Image by Keith Johnston from Pixabay

During play this Saturday past, I had reason to dissect a Save.

The entire process took only a few seconds at the time, thoughts following one on another so quickly that there was barely enough time to get a fleeting impression of one before it was chased out by the next.

I was helped in this process by the fact that it’s not the first time I’ve contemplated the subject. In fact, it’s something that every RPG designer knows quite intimately.

This prior experience permitted me to spare just enough mental capacity to recognize that each of those thoughts were mere surface impressions of much deeper subjects – and that, the next time I had to contemplate the subject, a quite different element within those subjects might be the one that’s relevant to the situation.

That means that every GM should have at least passing familiarity with the content of those thoughts so that when they need to do so, they have a foundation to build their own trains of thought upon.

1. Other Kinds Of Die Roll vs Targets

To start this discussion, I need to first address some laziness and imprecision of nomenclature. It’s quite common for all sorts of die rolls to be labeled as “saves” even when that terminology isn’t all that accurate, especially when the GM’s plot mandates that a check be made. These checks may be to comprehend a situation, receive a clue to a situation from one’s education, to carry out some obvious action, or for many other purposes – some of which I will describe later in the article (if all goes according to plan!)

For now, suffice it to say that there are two types of other die roll that are NOT necessarily saves.

Stat Rolls

The most like Saves are usually Stat Rolls. These are rolls that are employed to check on success or failure when a character is using native talent or raw capability, rather than an educated or refined expertise. Nowhere is this distinction more stark, more compelling, or more important, than when understanding the differences between Reflexes & Instinctive Reactions, Acrobatics (a trained expertise) and raw Dexterity.

Real life immediately complicates the situation; military training is aimed at implanting (amongst other things) certain triggers that cause a trained response to be employed instinctively. That’s often the difference between life and death for a soldier, and between the success and failure of an engagement. I can also accept the premise that training or native skill in certain expertises might imbue additional reflexes and instinctive behaviors – hunting comes to mind, as does the unconscious shifting of weight to retain balance on the deck of a sailing ship.

It’s also quite arguable that some instincts can be trained out of an individual. This is one explanation for why most people can overcome seasickness, for example. Certainly, most of us grow out of motion sickness in general, and pity those few unfortunates for which this does not occur even as we regard them – perhaps unfairly – as strange or unusual.

Skill Checks

The other major type of die roll is to determine the usage or application of trained expertise, usually in the form of skill checks. These are employed for two primary reasons: to determine how well something actually being done with the expertise, or to determine whether or not the character can recall and associate some theoretical, abstract, or learned knowledge or skill with the situation as presented to them. I am sometimes tempted to call the first “applied skill” and the second something else (such as “theory”) but that seems to shortchange the latter, suggesting that it is nothing but understanding of principles. The most accurate descriptive terms that I have found are “Applied Skill” and “Applied Knowledge”, to be honest – but that then gets hung up on some game systems defining some skills as “knowledges” and some not. This, of course, is the game designers groping around the same issues and trying to put in place a solution that distinguishes between book-oriented education and craft-oriented training.

Is it Really A Save?

Complicating the whole situation is that sometimes one of the above can actually be used as a saving roll. It all depends on circumstances and their interpretation within game mechanics.

In particular, one needs to contemplate the derivations of the various types of rolls, and what the different constructions might symbolize.

Stat rolls are a mathematical derivation of some measure of character ability.

Skill rolls are the total of a base score (usually derived from the stat roll of an appropriate foundation measure), plus an adjustment that reflects training and expertise. Further adjustments may reflect the quality and availability of useful tools and resources; while in other cases, it may be deemed that some tasks are simply impossible without certain tools or implements.

Saving Throws generally derive directly from a stat roll or from a different (but similar) mathematical treatment of a character ability measurement. The treatment that most strongly diverges from this fundamental principle was in an AD&D campaign in which the bases of the saving roll was the mean of two stats – INT and DEX for Reflexes, STR and CON for Fortitude, and WIS and CHAR for Will saves (the logic was strong in some cases and specious in others). Since this is the most extreme divergence from the principal that I can recall encountering, it’s fair to say that that it doesn’t diverge very far at all, and the principle is generally sound!

