7 +/- 2 chunks of information. If we are presented with a choice of more than about seven options, we can't comprehend all of them at once. See Delta's article on the subject at http://deltasdnd.blogspot.com/2007/04/magic-number-seven.html.

Note how WOTC has implemented this in some ways in 4E. A typical combat lasts about 7 +/- 2 rounds. A typical monster often goes down in 7 +/- 2 attacks.

When selecting powers, there are usually no more than seven choices per level (including MP).

Some places it is poorly implemented. For example, feats initially present far more than seven options. This requires the player to winnow down the list.

I find myself first eliminating feats I don't qualify for then chunking the remainder ones into offensive, defensive, and flexibility lists. I then decide which element I want to boost and then select from my sub-list. Equipment is another place it tends to be poorly implemented. There are far more than seven different types of viable magic armors or weapons which makes it tedious to hunt through all of them. I find that I don't have much patience any more for tracking GP or XP or even HP. I do it, but its tedious. I definitely need paper and pencil or a calculator. It is much easier to track AP or daily magic item usages remaining or even my stable of magic items (because I usually have 7 +/- 2 items!).

Magic Sevens should be used anywhere that a human is presented with a choice. It is also good for longer situations than the Rule of Three can handle. For example, one could either model combat on three 3-round "phases" or just use the Magic Sevens principle to steer combats towards seven rounds. The player should never be presented with a list of items longer than 7 +/- 2. Some exceptions might be made for round numbers like multiples of 10, as we are usually
quite familiar with the base 10 system. It might also be acceptable to use Imperial units of measure in some cases due to people's familiarity with them, even though they don't break down quite so nicely (12" to a foot, for example -- although many Imperial units are designed to be easy to use, such as a League!).

IMPERIAL UNITS

I like Imperial Units. On one hand they have an immediate weakness -- they're hard to convert. On the other hand, they're easy to use. Many are related to real-world experiences. For example, a league is about as far as you can walk in an hour or so. A stone is about as much weight as you can easily lift with a hand. A pint is what you drink at one sitting. Wherever they can be logically worked in, I think they are helpful.


GRANULARITY

Granularity refers to the amount of precision used in any mechanical check or decision point. See Delta's article at http://deltasdnd.blogspot.com/2009/01/granularity-in-checks.html.

This is one thing that 4E does not do well. Almost all checks are on a D20. This is good for consistency, but it implies that an out of combat streetwise check is as important as landing a life-or-death daily power. It forces the DM to call for rolls out of combat only when they are very significant, even though a DM might want to use dice to settle other situations. 4E is better than 3E in this regard because some powers and items allow auto-success; for example, for movement, many PCs will have a way to fly or teleport instead of relying on tedious skill checks.

One addendum for Delta's Article. Note that surprise was initially determined using D6s. As Gary's players reached the higher levels (D series instead of B2), surprise became more and more important. Notice how those later monsters (drow, spectators, duergar, etc) reduce the granularity of surprise checks from a D6 to a D8, D12, or even D100. As surprise became more important to life or death, it naturally deserved a more finely determined roll.

In general, I think one should use the smallest dice possible. One should use larger dice only when (A) the level of significance increases and (B) you intend to fully use the level of granularity provided (i.e., don't roll on a D20 but make all your modifiers or adjust target numbers in +3 chunks -- otherwise you might as well use a D6 and give a +1 modifier!).


ODDS OF SUCCESS

A poll on Dragonsfoot awhile back indicated that many respondents wanted about a 2/3 chance to hit when swinging -- that is, when doing something they're good at. Note that in original 1E, a strong character opened a door on 3/6 or 4/6, a sneaky character surprised others 4/6 (elves, halflings, thieves, rangers taking precautions). 66% are good odds pyschologically. One can push the envelope up to 75% and still retain tension. Once the odds get up around the 90% point there's not much point in rolling -- that's the point when you are just seeing if a "1" comes up.

SIMPLE MATH

The system should use the simplest math possible, especially for frequently repeated or time critical operations.

The easiest operation is to compare two numbers and see what is greater. After that, addition is easiest. Subtraction is also pretty simple. Multiplication is more complicated, and division is hardest. Single digit numbers are easier to deal with than double digit numbers. The more numbers are involved the harder it gets ("Well, I have 1d8 weapon damage, +2d6 sneak attack, +4 DEX, +3 magic weapon... etc.). Precalculation can help with some of this.

4E is terrible with this. I can't imagine playing above heroic tier without use of the computer or a calculator right in front of me.


ESSENTIAL ELEMENTS OF D&D

For a fantasy clone to feel like D&D to me, it needs certain elements. I guess these are "sacred cows."

- D20 rolls in combat.
- Fighters, Clerics, Wizards, and Thieves.
- Certain Races.
- Level-based system.
- Medieval weapons.
- Appropriate milleou.
- 3+1. By this, I mean for every 3 combat encounters, there should be 1 RP encounter/trick/trap/puzzle, in general. The original 1E is an outgrowth of

Chainmail, and later versions always reflect that wargame bias. A game like WoD is the opposte, 1+3 (1 combat encounter for every 3 RP/trick/trap/puzzle).

Obviously there are exceptions, but I think 3+1 works well for a typoical session worth of material.