Cinema Film by Alexandre Saes

On any number of occasions, I’ve referred to using a Cinematic combat style instead of the “full treatment”, but I’ve never gone into detail of how I go about that. I’ve explained why, but never how. (just in case, I’ll recap “why” as we go along).

I’ve always resisted doing so because I felt that the techniques that I use are too dependent on the game system that I am using, and so they would be of limited utility to anyone not using those systems.

It’s taken me a long time and a lot of thought to catch a glimmer of how I might be able to sufficiently abstract my techniques to give them a broader applicability, and even now I’m not sure that I’ll succeed. But I definitely won’t do so if I don’t give it a crack!

This is an article in three parts. Part one will deal with simplifying your game mechanics for attacks; part two deals with abstracted mechanics for damage; and part three will deal with the complete absence of traditional combat mechanics.

So here we go…

Reasons For Cinema

There are three central reasons why you might want to choose a more cinematic combat style.

1. It increases the Drama of the conflict.
2. It makes Combat secondary to Roleplay & Non-Combat Action.
3. It permits greater manipulation of the look and feel of the situation.

Drama

Cinematic Combat is essentially throwing away selected game mechanics and replacing them with narrative. Because the resulting mechanics are greatly simplified, Combat goes faster and is more dramatic, especially if the GM manipulates his narrative to emphasize that drama.

Combat vs Roleplay & Non-combat Action

Cinematic Combat leaves greater capacity for non-combat activities and elevates these to be equal or greater in importance than the combat that is taking place alongside such activities.

Look and Feel

And, by replacing relatively flavorless mechanics with flavor-rich narrative, the GM can adjust the content and style of his delivery in order to control the look and feel of the encounter.

Justification Examples

Whenever the fight should not be the center of attention, cinematic combat is the preferable approach. A group of PCs trying to hold off hostile forces while another PC attempts to solve a puzzle, pick a lock, reprogram a missile, hack a computer, defuse a bomb, negotiate with an enemy – the list is endless – for example.

Or perhaps the outcome of the combat doesn’t matter as much as the amount of time it takes while something else is happening, or how the PCs behave during combat, or the combat itself is relatively trivial compared to the fact that it has taken place – trying to prevent or intercept an attacker who is trying to reach a target protected by the PCs.

Whenever the PCs are up against a deadline, by the end of which the combat has to be resolved – indicating that the Pace of the action is more important than the action itself – cinematic combat is at least worth considering.

Or perhaps the need is to emphasize a particular look-and-feel rather than being slavish to the game mechanics. Barroom brawls, a duel while the combatants balance on a piece of rope stretched between two masts, a PC and group of NPC grunts vs another group of NPC grunts, combat between two starships using a game system that doesn’t explicitly cater for it, a dogfight – these are all valid reasons to employ cinematic combat.

What To Throw Away

So Cinematic Combat is all about throwing away game mechanics and substituting something simpler. The question is therefore “what to throw away?”

Cinematic Timekeeping

The first thing to get tossed aside like a chew-toy is any detailed timekeeping. I’ve actually reported previously on my adapting a D&D 3.5 Initiative/Combat-Sequence approach to the Hero System (Taking the initiative with the Hero System). You can’t afford a high level of granularity in a cinematic combat sequence; you want the fight to Flow. This is so much a sine-qua-non of cinematic combat that it can be taken as read.

The Elements Of Combat

After that, it gets more complicated. Combat Systems – all of them – essentially come down to Tactical Assessments, Number Of Attacks, Attack rolls, Defensive or Target Comparisons, Resolution, Damage Calculation, and Consequences Assessment. Attack Rolls may have Critical Hit and/or Fumble sub-steps. Damage Calculation may have Hit Location complications, and may also be different based on those attack roll variations.

The question can be simplified by uniting these into logical groups. Tactical Assessments usually modify attack rolls, so they can be combined, and an attack roll is nothing without the Defense/Target comparison and Resolution. And Number of attacks is just more of the same. Abstracting all of that into a single Combat Element makes perfect sense to me.

Because Critical Hits alter the Damage Calculation, and Fumbles are the other side of the coin to that, those three elements can be considered a second Combat Element. The same basic argument also integrates Hit Location into the damage assessment.