2. The Subdivision of a Stat

There have been attempts to subdivide stats into “more responsive” values that distinguish between different aspects of the rather broad and coarse stats that are commonly employed. I don’t want to burrow too deeply into that particular rabbit hole in this article, but some consideration is necessary. Because it’s relevant to the example that I intend to offer (the one which derived from last Saturday’s play and sparked the entire article), and because is the most definitive in terms of the differences between types of roll, I’m going to focus on Dexterity and the Reflex Save and why the two exist separately at all.

Dex: Fingers vs Feet

Dexterity conflates capacity in two quite different and distinctive areas – deftness with the fingers and agility. Ever since AD&D was published, people have been questioning that conflation. There is no doubt that the two are related, but they are also quite different and ability in one area does not necessarily translate into ability in the other. Quite the contrary!

The situation is rendered even muddier by the sloppiness exhibited in some early game designs in which the stat was labeled Dexterity but then defined as “acrobatic potential” or “nimbleness of foot” or something that clearly referred to Agility, or was labeled as “Agility” only to include in it’s definition “Deftness of fingers” or “Usage of tools”. I tend to be ruthless when encountering such, getting out a bright red pen and making corrections, often accompanied by exclamation points! But, for the purposes of this discussion, let’s ignore such sloppy designs.

I have even encountered at least one example in which the ability “to use a keyboard or numeric keypad” is part of the definition of Dexterity, as though that were somehow distinct from things like Carpentry. I, for one, have seen absolutely no evidence that the two are in any way related.

What most people mean by this stat has remained more or less unchanged since those days of AD&D.

Dex: Deftness, Agility, and Reflexes?

But the inclusion of a Reflexes Save, while an obvious requirement in some form or another, muddies this neat picture. The fact that this saving roll is also derived (in most systems) from the Dexterity score implies, in the eyes of many, that Reflexes are also a part of the melange that is the Dex stat.

What’s more, the breaking out of that save, in effect, partially subdivides the stat already – why not complete the job and subdivide the whole thing into its constituent parts?

The mechanics of doing so can vary, but they all boil down (in the end) to ensuring that the average of the three components equals the compound score.

This is true even if the rules require that the constituents be generated individually and then averaged to get the compound score. Though, the question then needs to be asked, if you have the three components, what do you need the compound score for?

The 3.x (compromise) standard: Dex to Dex and Reflex Save

I’m not entirely sure that the analysis in the preceding section has actually clarified matters all that much, though it has shed some additional light on part of the problem. To keep from getting bogged down or side-tracked, for the rest of the article I’ll be using the 3.x rules as a standard for discussion whenever it’s relevant, and contrasting with those rules as necessary. There are lots of reasons why this is a good idea, not least of which is that the ubiquitous of the 3.x/d20 system at least means that most readers will be familiar enough with it to understand the discussion and its relevance to them.

Why Subdivide at all?

I would argue that the compound score needs to be the basis of all its derivative scores so as to preserve the relationship between them – there might not be a direct correlation between deftness of hand and speed of reflexes, but there is a difficult-to-pin-down relationship – and by acknowledging that, and building it implicitly into the game system in the way 3.x does, you avoid the messy business of actually having to define the relationship and relevance of the components to each other.

But that only turns the question on its head, and takes us back to the issue that’s been lurking in the weeds this whole time: why subdivide at all? I have two answers, one founded in what the game rules are supposed to simulate, and the other in a far more metagame head-space – and, on this occasion, I think the metagame answer is the “real” one.

The “real world simulation” answer is that Deftness, Reflexes, and Nimbleness are all different things that are interrelated but not indistinguishable; so separating them permits greater fidelity of characters. You can have an acrobat who is all thumbs, or a carpenter who has two left feet but who makes the wood with which he works sing. Of the three, it also seems clear that reflexes and agility are more closely entwined than deftness and reflexes. If you assume that these two closely-entwined concepts are conflated for reasons of practicality, you end up with something that very closely resembles the 3.x handling of the question (which in turn derives from the AD&D version, and may even predate it)..

The parts of the picture that this overlooks are that the Reflex Save is never used as the basis of skills that are more Agility-oriented than Deftness-related. Of course, explaining the reasons for such a rules construction (which would be inherently confusing to an inexperienced gamer) would be difficult and possibly dull. The rules, as presented, avoid all that confusion by not treating a Save as a Stat.

