Hammercalled is probably the most complex roleplaying game I've ever made in the sense that it has a lot of moving parts, though I think that it should be relatively simple for players and GMs to run.

One of the things that I've been putting off for a while is talking about the framework for the story and how to make those things work together in the pursuit of storytelling, because I haven't generally been as focused on explaining to others how to get the system to work with their own tweaks. I've put out some information about the process I use, but it hasn't been systematic in any way. So, without further ado, I want to break down a lot of the major systems of the game and how to use them.

This is important because it shows how the game is designed to work; it's not necessarily a "do this or else", but it can help people who want to make their own Hammercalled derivative or add new rules and concepts to the system.

Abstract Mechanics

Abstract mechanics cover all the game systems that exist to serve as a foundation for play. The main mechanic in Hammercalled is a d100 percentile test.

Tests

Tests are the main way that things get done in Hammercalled, and they make up the bulk of the conflict resolution and emergent elements of the game.

A sample test is built using a number of character elements, which are stacked together to achieve a Target Number.

The primary reason why Hammercalled uses a d100 test is to allow it to have a high degree of granularity (100 possible results in a linear spectrum of mediocre to incredible).

This means that in general a test is entirely self-contained. The d100 system allows us to take three different pieces of information: raw total, margin, and individual die results, though we currently only use raw total and margin results.

Because of Hammercalled's philosophy, NPCs never make a test. Players may make a test, with a penalty or bonus, against an NPC. Occasionally they roll for initiative or to remove a status condition during combat.

Target Number Tests

The primary reason that you roll a die in Hammercalled is when there is a chance that an action fails, and this test serves as a way to determine whether or not a character can do something. Narratively, this helps to settle disputes between the GM and the player regarding a character's abilities, but it also adds some random chance and risk to situations and plays into the resource economy of the game.

TN Calculation

TN is calculated using up to six elements: attributes, talents, specializations, gear, Adrenalin expenditure, and circumstantial modifiers.

The first five elements are provided by players, and for most rolls only two or three are actually included.

Characters can potentially achieve all five during combat, but this is generally unavailable; most talents offer very selective combat boosts, and Adrenalin is a limited resource.

In practice, most of the time it is attributes, specializations, and gear that contribute to a TN outside of combat, and attributes, specializations, and Adrenalin expenditure that contribute to a TN in combat.

Talents that interface with the TN system are relatively rare; they tend to be a source of potential confusion, since a GM may not be as immediately familiar with them and players might not be used to how to apply them. Still, a straight bonus to TNs is useful for noncombat talents and allowing some special circumstantial benefits.

Circumstantial modifiers are at the GM's discretion, but their omission is fine in the majority of cases; the narrative focus of the game makes it so that the GM can use circumstantial modifiers to nudge players in a particular direction.

Currently the philosophy for combat is that NPCs that the player are fighting provide a circumstantial modifier, but testing has shown that while this is very mechanically effective in influencing outcome there isn't much evidence that players really seem to notice, indicating that this may be relatively meaningless in a story context unless the circumstantial modifier is rather large.

Of course, the use of circumstantial modifiers as a way to keep play interesting by rewarding clever behavior and discouraging risky behavior is still valid.

Mechanically, the role of a determined TN is very simple to assess by both players and GMs, since the raw number from 1-100 is the percentage chance of a task succeeding (assuming that the RNG used is fair).

Rerolling Tests

Rerolling tests is a strong resource for players and one that interfaces strongly with the basic TN system. Right now it is used to reward exceptionally clever or appropriate actions, but this may be added to or codified in the future.

Currently the main option for rerolling is a voluntary reroll after the first roll of a test. This impacts the probability calculations in a relatively uncertain manner, since there is currently not enough information to determine at what rates a player will reroll a successful test in hopes of getting a greater success.

The reason why we use this system is because the Margin is calculated using a blackjack system. This is not set in stone, but has a number of advantages. It's essentially a tens-place taken from a successful roll, which is a very easy mental calculation, and it permits the GM to have players roll against an unknown TN without revealing an estimated value by calculating a Margin using the difference of the result and the TN.

However, there may be some circumstances that have players roll twice and keep the best/worst result. When this happens, a handful of changes are applied to the probability curve of the standard TN test.

The expected value of the result increases or decreases as if there were a 17 point modifier applied to the test, and standard deviation is decreased by just over 5%.

What this means, effectively, is that a character trying to roll against a TN of 40 has a 40% chance of succeeding in general.

If they roll twice and keep the lowest, that likelihood increases to 64%, and the likelihood with keep highest is a mere 16%. It is not coincidental that the difference away from the normal chance of success is symmetrical (i.e. goes up when lowest results are kept, and down when highest results are kept by an equal amount), and the result is most significant by proportion near the extreme ends of the result set, but most significant by expected value near the middle of the result set, due to the way that the changes to the distribution and expected value are influenced by having two rolls to choose from.

In any case, the current reroll system is likely to be the one with the "best" value for players, with the chance to have an unmitigated success on the first roll and then abstain from a second roll, or a second roll to try to recover from a success when players have a low TN, and a chance to optimize results when players enter a higher TN bracket (i.e. reroll a 04 result when the TN is 83, but not when the TN is 20).

Margin

Hammercalled uses a "blackjack" margin system: the higher a roll gets without exceeding the TN, the better the Margin.

For the sake of simplicity, the Margin is simply the tens-place of the roll, but this can be modified in certain circumstances (e.g. the Critical Strike talent, which gives 1 bonus Margin on an effective attack; this change went in recently so you might need a new version of the Rules Reference, and it's not updated in the Quick-Start yet.).

In a similar vein, we utilize results over a certain number when determining whether gear degrades or not. This isn't particularly important, though it's important to note that this happens on attacks where the PC is going to be unlikely to suffer any other significant penalty for failure.

Doubles

One potential degree of resolution that comes up in Hammercalled is taking doubles.

Individual Die Results

Right now the Margin takes the value of the tens-place die. However, the ones-place can be used for other functions, though it is important to note a few important distinctions on when this would be advantageous and potentially deleterious.

For starters, a ones-place result is skewed by having a fractional chance of success. Because Hammercalled allows individual point variance in attributes, this means that any time you expect the result of 1-10 (or 0-9, depending on how the die is read) on the ones-place die to be strictly equal in chance of outcomes you need to remember that this doesn't provide equal chances of every outcome relative to the success and failure of rolls (e.g. someone rolling a 46 with a TN of 45 has a different experience than someone rolling a 46 with a TN of 46).

Opposed Tests

Opposed tests involve two player characters rolling against each other to attempt an action.

This is handled simply as regular tests against a TN, with successful characters comparing Margin.

There is a tie-breaker mechanic in the form of a character with the highest Attribute always winning.

Automatic Tests

It is strongly encouraged that players not roll for everything, but there are times when a player may roll when something is a foregone conclusion. This occurs only under two circumstances: Margin rolls, which are simply used to determine how well someone succeeds, and opposed rolls involving player versus player action.

Margin Rolls

Margin rolls are rare, and right now the system doesn't really use them explicitly, but they can be used by GMs in situations like wilderness survival where a tracker needs to use them to determine how many people they can care for, or perhaps as part of a climactic battle where the Margin decreases the time it takes to complete some crucial task.