That leaves Consequence Assessment to stand alone as a third combat element, though you could argue that Critical Hits and Fumbles should be in a group with it. Either way, we’ve – in theory – simplified combat mechanics down from nine elements to three, and that’s good enough for the first level of abstraction.

But we’re a long way from being able to actually accomplish that. How do you combine these elements? What are the effects on combat? What are the traps?

The Attack Element

This is actually a lot simpler than it might seem. There are four elements to be combined into a single abstract quality. The trick is to remove the complications, building everything into a single die roll. For many reasons, the best choice of die roll to work from is a d20; it’s linear probability nature, and degree of granularity permits substantial finessing of results.

Relative Attack

We start with the character – PC or NPC – who is both participating in the combat sequence and who has the lowest attack value, ignoring any tactical or circumstantial modifiers. I tag his score as “Base”. Everyone else is rated relative to that score. If you are using a game system that already employs linear probability to determine combat results, that is a simple subtraction; if your game system uses 3d6 then you might need to take into account the “bulge” in probable result centered on the 10.5 mark; but that’s much more complicated, so we’ll deal with it separately.

What you want to know is an approximation of the linear-probability gap from the base combat score to the individual scores of all the other combat participants. If there is a substantial gap between the scores of the PCs and those of the NPC combatants / monsters, you may need to choose separate base values. In the case of multiple attacks at different attack values, as happens in the d20 system, D&D, and Pathfinder, ignore the other attacks and just calculate from the first attack score.

For example, assuming that higher is better, if the PCs have d20 attack scores of 15, 16, 13, 9, and 10, the base score is 9, and the characters have values of +6, +7, +4, +0, and +1, respectively. Jot them down on a piece of paper, a notepad, or a whiteboard. If lower is better – and some counterintuitive game mechanics operate that way – then the base value (ie worst attack value) is 16, and the modifiers are -1, -0, -3, -7, and -6 respectively. The way to tell: “higher is better” translates into “roll this or more to hit”, while “lower is better” translates into “roll this or less to hit”.

Continuing the example, and making the same assumption, if the NPCs/monsters had attack scores of 12, this fits nicely into the same range of results, so use the same base roll and a value of +3. If the NPCs/monsters had an attack score of 2, or of 25, though, it would not fit so comfortably in the same range; you would end up with values of -7 and +13. The first can be lived with, but just barely; the second is more problematic.

Relative Defense

Attack rolls are meaningless until compared to a defensive target. Some systems add the attack value to a die roll to compare with the target, others require the gap between die roll and attack value to exceed the defenses of the target in order to penetrate the defenses. There are other variations; the Hero System adds a fixed +11 to the attack value and subtracts the defensive value of the target to determine the number that must be rolled on 3d6. There are a lot more variations in defensive simulation “theory” than in attack models.

We don’t care about any of that. All we need is a relative assessment of the defensive or target values presented by the different targets – a base value and an assessment of how much better or worse the other values are. If the combat system is non-linear, we also need to account for the potential for the better defensive values to shift combat target values to the “wrong side” of the probability bulge.

Depending on which yields the most convenient results, you can either employ a “highest equals base” approach or a “lowest equals base”. A “lowest=base” approach produces negative modifiers for the better defenses, reducing the relative attack values, sending some into the negative; the “highest=base” approach yields further pluses to attack, which is often easier for GMs to calculate but may be harder to translate into meaningful results. As a rule of thumb, you don’t want the total to exceed the size of your die’s range of results – a d20 has a range of 19, so that’s the highest that can be accommodated. However, in a non-linear combat system, 6 of that range have to be set aside to accommodate the probability bulge, as discussed below – so the biggest range that can be accommodated is 13.

My approach is therefore to consider the “highest=base” approach first, and if the highest combination has a total of 20 or more (14 or more in a 3d6 system), use the “lowest=base” approach.

There’s a slight twist in the logic applied above that might not be clear to the casual reader.

A better defense means that it’s harder to score a successful hit; that means that a “highest=base” approach means that all relative values to the base make it easier to score a successful hit, and therefore add “+modifiers” to the relative attack chance.

Conversely, it also means that using the “lowest=base” means that the modifiers reflect a reduction in the chance of a successful hit, and therefore operate in the opposite direction to the relative attack modifiers; those were “+modifiers”, so the relative defensive values have to be “-modifiers”.