The Metagame explanation has no such deficiency, and preserves the existing rules structure perfectly: the two are separated in the rules because it permits them to be treated differently in other parts of the rules.

Under this theory, you can forget all the deep naval-gazing about what the rules simulate or quantify or represent; they are simply ad-hoc constructs that yield a playable game, and the rules are a far more abstract representation of reality than most simulationists are willing to admit – or willing to accept, perhaps.

These decisions, and the thinking behind them, are inevitable in any game design. The questions posed are relevant to every rules system, only the specifics and parameters vary. The problem of how to represent reality in a reasonably-accurate-but-playable way are at the heart of every RPG rules system.

Nevertheless, the differences and distinctions between these different applications of the general concepts – Reflex Saves, DEX checks, and Skill Rolls – are essential to interfacing a character with a game situation, and have to be clearly understood by the GM so that he can ask for the right kind of check, if nothing else!

3. The Situation

The character in question was walking down a New York dock when he heard an insistent car horn beeping behind him. Turning, he saw a sleek black sedan racing down the narrow dock between the waterside and warehouses and many stacked crates being loaded and unloaded. Workers were forced to dive for cover to this side or that, onto or behind crates, or over the side and into the East River as the seemingly out-of-control vehicle careened toward the PC at high speed!

Let’s put this into context. This is a Pulp campaign, which means that dangers are hyped up, action is dramatic and energetic and stretches reality all out of shape. Think “swashbuckling” but in a 1930s context – something like Indiana Jones meets Jerry Bruckheimer production.

Solution Options

You may have noted that there was little in the way of detailed description of the scene – not enough for the character to even know what his options were. We rectified this situation by spelling out three obvious choices for the character, which also provided some of those details – a compressed form of writing that accelerates the action.

The options were (1) leap for cover behind some crates, with no idea what might be in them; (2) leap in the other direction, ending up in the river; (3) leaping for a hook attached to a crane, about 1m (about 3 1/4 feet) overhead, by bouncing off one of the aforementioned crates.

Always implicit is Option (4), Something else. It should also be noted that the distance to the vehicle, and its speed, are never mentioned. That implies that the options we have offered are the only ones for which there is time, and any “Option 4” solution must take only the second-or-two of duration that these require.

More context: an ordinary person can get about 1/3 of a meter (1 foot) off the ground in a standing jump. An athlete might be able to manage 1/2 a meter (about 20 inches). You can get a little higher with a running start, and a lot higher if you have a pole that you can use to translate your horizontal motion into vertical. The PC had no such aids, but did have the crate to use as a trampoline / launch platform, was naturally physically-capable, and “within his element” as it were. So two half-meter leaps were quite reasonable as solutions.

It can be argued that the player should have no say in determining a reflex action, but in the opinion of both myself and my co-GM, that takes too much free will from the character. Presenting options for the player to select between is a compromise that works for us. In a different campaign, where the “action standards” are different, the choices might be anything that the player can think of, perhaps within a real-time limit, or might be fully-dictated by the GM, subject only to a veto from the player that then has to be justified. Depending on the situation, I have used all of the above in my superhero campaign, for example, and have even permitted / required the character to make one or more skill checks to see if they think of any other (prepared) options.

The player immediately chose the most difficult third option, because it would permit him to drop after the vehicle’s passage and leave him in a position to take further action if it was warranted.

Once such a decision is made, regardless of who is making the choice, the next step is to engage the relevant game mechanics.

The Game Mechanics

The character in question had no acrobatics skill – something he may or may not choose to address in the future. Under the game mechanics (and most games will work in the same way), if you don’t have the relevant skill, the roll required defaults to a stat check; in this case a dex check.

This is simple enough in the game system that we were using, but would have been more complicated if we were using a d20 game system for some similar action (a runaway cart barreling down a hill toward a PC, for example), because everyone has a REFLEX Save. It can be argued that if you have the Acrobatics skill, that should take the primary role in determining the outcome – especially since we had given the character a choice of actions. “Volition means that this is not a reflex action” is the line of argument. But what if the character is not very good at Acrobatics, but has a very good Reflex save? Or perhaps the character should get a synergy bonus from the Acrobatics on his Reflex Save?

Personally, I find the Volition argument compelling – but interpret it as meaning that a character must have the relevant skill in order to have choices of action.

