A Time Based Combat System

Let's talk about initiative systems and how Virtually Real's solution works. This is just one aspect of the combat system, not an exhaustive discussion of the combat system itself.

History of the TTRPG Combat

Rather than a comparison and contrast of all the systems available, let's begin where all this started, the first Dungeon's & Dragons system. This was written by wargamers, used to battling with scores or even hundreds of men. The individual fighting styles of an individual didn't really matter in the scheme of things. What mattered was how many men you had and a roll represented a lot of action, all being abstracted. Initiative was not really a mechanic. You just rolled a die to randomly figure out which side went first.

Individual combat was just a slight modification to the rules and designed to fit with an existing mass combat system. These "heros" were the generals and major players of the combat, exceptional people. Sound like D&D yet? A round was an entire minute, driving home the abstract nature of combat. Actual turn order was just going around the table.

Then they decided to drop the round to 6 seconds. It was supposed to still represent multiple actions in combat, but the feel was a single attack. This was made worse by terminology like a "to hit" roll, which speaks to a single attack missing rather than multiple attacks that were parried or glanced off armor, or did minor damage. Initiative was also individualized to allow some characters to react faster than others while adding more attacks to represent the additional speed of experience and grant additional player agency. This mash-up of abstractions doesn't quite mesh. A number of alternatives have been proposed to try and address these limitations, but let's look at what the main barriers are. Most of these have adjusted the number of attacks per turn.

The Bookends

If you have 1 attack per turn and then gain a second, you have doubled your speed. If you have 2 attacks more than someone else, you might be 3 times as fast. This does not scale well! D&D and others have added penalties to second and subsequent attacks while Palladium changed from everyone starting with 1 attack per round to 2 to lower this disparity. An additional attack is now only a 50% gain rather than 100%.

If you want a wide range of speeds for a smooth increase in ability, then you should have a large number of attacks to begin with so that the additional attacks don't scale out of control.

The other "bookend" is what happens when you compare your speed to your opponent. If we have 4 or 6 attacks per round, then somehow the winner of initiative gets 4-6 attacks off before you can even get 1! Further, the amount of time you have to wait between turns quickly gets out of control. The time you have to wait is multiplied by both the number of players and the number of attacks per player.

Conventional designs seek to stay between these two bookends, trying to keep some balance that is fair to everyone. However, those barriers still exist because you are taking turns of multiple attacks each.

The Virtually Real Solution

I wanted to solve both of these issues in the simplest way possible. First, let's assume we want a large number of attacks and a smooth transition in power levels as people get faster with their weapons. Further, let's assume that some weapons are faster than others. We'll need a large number of attacks per round to do this, so the round was set to a larger duration. I chose 15 seconds. People typically begin the game with 6-8 attacks per round from a base of 4 (for humans).

We don't want to wait while 4-8 players all make 6-8 attacks. That is 24 - 64 attacks plus 24-64 defense rolls. No way! So, let's think this through!

If someone is only a microsecond faster, they should not get all their attacks before you get your first! The solution I chose was to simply divide the 15 seconds of a round by the number of attacks per round. This gives you the time for each attack, which is rounded to the nearest quarter second (don't worry, there is a table in the book, so you don't have to divide, and you write the final time on your character sheet). This is calculated per weapon so that your speed is different from weapon to weapon.

So, if someone wins initiative, they attack first and the GM marks off that much time using a grid with 1 box per second and a hashmark system to record quarter seconds. So, 2 1/2 seconds is a line through 2 boxes and a slash through the third. Easy?

Now we move on to the next person, whoever reacted just a few milliseconds later, and they get an offense.

Once everyone has acted, we look at who has used the least amount of time and they act next, catching them up with the action. Initiative is used to break ties, and each time you roll initiative, this defines a new attack wave for you, which is an unspecific length of time used to make certain mechanics more associative. Rounds would be confusing because an action can span from one round to the next and rounds are based on "clock time" rather than the action in the narrative. Waves are per person based on the dramatic moments when you are tied with someone else and a new initiative decides your fate.

Results

Not only does this fix both of our bookend barriers, but it leads to additional possibilities and strategies that aren't possible or would be too cumbersome to implement in an "initiative order" type of system. You'll also note that the mechanics tend to line up with the narrative for a more intuitive feel. Things just seem to make sense and extrapolating new situations becomes easier.

Take defense for example. If you are fighting two people at once, how does that affect your ability to defend yourself from multiple attacks, and how does your speed and the speed of your enemies affect your ability to do so? How many people can you hold off on your own? This sounds like it would be really hard to model, so most systems leave it out, or have special-case rules that provide modifiers. Often, modifiers result in only a minor change to the combat while having to remember when to apply the modifier and how much of a modifier, and as more and more modifiers are added, you end up with a ton of math. So, its just not worth it to attempt to model in most systems because they don't have enough granularity in time resolution! This gives rise to the idea that "crunchy" systems are full of modifiers and tons of math. It doesn't need to be that way.

The Virtually Real approach is to define everything as an action, which costs time, delaying your ability to make an offensive attack and control the combat, or you have a "maneuver", like a saving throw. Any maneuver adds a "maneuver penalty" (like a condition) to your next action, simulating the decreased likelihood that you can defend over and over again within a short period of time.

Other Solutions

You may have heard of "action points" which is just a different name for number of actions. Some versions allow you to combine action points into a heavier attack much like Virtually Real has a power attack cost an additional second. APs are still bound by the bookend problem.

How about "Popcorn Initiative"? Again, this is mainly to reduce record keeping and ease the DMs job. It also makes initiative modifiers and such pretty much useless and invites a lot of metagaming while also not addressing the bookend problem. Further, its based on selecting the next person to act, so not only does the loser of initiative have to wait on on all the winner's attacks before they can act, but the winner can just select someone from his team to act next! This means the entire side can act before you can do anything. You might as well just go around the table like the old D&D system! Being able to choose who goes next may be cool, but it's about as dissociative as you can imagine. There is a difference between coming up with a justification for why a dissociative mechanic works, and an associative mechanic that is invoked as a result of character action!

Another good one is a countdown initiative. Basically you roll initiative and each attack lowers your initiative by some value which may be dependant upon your weapon (small weapons reduce initiative by 2, medium by 3, two-handed by 4, or something similar). Your number of attacks will vary from round to round based on how well you roll on initiative. It works pretty well and solved the bookend problem, but at a cost.

You end up with each player being put on a list in initiative order and having to change that list order or change the initiative numbers every action. Its really the reverse of popcorn initiative as far as the goals.

I feel Virtually Real's solution has more benefits and is less cumbersome to keep track of while also integrating movement and other aspects. The tight integration between time and offensive and defensive capabilities is missing from the other solutions. Plus, Virtually Real was designed with automation in mind for even faster results.

For more on how this works, see the this other post that goes into more detail.

Conclusion

There is no one-right-way to do something. I don't have all the answers, but presented here is the most exciting solution I have come across to date (and I acknowledge my own bias here), but it was done by reimagining the abstractions in new ways rather than continuing to add duct tape around abstractions that haven't held true for the past few decades. I encourage you to think outside the box and see what you can come up with.

I also think that the design goals of 100% associative, character-driven mechanics, with direct mapping to the narrative tended to push the design goals and helped make the system easier to understand and remember.

Dissociative mechanics (such as popcorn initiative) are easy to create and balance, but often break immersion and lead to complications when corner cases emerge.

Don't be afraid to do something different, just make sure its easy to understand and character driven!