For example: Defensive values of 16, 20, 22, 14, and 17. “Highest=Base” gives a base of 22, and modifiers of +6, +2, +0, +8, and +5. The highest attack modifier that we have from the d20 (linear) combat model is +7; combining that with the highest defensive value of +8 gives a +15 total. This is well within our 19-point range, so that works fine, and this will be the case 99% of the time.

The end product

We have defined a “Base vs Base” result that is the worst attacker vs the strongest defender. If a character has a better attack, he will have a +modifier to the likelihood of successfully hitting that defender; if the defender has a weaker defense, whoever is attacking them will also have a +modifier to the likelihood of successfully hitting that defender.

You could draw up a table showing all the combinations, as shown to the left, but that sounds too much like work to me, and it’s totally unnecessary. All you need is a pair of lists: PC1 A+6, PC2 A+2, PC3 A+0 (and so on), and Tgt1 D+6, Tgt 2 D+7, Tgt 3 D+4, and so on. Of course, you’ll also need attack and defense values for the NPCs attacking the PCs – but that’s simply a matter of extending the lists.

When an attack is made, the attacker simply rolls a d20, the GM mentally adds the attacker’s attack modifier and the defender’s defense modifier, and interprets the result.

Ahh, if only it were that simple.

The complicating bulge

The biggest wrinkle to be tackled is the non-linear nature of some combat systems’ die rolls – typically 3d6, but there are all sorts of variations. That means that there is a probability “hump” around the average roll result. If your required roll is higher than this, and you have to roll higher than a target, your chances of success are considerably lower than a strict linear accounting (such as I’ve been using) shows. If your required roll is higher, and you have to roll less than a target value, the hump significantly boosts your chances of success.

While an exact accounting of the changes is far too complicated and messy to be practical, some notional adjustment is needed – a tip of the abstract hat to the greater or lesser chances of success. Because the actual adjustments vary too much with specific game systems, I don’t think it’s possible to offer a general solution that is applicable in every case, or even in most of them.

I can offer some broad advice, however. What that comes down to is a three-step procedure:

• Assess a base-attack-vs-base-defense attempt to hit – what needs to be rolled for success?
• Use the result to assess where the base-attack-vs-base-defense combination falls on the 3d6 curve, relative to the “hump”. On 3d6, I consider results of 9-10-11-12 to be that “hump”.
• A better attack will move the hump one way or the other of the succeed/fail division point; a better defense will move it in the other direction.
• Use this information to assess each combined adjustment (attack and defense) at the time an attack is made; use a quick rule of thumb (given below) to assign an adjustment to the attack chances that gives a rough approximation of the corrected chances to successfully hit. A “+modifier” represents an increased chance to hit; a “-modifier” represents a decreased chance to hit.

So, to that rough rule of thumb:

• 9 to 11 or 12, or 10 to 12 – i.e. moving from one side of the hump to the other – is worth plus-or-minus 1.
• 7 or 8 to 11 or 12, i.e. moving from outside the hump to a chance that includes the hump is worth plus-or-minus 2.
• 6 or below to 11 or 12, i.e. moving from very early on the curve to include the hump is worth plus-or-minus 3.
• any change from 11 or 12 to 13 or 14 is worth an additional plus-or-minus 1.
• any change from 11 or 12 to 15 or 16 is worth an additional plus-or-minus 2.
• any change from 11 or 12 to 17 or better is worth an additional plus-or-minus 3.

An Example

That might not make a lot of sense without an example. So let’s say that we have a “roll X or less” combat system, and that our base-attack-vs-base-defense combination attempt to hit requires 7 or less to hit. This is slightly to the left of the hump, indicating a relatively poor chance of success. A better attack will increase this number, so more of the hump will act to improve chances to hit; a worse defense will do likewise. Assuming a “higher=base” defense assessment – which is what I prefer to use, because it’s all addition – that means that assessing what the modifier due to non-linearity is simply a matter of getting the total modifier.