Sidebar: Should Characters Be Penalized For Not Having The Relevant Skill?

In most systems, this is the case automatically – to some extent. But I want to focus on two in particular as case studies.

In 3.x / d20, Saving Throws improve with character level and skills improve when skill points are allocated to them – and don’t improve otherwise, except from equipment modifiers and magic and the like. Because they are advancing at different rates, it’s easy for Saving Throws to exceed skill totals in relatively short order. Which brings back to prominence the question I posed earlier: if a character has both Acrobatics and a Reflex Save, which one should be checked in circumstances where both are relevant? And what if the other choice has a noticeably higher score than the one deemed most appropriate? One way to solve that problem is to assume that the two scores are on different scales and apply a correction – an increase – to the DC when employing the lower of the two. This is quite justifiable because of the distinction that can be made between a deliberate choice of action and an instinctive reflex. It can be quite plausible – depending on the action – that the difficulty of performing some tasks is much greater when they are performed instinctively compared to when someone with the appropriate training attempts to perform them. If the Saving Throw is 18 and the Skill only 14, then the same task might be DC19 when performed instinctively and DC15 when performed deliberately by someone who knows what they are doing. The implication of that is an answer of ‘yes, characters should be penalized with a higher DC if they lack a relevant skill, even when making a Reflex Save.’

Things are a little more complex in the Hero System:

— Stat saves are calculated as 9+(STAT/5), round in the character’s favor.

— For 1 character point (1 XP, aka 1 build point) the character gets 8 or less (written 8/-) in a skill.

— For 2 XP, the character gets 11/- in a skill.

— For a certain number of XP that varies from skill to skill (usually 3, 4, or 5) the character gets 9+(STAT/5), round in the character’s favor.

— For additional XP, the amount varying from skill to skill (usually 1, 2, or 3) the character can increase this score by +1.

— Checks are made on 3d6, with the goal of getting under the skill level. The GM can apply modifiers to the target to increase or decrease the likelihood of success.

That means that it’s routine for Stat Check values to exceed skill rolls in things that the character isn’t very skilled in, but might or might not be less in subjects in which the character IS very skilled – and you can’t put the Stat Check value up without also increasing the Skill Checks (all of them) that are based on that stat.

This can actually provide a disincentive for the purchase of skills if the character has a very high or very low stat value. The GM using this game system has to actively compensate with an incentive that matches or exceeds this disincentive – but there is absolutely nowhere in the rule book that tells you this, let alone gives you guidance as to what that should be.

For example, consider DEX 40 – a quite reasonable value in a superhero campaign for some characters. That gives a Stat Check of 9+(40/5)=9+8=17/-. Skill roll progression, depending on how much you spend on a skill, will be 8/-, 11/-, 17/-, more. You can argue that unless you intend to put a skill roll into the “more” category, you are better off spending nothing on the skill and defaulting to the 17/- skill roll.

I counter this with two – no, three – propositions:

— First, there are some “everyman” skills that everyone has to a minimal standard unless they take a disadvantage to give them a deficiency. These skills are at 5/- for no points.

— Second, some tasks are only possible are only possible if you have 8/- in the relevant skill, some are only possible if you have at least 11/-, and some are only permitted if you have spent the points to get better than the two basic levels of expertise. Other tasks are equally possible regardless of skill level, but take twice as long at the 11/- level, four times as long at the 8/-, six times as long at the 5/- level, and ten times as long if attempted “unskilled”. Furthermore, any difficulty penalties are doubled if you don’t have the appropriate skill.

— Third, I am a big fan of “quality of success” / “degree of failure” assessments – and some “quality of success” outcomes are off the table unless you have the appropriate skill.

The latter two propositions are applied depending on whether or not success will eventually be achieved with persistence. In my home-brew system, as used for my superhero campaign, many of these details are explicitly stated within the rules, but are all employed in a common-sense manner that is so predictable that it’s rare that I need to consult the rules for any specific skill. In the Adventurer’s Club campaign, we might occasionally use the “greater time required” option, but for the most part, we will do it all with quality of success, which is assessed with an awareness of the probabilities of a 3d6 roll.

Thus, missing a target by 1 will usually be a lot worse if you only have 11/- in a skill than doing so if you have 17/- in that skill – and worse still if you are attempting to use a skill that you don’t actually have.