So, if we have a +5 attack (PC 5) and a +4 defense (NPC 3), indicating that PC 5 is attacking NPC 3, we have a total of +9 from a starting point of 7, i.e. a shift from 7 to 16 on the table above. “7 or 8 to 11 or 12” gives +3, and the further change from “11 or 12 to 15 or 16” is worth an additional +2. So the total modifiers for this particular attacker/defender combination is actually +9+3+2=+14.

non-linear Practicality

To be honest, I know the 3d6 probability curve well enough that I don’t bother with the rule of thumb given above; tell me “7 or less” and “+9” and an answer of “+5 more on a d20” pops straight out (That’s one of the benefits of Gaming for 34 years). What I’ve described above is the best approximation of my subconscious number-crunching that I can capture.

Tactical Modifier

Tactical Modifiers are now simplicity itself. The GM simply takes ALL the circumstances, in aggregate, into account, and decrees “+0” or “+1” or whatever feels right. DON’T let a player begin to rattle off “book values” or mechanics: “I’ve got reach, I’m attacking from behind, by surprise, I’m flanking him, and I’m invisible so that’s a modifier of…” To all such, the answer is “I’ve already taken all that into account”. You can even consider giving the enemy attackers an extra +1 if the player repeats his litany. The character is already in as advantageous a tactical position as he can get, so far as you are concerned, and the modifier you’ve mentally assigned is appropriate to that determination.

Number Of Attacks Modifier

Some game systems give characters multiple attacks at decreasing chances of success. Some give a character multiple attacks at the same attack value in a given time frame according to some stat.

Here’s the truth about abstracting such situations: An increased number of attacks gives a better chance of at least one of them hitting, so the character gets a bonus to hit. I use +1 per extra attack per +5 or less of other modifiers. And it increases the average total amount of damage done, so that needs to get taken into account when abstracting the damage part of the combat.

Example Continued

The highest combined modifier we have is +15. If the attacker in that case has two extra attacks, he gets +3 attack modifier to represent each of those extra attacks (+15/5=+3). If the total was only +14, it would be +2 attack modifier per attack.

The Universal Success Target

Success is a modified roll of Twenty or more. This is the target for ALL characters – apply the case-by-case modifiers to the actual die roll to determine the outcome of an attack – success or failure.

The upshot

The net effect of all this is to take the entire mechanism of determining whether or not a target has been damaged to a single yes-or-no determination based on whether or not the player rolls a target number or less.

It takes only:

• a second or two to list the attack and defense values of the combat participants, per participant;
• another second to identify the lowest attack;
• another one or two per participant to list the differences between this value and the other participants;
• another to identify the highest defense, and one or two to identify the lowest and get that difference, telling you whether or not you can use the preferred “highest=base” defense approach, or need to use the “lowest=base” approach; allow one more to interpret the result;
• another second or two per participant to list the differences between the base defensive value and the defensive values of the other participants.

Total prep time for combat between 5 PCs and 5 NPCs: 10-20 plus 1 plus 10-20 plus 1 plus 1-2 plus 1 plus 10-20 equals a grand total of 34-65 seconds.

But it’s in conducting combat that the real benefit emerges. Instead of identifying and analyzing who-knows-how-many tactical modifiers, rolling a die per attack (or 3 dice and getting a total), applying (possibly different) combat values to each, looking up the defensive value of the target, comparing each total to that value, and interpreting each result, there is ONE roll, to which two or three modifiers are added, and an interpretation of the result. You don’t need a second-by-second breakdown to see that the abstraction is a LOT faster.

I didn’t consciously set out to dedicate Mondays to writing article series and Thursdays to standalone articles, but I’ve found that it’s a lot harder writing two series at once. So – for now, at least – that’s the pattern that I find myself in. But that’s not a bad thing – consistency of subject on the one side balanced with something with a greater chance of finding favor an audience who aren’t into that series on the other.

The alternative – seriously contemplated – was to use both Monday and Thursday for a series, so that whatever the subject, it gets dealt with in half the real-world time. But sometimes you need that extra time up your sleeve; a number of times I’ve only been able to get the next part in a series finished by scheduling a relatively short and simple article for the other part of the week, freeing up time to work on the series. So this seems the best compromise. But it will only take one filler article when the next part of a series is nowhere near ready, and the pattern will swap ends of the week.

So the plan is to present part two of this series, dealing with abstracted damage handling, next Monday…