So the answer to the question posed by the section title is an emphatic but unofficial “yes” in the case of the Hero System. The specifics may vary, but there is always a benefit to spending the points required to buy a skill, regardless of how good your “unskilled default” might be.

The Odds Of Success

Even though I’ve labeled the discussion as a sidebar, the above provides essential context to the application of game mechanics to the proposed task at hand – leaping to a hook on a crane 1m overhead by using a crate as a stepping stone.

In this case, the character does not have the skill deemed relevant – Acrobatics – and the game system subsumes Saving Throws into Skill Checks. Nevertheless, by the time his general competence and skill levels were taken into account, he had a 17/- chance of success on 3d6, less any penalties for difficulty that we GMs chose to access.

If we consider rolling or leaping to the side to be an essential step to all the different solutions we had placed before the player, they were all of similar difficulty magnitude.

The Roll

The player rolled three sixes – an automatic failure – and that meant that we were directly into interpreting the magnitude of his failure and what actually happened. It was this outcome that triggered the thought process described at the start of this article.

4. Moments Of Interpretation

Many rolls don’t require interpretation, for one reason or another.

Success

For example, if the character succeeds, on many occasions he simply achieves whatever he set out to do. Narrative need only begin with the consequences. This is a Simple Success.

There are times, however, when a simple yes/no outcome is not good enough. A character painting a portrait, for example, or laying out a book, or just about anything creative or craft-related – These require some assessment of Success Degree.

Failure

Likewise, there are lots of times when failure simply means that the character did not succeed in whatever they were attempting – Simple Failure. But, equally, there are times when Degree Of Failure is a relevant consideration.

Criticals

To some extent, this practice is merely nuancing something that many GMs already implement in their games – Critical Successes and Failures. It’s simply adding more shades in between those outcomes and a “bare minimum” success.

Relative Interpretation

Once you have multiple levels of success open to you, the next thing that you need is some method of selecting between them. There are many possible approaches to this, but the simplest is to simply consider the degree of success or failure revealed by the die roll. If you need 14 or less, and you roll 17, that’s a failure by three. If you take the relative probabilities of those outcomes into account, it’s a considerably bigger difference on 3d6 than it is on a d20. So, depending on how many nuances you have broken the skill into, you might interpret this as a far worse failure than if you had needed 11/- and rolled a 14. Or vice-versa. It’s all a matter of interpretation.

There are other ways of indexing possible outcomes. Distance from a critical failure, for example, places all the possible results on a continuum, and means that if you succeed by enough, you effectively get the benefits of a critical success even if you don’t roll one – a six might be good enough. Or Distance from a critical success. Either way, you are committed to either bunching results together into ‘bands’ of rolls that lead to the same outcome, or into creating many more layers than you probably need.

For my superhero campaign, this no problem – explicit interpretation structures are woven into the description of each individual skill, once again based on a common-sense interpretation of the situation. For the Adventurer’s Club campaign, that’s not much help.

This gives a great deal more flexibility to the campaign, but also places a premium on improvising outcomes.

5. The Anatomy Of A Reflex Action

So, looking at this particular roll and situation, the immediate response should be to dissect the proposed course of action. Many of these will be the same every time, regardless of the situation, especially if one adopts a less literal interpretation. This is important because each represents a potential failure point.

a. Awareness

The character needs to become aware of the cause of a need to act on instinct.

b. Reaction

The character needs to associate that cause with the need to act on instinct.

c. Opportunity

The character needs to subconsciously evaluate possible responses.

d. Target

In this case, the character needs to aim himself at his proposed landing point on the crate.

e. Leap

He then needs to carry out the leap to that targeted point.

f. Landing

There are times when landing will end the instinctive action of a leap this way or that. But not this time – instead, the character needs to land, dropping to his haunches, still balanced, and ready to leap off again.

g. Rebound

That second leap can be considered the rebound of the character. It has to be in the right direction and deliver enough upwards motion to reach the eventual target. You will note that there’s no re-targeting – no time, it all has to happen in one smooth motion – the ‘aiming’ of this motion was done back in step ‘d’, and the intermediate point reached in step ‘f’ is only a means to that end.

h. Grab

It’s quite likely that the hook has a motion of its own; the grab is not just to hold onto it, it’s to get it into a position to be held on to.

i. Hold

Finally, the character has to hold himself in place long enough for the danger to pass beneath him.

6. The Metagame Factors

Before any assessment of how badly the character fails, and what happens as a result, takes place, there are always some metagame considerations that have to be taken into account. These always function as mitigants, if they have any effect, but they can also rule some potential failure modes out of the question.

How important is it to the campaign that the character succeed?

This is often a critical factor in terms of restraining the GM from being remotely as lethal as he could be. In this case, it immediately ruled out killing the character for all sorts of reasons – inappropriateness to the genre (not being heroic enough for a PC) being a main one.

How important is it to the adventure that the character succeed?

While there would have been several ways around the problem if the character were to be left temporarily infirm, they would all have involved a player sitting around twiddling his thumbs for the next several game sessions. My Co-GM and I both consider this unacceptable, and rewrite adventures repeatedly if necessary to avoid it. “Everyone plays, everyone contributes” is our goal for each adventure (and each game session). So this ruled out incapacitating the character.

How important is it to the encounter that the character succeed?

Which leaves us with a minor injury or inconvenience as the absolute worst-case scenario. Within the bounds of possibility remaining, we needed the car to get past the character, one way or another, in order to complete the plot-relevant actions we had plotted for it – after that, we didn’t care one way or another. So a twisted ankle on landing? Okay, a little inconvenient adventure-wise, but that’s as bad as it can get.

Information explicitly conveyed

By describing the threat, we explicitly conveyed that the character had succeeded in steps (a) and (b) – he knew that there was a threat and that he needed to take immediate action to get out of the dangerous situation.

Options Explicitly Offered

Furthermore, by explicitly offering a range of options, and getting a decision from the player, we had explicitly ruled out a failure of step (c). We had presented the results of the characters assessment.

That meant that the failure lay in imperfect action of some sort. I instinctively went for the most dramatic option – the character leaped, but underestimated the motion of the hook, grabbing it only with their little finger, which could only hold for an extremely painful half-second or so. Just long enough, in fact, for the car to pass beneath the PCs’ feet.

7. How Long Did The Hold Have To Be?

Before I could announce that, however, it needed to pass a plausibility test.

How fast was the car going? How long was it? If I didn’t like the results, I would have to amend my first instinct and have the character land on the top of the vehicle and then try to grab hold.

We’re talking 1930s, so 60-70mph was a reasonably good top speed – but it takes time to get up to that sort of speed, and time translates to distance – and there wasn’t all that much distance on the dock. Even giving the car an initial speed as it screamed around the corner onto the dock, it wouldn’t be anywhere near whatever it’s top speed was.

Having said that, the car was clearly a powerful one, probably with a V12 aircraft engine (based on the length of the hood in the image we had chosen), so it wouldn’t be slow, either. I couldn’t see that it would be much faster than 50-55 mph – or much slower than 40 mph.

Fifty is the most convenient of these numbers to work with, easily translating up to 55 (take off 10% of the time) and down to 40 (add 20%). So that’s where I started.

The image on the screen was about 6.5 inches in length, with the width of the vehicle estimated to be about three inches, allowing for the somewhat side-on perspective. I also knew that this was a Bentley, a quite wide and heavy car, better known these days as a luxury vehicle but considered a performance car at the time. So, roughly 1.3m wide (4 1/4 feet). Which gives a length of 2.82m (9 1/4 feet).

9 1/4 feet is 0.0017390152 miles, according to Google. A car traveling at 50 mph will take 0.0017390152/50 hours to pass a given point, given it’s length = 0.000034780304 hours = 0.00208681824 minutes = 0.1252090944 seconds. Call it an eighth of a second.

Even tacking on an extra 20% to that isn’t going to be very much. The PC could hang on for as little as 1/4 of a second and be well clear.

8. Interpretation Of A Failure

Given the metagame considerations that were in play, and the high native ability of the character – you don’t expect to fail 17-or-less rolls on 3d6 very often, it’s well under 1% of the time – it also seemed reasonable that the character would be able to do most of what he had attempted. 17/18ths of it, to be exact! But that would put the point of failure on the holding on to the hook, and that didn’t seem reasonable if the character got any sort of decent grip on it. So, logically, the grip had to be the point of failure.

But if the grip was the complete point of failure, he would have dropped to land right in front of the oncoming car – an outcome ruled out by the metagame circumstances. So I would need the grip to be an almost complete failure, and the hold to be an ever-so-slight success – effectively taking part of the failure from step (h) and transferring it to step (i).

He wouldn’t have to hang on for very long, after all. So I ruled that he grabbed on with his little finger only, and that his pixie could only hold him for half a second.

Hanging On By His Little Finger

Of course, the character can’t start his leap as the car gets to his position, he’s going to have to anticipate it’s arrival to be out of its way. But if he leaps when it’s about twenty feet away, that doesn’t increase the travel time to much more than 3/8 of a second. So 1/2 a second would be plenty, and he would land a couple of feet behind the spare tire.

I also factored in that this character is one of the physically strongest of the PCs – his preferred ‘soft weapon’ is a capstan bar – a heavy iron bar over 2′ in length and more than an inch thick. If anyone could hang on by his pinky for half a second, it would be this PC.

Reasons To Interpret A Failure (or a success) When You Don’t Have To

#1: In a word, Flavor. Without interpreting success or failure, you have sterile outcomes that provide no toe-hold for consequences beyond the immediate success or failure. Such interpretations provide a foundation for flavor text that transitions from game mechanics to roleplaying within the game.

#2: I’ve mentioned a number of times the degree to which game mechanics is disruptive to the pacing, tone, and intensity of a game. Anything that softens or counters that impact is a significant improvement.

#3: It gives you a hint, a starting point, and that makes the job of devising the flavor text easier. So it can save you work in the long run.

#4: It adds an enhanced sense of realism, even to the unrealistic and fantastic.

There are probably more, but those are enough to be going on with!

9. Other Kinds of Save

That brings to an end the example of the process of dissecting a Save. Along the way, I’ve looked at when you might need to do so, and why it’s reasonably desirable to do so. But almost all of the discussion has been framed around that example, and Reflex Saves. With that in mind, let’s look at the different kinds of save there are – and these are almost ubiquitous, system to system, regardless of the specifics and mechanics involved.

FORT saves (passive)

Fortitude saves are generally made to resist some ill effect or to resist the onset of an undesirable condition. That can include exposure to diseases and environmental pathogens; it can include resisting blacking out from high g-forces; it can include withstanding shock, or blood loss; or simply maintaining activity despite extremes of temperature.

A key point about FORT saves is that they are passive. The character doesn’t have a choice about making the save – it’s completely involuntary, a test of how the character is reacting to the adverse conditions that have triggered the role.

This contrasts markedly with Reflex saves or their skill substitutes like Acrobatics, in which a character can attempt something knowing that it will trigger a Reflex save or other check.

WILL saves (passive)

WILL saves are also passive, and used to test a character’s determination despite adverse conditions, more than anything. That includes the determination to do what the character wants and not yield to some external mental influence. A lot of the time, WILL saves are underutilized; for that reason, when GMing, I never require a FORT save when a WILL save is just as appropriate.

PERC checks

Okay, so now we’ve moved beyond the “official” types of Save and into the realm of checks that GMs frequently use as saves. One player I know describes Perception checks as “Save Vs Surprise”, and in many ways, the term is appropriate.

Perception Checks come in two flavors, and for metagame reasons, it’s usually necessary to treat them completely differently.

Passive checks deal with the character noticing something without an active examination of the scene or locale. Example: PCs discover a building with a huge number of carved demonic images on its walls as reliefs, heavily overgrown. After they spread out to find the entrance, each has the chance to notice that one of the images blinks every now and then. This is a passive Perception Check. Obviously, the GM can’t have the players roll these checks, or if he does, they have to be made without the players knowing what the rolls signify; anything else tells the players that there IS something to perceive, which means that they will keep looking until they find it.

Which brings me to Active Perception checks. These are usually quite separate from, and different to, a search roll, but that depends on the game system. If a PC says, “I’m looking closely at the gargoyle with the bright red eyes – what are those eyes made of?” then they are using an Active Perception Check – specifically looking at something and asking for more detail about what they can see.

I can’t speak for anyone else, but I absolutely hate having nothing to say when an active perception check is successfully made. It immediately makes the game world’s artificiality obvious – and makes it look as though I haven’t done my homework, to boot. Of course, you can always make something up on the spot off the top of your head – but that’s even more likely to get you into trouble because the rest of the adventure that you have planned makes no allowances for your little exercise in unbridled creativity. I solve this particular problem by carefully visualizing whatever the PCs are looking at before I start to answer, assuming that I don’t have more information prepped for delivery.

Another example of an active perception check is “This looks like good country for an ambush – I’m looking for any signs of trouble waiting for us.” That does no good if the ambushers are invisible (a particularly nasty trick to use) but under normal circumstances, you would have to think very carefully about whether or not there was anything to see before you could give an answer. (I’ve sometimes gotten good mileage from deliberately understating or extremely heavy-handedly overstating the absence of anything to make the players paranoid).

Sidebar: Extra Senses

Again, I don’t want to get too deeply into it – this article is quite long enough already – but it would be remiss of me not to at least mention the issue of characters with unusual senses that most characters don’t have. Does this confer an advantage when making a perception check? Does it only do so when making an active or a passive check? Does a Perception score function identically and equally for all senses? Those are just some of the minefield of issues waiting to blow up underfoot when extra senses come into play.

Oh, and while I”m within shouting distance of the subject: how about ordinary senses other than sight (spot) and hearing (listen)? Say, flavor, or enhanced odor discrimination?

Comprehension (INT) checks (passive vs active)

An NPC explains something complex to the player character, simplifying it as much as possible because the PC doesn’t have the relevant skill. The test for whether or not the PC understands – and how much they understand – is a Comprehension Check.

These come in two flavors – there are the comprehension checks that deal with implied statements and subliminal messages – “reading between the lines” activities, in other words – and there are comprehensions that come from reading things, or listening to lectures or speeches – “face value” activities. Since the first involves actively analyzing what you have heard and relating it to other information, that’s an “active” comprehension check, while the second is clearly a “passive” check.

There’s no official practical difference between the two, and many rules systems don’t even actively list this application of an Intelligence check. The unofficial difference would lie in the relevance of other related skills, in particular something like deduction or detective.

Idea (INT) checks (passive vs active)

You know a player is really stumped (or has given up) when he utters that immortal phrase that every GM encounters sooner or later: “Can I roll to have an idea?”

Sometimes the GM can see this coming and simply has the player make a roll to get a hint without being asked – which enables them to weave the question into the game narrative more strongly.

Idea checks are prime candidates for “will eventually inevitably succeed” interpretations.

Education checks (Skills) (passive vs active)

And then there are skill checks that are actually saves. I’ve already discussed active skills like Acrobatics so instead I want to move directly onto passive skill use, where checks are used to associate flavor text with significance, beyond that which is possible for the player alone. For example, a character might describe the animated corpse of a creature with the ears of an Elf and glowing blue eyes. While the character has a skill level in an ancient society – call it Atlantamuria – since this is an original creation of the GM, the player doesn’t know very much about it beyond what the GM has already revealed. So it would take an Education Check on the relevant skill to determine that the description matches a creature from Atlantamurian legend.

Like a passive Perception check, the GM can’t ask for the roll directly without giving the game away. And that’s why these checks should be considered Saving Throws and not ordinary skill checks – they are handled differently, or should be.

The same can’t really be said of active Education Checks (“My character knows Golthrokbin Legends, can I make a roll to see if there’s anything similar in their mythology?”) – they get included only for the sake of consistency.

10. The Save vs Uncertainty

One of the worst things to do is to hold up the game while you attempt to make up your mind about something that isn’t all that critical. If all your prep is for the city of Castlemaine and the PCs make a unilateral decision to head to Gracemeyre chasing some off-the-cuff rumor that the GM tossed out last week, taking pause to think is justified; but if it’s a saving throw, there should be no more than a couple of heartbeats between the announcement of the roll result and your delivery of an explanation. Equally, if the situation demands a saving throw be made, awareness of how to handle that save – both written and unwritten rules – should be as close to instinctive as you can make it. You have to make your (metaphoric) save vs uncertainty.

The more you’ve thought about this stuff in advance, the less time you have to spend thinking about it at the game table. You have to be able to dissect a save at lightning speed, and use the results to make a decision as quickly as the player can roll his dice. This article gives you all the tools that you need; how you use them is now up to you.